Professional Documents
Culture Documents
A.Penin
Analysis of
Electrical Circuits
with Variable Load
Regime Parameters
Projective Geometry Method
Second Edition
Power Systems
A. Penin
123
A. Penin
D. Ghitu Institute of Electronic
Engineering and Nanotechnologies
Academy of Sciences of Moldova
Chiinu
Moldova
ISSN 1612-1287
Power Systems
ISBN 978-3-319-28450-7
DOI 10.1007/978-3-319-28451-4
ISSN 1860-4676
(electronic)
ISBN 978-3-319-28451-4
(eBook)
Foreword
The classical electric circuit theory is widely presented in modern educational and
scientic publications. The properties of circuits, which allow simplifying the
analysis of complex networks in the case of change of one (or several) resistance or
load, are known. These properties make appropriate sections in the circuit theory
books. The problems of application and development of basic properties of circuits
in an analytical view draw the attention of researchers at present times, despite
opportunities of simulation computer systems.
It occurs in science that someone looks at traditional problems from another
viewpoint and a whole new direction with a new mathematical apparatus appears in
this traditional area of science and engineering. The book, written by A. Penin, is an
example of such a new approach in circuit theory or methods of their analysis.
This book includes the authors main publications, results of more than 30-year
searches, reflections, and doubts. All this determine a new, unusual approach to the
solution of a number of tasks of circuit theory.
The combination of such concepts as circuit theory and projective geometry (in
general, non-Euclidean geometry), is a new representation in electrical engineering
and circuit theory. There are almost no modern publications of other authors using
such a geometrical approach for circuit analysis. At the same time, similar geometrical methods are widely used in other areas of science. For a wide audience, it
is necessary to explain that geometry is a much deeper concept than a simple
graphical representation of studied dependences.
In the foreword, there is no opportunity and no need to speak about the content
of the authors main results, but one important circumstance should be noted. The
used mathematical apparatus gives a successful interpretation for observed
dependences, validation of used denitions and concepts, removes a tradition, and
their formal introduction. Therefore, the book supplements and develops basic
methods of circuit analysis, makes them evidential, and simplies understanding
of the complicated interconnected processes happening in electric circuits.
The author restricts his analysis by direct current circuits (as models of power
supply systems with one, two, and more loads), which gives the possibility to show
vii
viii
Foreword
in the simplest way the new things or results that geometry can propose for circuit
analysis.
The theorem of the generalized equivalent circuit is such a result. The author
determines the parameters of this equivalent for a network with changeable
parameters and load.
Another result is that mutual changes of resistance and current are set not in the
form of generally used increments (typical for mathematical analysis and a concept
of the derivative) but as well-founded fractionally linear expression. On this basis,
convenient formulas of recalculation of currents turn out. It is possible to carry out
the normalization of running regime parameters and to prove a technique for
comparison of regimes for circuits similar to each other.
Using own scientic and practical experience in power electronics, the author
applies new representations to a number of traditional problems in this area. It is
natural that the author uses idealized models of considered power electronics
appliances and therefore does not give the nished practical decisions. Also, the
possible directions of development of this approach are shown in this book.
As a whole, the book represents a theoretical, methodological, and methodical
interest for those who study circuit theory or work in this area directly. It is possible
to hope that the presented geometrical approach to known tasks and arising new
problems will cause the active permanent interest of experts in various areas of
electrical engineering and radio electronics.
Prof. Anatolie Sidorenko
Preface
The circuit regime analysis is one of the main problems for electric circuit theory.
The nding of the actual (absolute) value of regime parameters (voltage, current,
power, and transformation ratio for different parts of a circuit) is the simplest
analysis task. If a circuit has variable elements (loads and voltage regulators),
additional analysis tasks appear.
The interest in such circuits is dened, in particular, by the state and tendencies
of development of power electronics, modular power supply, or distributed power
supply systems with renewable power sources. Similar devices, in general, represent the complex multiple inputs and multiple outputs systems and their loads can
change from the short circuit to open circuit and further give energy. In turn, the
loads can be subdivided into high priority and additional (ballast) loads. For deniteness, it is possible to accept that such systems, for circuit theory, present linear
mesh circuits of a direct current or multi-port networks.
We will consider some of the arising additional tasks of analysis. For example, it
is important to confront operating regime parameters with characteristic values; that
is, to represent these parameters in the normalized or relative form. In this case, the
informational content of these parameters is increasing; it is possible to appreciate
qualitative characteristics of an operating regime or its effectiveness, to compare
regimes of different circuits, and to set a necessary regime.
The other task of analysis is the determination of the dependence of the regime
parameter changes on the respective change of elements parameters (for example,
the problem of the recalculation of load currents). Thus, it is necessary to set the
form of these changes reasonably; that is, to determine whether these changes are
increments or any other expressions.
Another task of analysis is the denition of the view or character of such an
active circuit with a changeable element (as a power source concerning load); that
is, this circuit shows more property of a voltage source or current source.
In the electric circuit theory, a range of circuits properties, theorems, and
methods is well known, and their use simplies the decision of these problems.
ix
Preface
Preface
xi
A. Penin
Contents
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.1 Typical Structure and Equivalent Circuits of Power
Supply Systems. Features of Circuits with Variable
Operating Regime Parameters . . . . . . . . . . . . . . . . . . . . .
1.2 Disadvantages of the Well-Known Calculation Methods
of Regime Parameters in the Relative Form for Active
Two-Poles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.1 VoltAmpere Characteristics of an Active
Two-Pole . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.2 Regime Parameters in the Relative Form . . . . . . . .
1.2.3 Regime Change in the Relative Form . . . . . . . . . .
1.2.4 Active Two-Port with Changeable Resistance . . . .
1.2.5 Scales of Regime Parameters for Cascaded
Two-Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3 Analysis of the Traditional Approach to Normalizing of
Regime Parameters for the Voltage Linear Stabilization . . .
1.4 Active Two-Port . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.4.1 Volt Characteristics of an Active Two-Port . . . . . .
1.4.2 Traditional Recalculation of the Load Currents. . . .
1.5 Nonlinear Characteristics. . . . . . . . . . . . . . . . . . . . . . . . .
1.5.1 Efciency of Two-Ports with Different Losses . . . .
1.5.2 Characteristic Regimes of Solar Cells . . . . . . . . . .
1.5.3 Quasi-resonant Voltage Converter. . . . . . . . . . . . .
1.5.4 Power-Source and Power-Load Elements. . . . . . . .
1.6 Regulated Voltage Converters . . . . . . . . . . . . . . . . . . . . .
1.6.1 Voltage Regulator with a Limited Capacity Voltage
Source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.6.2 Buck Converter . . . . . . . . . . . . . . . . . . . . . . . . .
1.6.3 Boost Converter . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
...
...
...
.
.
.
.
.
.
.
.
3
4
7
9
...
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
11
14
14
14
17
17
20
20
21
21
.
.
.
.
.
.
.
.
.
.
.
.
21
23
24
25
xiii
xiv
Contents
Part I
2
..
29
..
..
..
29
29
37
..
42
..
42
.
.
.
.
.
.
.
.
.
.
.
.
44
47
48
50
52
54
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
55
55
56
56
57
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
60
65
70
70
71
.....
.....
74
78
.....
.....
.....
82
83
84
.....
.....
.....
87
91
94
Contents
xv
Two-Port Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1 Input-Output Conformity of Two-Ports as Afne
Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1.1 Conformity of a Two-Port . . . . . . . . . . . . . . . . .
4.1.2 Conformity of Cascaded Two-Ports . . . . . . . . . .
4.2 Input-Output Conformity of Two-Ports as Projective
Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.1 Conformity of a Two-Port . . . . . . . . . . . . . . . . .
4.2.2 Versions of Conformities, Invariants, and Cross
Ratios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.3 Conformity of Cascaded Two-Ports . . . . . . . . . .
4.3 Use of Invariant Properties for the Transfer of Measuring
Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.3.1 Transfer of Signals over an Unstable Two-Port . .
4.3.2 Conductivity Measurement by an Unstable
Two-Port. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.4 Deviation from the Maximum Efciency of a Two-Port . .
4.4.1 Regime Symmetry for the Input Terminals . . . . .
4.4.2 Regime Symmetry for the Output or Load. . . . . .
4.5 Effectiveness of Modular Connections. . . . . . . . . . . . . . .
4.5.1 Complementary Knowledge About a Two-Port . .
4.5.2 Parallel Connection of Two Converters . . . . . . . .
4.5.3 Connection of Two-Ports with the Interaction . . .
4.6 Effectiveness Indices of a Two-Port with Variable Losses .
4.6.1 Problems of Energy Indices . . . . . . . . . . . . . . . .
4.6.2 Inuence of Losses on the Load Power . . . . . . . .
4.6.3 Inuence of Losses on the Efciency . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
....
97
....
....
....
97
97
99
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
119
120
121
123
126
126
127
130
131
131
131
135
138
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
141
141
141
143
143
145
.
.
.
.
.
.
. . . . 102
. . . . 102
. . . . 106
. . . . 109
. . . . 116
. . . . 116
. . . . . 148
. . . . . 153
. . . . . 153
. . . . . 154
. . . . . 157
. . . . . 162
xvi
Contents
Part II
6
. . . . . . . . 167
. . . . . . . . 167
. . . . . . . . 167
. . . . . . . . 173
. . . . . . . . 175
. . . . . . . . 183
. . . . . . . . 183
. . . . . . . . 190
. . . . . . . . 198
. . . . . . . . 205
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
207
207
207
210
. . . . . 228
. . . . . 235
.
.
.
.
.
.
.
.
.
. . . . . 213
. . . . . 213
. . . . . 219
. . . . . 224
. . . . . . . 237
. . . . . . . 237
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
244
244
246
251
252
254
260
260
262
Contents
xvii
Part III
. . . . . . . . . . . 275
. . . . . . . . . . . 275
. . . . . . . . . . . 275
. . . . . . . . . . . 276
. . . . . . . . . . . 279
. . . . . . . . . . . 282
. . . . . . . . . . . 286
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
289
289
295
296
299
305
306
308
312
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
313
313
313
315
322
. . . . . . 324
. . . . . . 328
. . . . . . 330
. . . . . . 333
. . . . . . 334
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
337
337
337
348
348
354
357
xviii
Part IV
Contents
. . . 361
. . . 361
. . . 361
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
363
363
364
370
372
374
374
375
381
388
. . 389
. . 389
. . 392
. . 396
. . 400
. . 401
.
.
.
.
.
.
.
.
405
405
405
410
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415
xix
This book introduces electric circuits with variable loads and voltage regulators.
It allows to dene invariant relationships for various parameters of regime and
circuit sections and to prove the concepts characterizing these circuits. Generalized
equivalent circuits are introduced. Projective geometry is used for the interpretation
of changes of operating regime parameters. Expressions of normalized regime
parameters and their changes are presented. Convenient formulas for the calculation
of currents are given. Parallel voltage sources and the cascade connection of
multi-port networks are described. The two-value voltage regulation characteristics
of loads with limited power of voltage source is considered. The book presents the
fundamentals of electric circuits and develops circuit theorems.
The second edition was extended and contains additional chapters to circuits
with nonlinear regulation curves, circuits with nonlinear load characteristics, concepts of power-source and power-load elements with two-valued characteristics,
quasi-resonant voltage converters with self-limitation of current as well as the
similarity of characteristics of converters and electronic devices.
This book is useful to engineers, researchers, and graduate students who are
interested in the basic electric circuit theory and the regulation and monitoring of
power supply systems.
xxi
Chapter 1
Introduction
1.1
The circuit regime analysis is one of the main problems for the electric circuit
theory [1, 7]. The nding of the actual (absolute) value of regime parameters
(voltage, current, voltage transformation ratio, and so on), is the simplest analysis
task. In turn, the determination of the maximum load power and maximum efciency is an essential energy problem [6, 14], including distributed power supply
systems. Such systems, shown in Fig. 1.1, contain multiple power sources, energy
storages, direct current DC/DC voltage converters, charge/discharge units,
point-of-load POL regulators, and loads [4, 13].
Usually, the static characteristics of all the incoming units are used for energy
analysis [9, 12] and analog computation of power systems [11]. Therefore, this power
supply system is the complex DC circuit with a given number of input and output
terminals. For example, such a circuit is shown in Fig. 1.2. The voltage DC/DC
converters are simply DC transformers. The resistive network determines losses of all
the converters and supply lines.
Let us consider the feature of the show circuit. The interaction both between
power sources and loads is the main feature of this system. The regime of the loads
and energy storages may change from the energy consumption to return of energy.
The loads may be also subdivided into basic (priority) and additional (buffer) loads.
Therefore, additional analysis tasks appear for these circuits with variable elements (loads, voltage regulators). For example, it is important to confront operating
regime parameters with the characteristic values; that is, to represent these
parameters in the normalized or relative form. In this case, the informational content
of these parameters is increasing; it is possible to appreciate the qualitative characteristics of an operating regime or its effectiveness, to compare regimes of different circuits, to set a necessary regime in the sense of similarity theory.
Springer International Publishing Switzerland 2016
A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4_1
Introduction
Usually, relative expressions are constituted by using of the characteristic values (as
scales) for the corresponding regime parameter [19]. Similarly, the regime change is
dened by the difference or ratio of subsequent and initial regime parameters; there are
changes in the form of times or percents. For example, the open-circuit OC
voltage and short circuit SC current will be the corresponding scales for the load
voltage and current of the simplest circuit. In the same way, the maximum load power
or maximum source power will be the scales for the running load power.
The other task of analysis is the determination of the regime parameter changes
via respective changes of element parameters (for example, the problem of the
recalculation of load currents). Thus, it is necessary to set the form of these changes
reasonably; that is, if these changes are increments or the other expressions.
Also, we have the next task. The change of the load parameters or parameters of
circuit denes the corresponding regime change and its effectiveness. Therefore, a
deeper analysis and introduction of changes in the valid form of both regime
parameters and effectiveness indicators are necessary.
In the electric circuit theory a range of properties, theorems, and methods are
well-known, and their use can simplify the solution of these problems.
However, the known approaches do not completely disclose the properties of the
circuits with variable elements that reduce the efciency of analysis.
Using the equivalent circuit in Fig. 1.2, we will choose the simplest and
important circuits. The analysis of such simple networks shows the disadvantages
of the known methods.
1.2
1.2.1
The simplest circuit of an active two-pole is shown in Fig. 1.3a. At change of a load
resistance RL from the short circuit SC to open circuit OC, a load straight line or
voltampere characteristic in Fig. 1.3b is given by a linear expression
IL
V0 VL
VL
ILSC ;
Ri Ri
Ri
(a)
(b)
1:1
(a)
Introduction
(b)
Fig. 1.4 a Influence of loads to each other. b Influence of voltages sources to each other
In turn, an internal resistance Ri in Fig. 1.4a determines the influence of the load
resistances RL1 ; RL2 to each other. Similarly, the influence between the paralleling
voltage sources V01 ; V02 takes place in Fig. 1.4b.
Let us consider the straight lines of the initial circuit and a similar circuit with the
~0 , R
~ i in Fig. 1.5. The regimes of these circuits will be similar or
other values V
equivalent if the correspondence of the characteristic and running regime parameters takes place. For the given case, this conformity is specied by arrows.
1.2.2
Let us constitute the relative expression of the load straight line for our simplest
circuit in Fig. 1.3. For that, it is possible to use the values ILSC ; V0 as the scales.
We may rewrite Eq. (1.1) in the form
IL
VL
1 :
V0
ILSC
1:2
IL
IL
SC
IL
SC
IL
RL
RL
V0
V0
VL, VL
~IL
IL
SC ;
SC
~
IL
IL
VL
~L
VL V
:
~0
V0 V
1:3
1:4
Therefore, the relative values I L ; V L allow evaluating the use of the voltage
source for current and voltage of the running regime.
From (1.3), the recalculation formulas of the actual regime parameters are as
follows:
mI
IL ILM
;
~IL ~ILM
mV
VL V0
;
~L V
~0
V
1:5
1:6
PL
VL
P0
1:7
RL
:
Ri RL
1:8
Introduction
1.00
0.75
0.50
PL
0.25
0.00
0.0
0.5
1.0
VL
~L
RL R
:
~i
Ri
R
1:9
RL
:
1 RL
1:10
RL Ri
;
~L R
~i
R
1:11
1.2.3
Let an initial (the rst) regime correspond to values R1L ; VL1 ; IL1 . In turn, a subsequent (the second) regime corresponds to R2L ; VL2 ; IL2 . For convenience, we consider
V0 10; Ri 1. Using (1.1), (1.8), we get the concrete or absolute values of all the
regime parameters shown in Fig. 1.7.
Let us obtain these changes of regime parameters in the absolute and relative
form. In the given case, we use the difference of values that corresponds to the
known variation theorem of resistance and current into a circuit.
Then, we get the relative changes for the currents and voltages
IL2
IL1
DI 21 1
21
I L 0:1:
SC
SCL
SC
10
IL
IL
IL
VL2 VL1 DVL21
1
21
V L 0:1:
10
V0 V0
V0
1:12
Using (1.9), we constitute the following relative changes for the resistances:
R2L R1L DR21
0:5
L
0:5:
1
Ri Ri
Ri
1:13
RL
RL
1 R21 1.5
L
RL
32
RL
RL
5.666
RL
31
RL
10
IL
IL
21
IL
IL
32
IL
IL
2.5 1.5
IL
31
IL
VL
VL
0
21
VL
VL
7.5
VL
6.5
10 V
L
DIL21
1
DR21
0:5=2
L =RIN
IL1
5 1;
21
1
1 0:5=2
1 DRL =RIN
Introduction
1:14
where R1IN Ri R1L 2 is the input resistance of our circuit for the subsequent
load.
It is possible to rewrite expression (1.14) in the relative view
1
DIL21
DR21
L =RIN
:
1
21
IL
1 DRL =R1IN
DIL21
0:2:
IL1
1:15
Hence, the practical question arises, which expressions may we use? Because
both initial expressions (1.12), (1.13), and complementary expression (1.15) have a
clear physical sense.
Group Properties of Regime Changes
Let the next regime correspond to the values R3L ; VL3 ; IL3 shown in Fig. 1.7. Then,
corresponding to (1.15), we get the changes relatively to the rst regime
DR31
L
1;
R1IN
DIL31
0:5:
IL1
Let us obtain the changes relatively to the second regime. To do this, we must
calculate R2IN Ri R2L 2:5. Then,
DR32
L
0:6;
R2IN
DIL32
0:375:
IL2
We note that the resultant regime change DIL31 =IL1 0:5 is not expressed
through the initial change DIL21 =IL1 0:2 and intermediate change DIL32 =IL2
0:375 by the group operation (addition or multiplication). In this sense, regime
changes (1.12), (1.13) hold the group properties that are convenient for analysis.
Anyway, the next regime changes will lead to the increase of a number of the
relative values for various parameters and determinations of changes that complicates analysis.
1.2.4
Let an active two-pole A with a load resistance RL1 contain a changeable resistance
RL2 shown in Fig. 1.8a.
Therefore, values R1L2 ; R2L2 and so on determine a family of load straight lines
with the various values of SC currents and OC voltages in Fig. 1.8b. Then, we have
the set of the base or characteristic values as the parameters of the Thvenin
equivalent circuit. Also, the corresponding variations of scales (1.3) complicate
analysis.
Hence, it is necessary to obtain such base values which do not depend on the
variation of an additional load. Also, we may get the relative values and relative
changes of this additional load for comparing regimes for different circuits.
1.2.5
Let us consider a cascade connection of two two-ports TP1; TP2 with a variable
load conductivity YL2 in Fig. 1.9a. At change of load conductivity, we get a family
of load straight lines for various sections; that is, we have some correspondence
between these load lines by dashdot lines shown in Fig. 1.9b.
Fig. 1.8 a Active two-pole
with a changeable resistance
RL2 . b Family of load straight
lines
(a)
(b)
RL2
RL2
I1
SC1
I1
I1
A
V1
RL2
SC2
I1
RL1
RL1
OC2
V1
OC1
V1
V1
10
Introduction
(a)
TP1
V0, I0
TP2
V1, I1
V2, I2
y10
y0
y1
YIN1
YL1=YIN2
I2, I1, I0
(b)
YL2
I1(V1)
I0(V0)
SC
I0
SC
I2(V2)
I1
SC
I0
SC
I2
V1
1
I2
OC
I0
I1
YL2
2
I2
YL2
OC
I1
OC
V1
OC
V2
V0
V2, V1, V0
Our cascaded circuit, relatively to the load YL2 , is similar to the simplest circuit
in Fig. 1.3. Then, the characteristic values V2OC ; I2SC are the scales or base values.
Therefore, the relative values of the load current and voltage are determined by
(1.3)
I2
I2
;
I2SC
V2
V2
:
V2OC
RL2
:
Ri2
So, we have the following base values for the load section V2 ; I2
V2OC ; I2SC ; V2SC 0; I2OC 0:
1:16
11
These values determine the corresponding base values for the other section of
the cascaded circuit. For the section V1 ; I1 , we get
V1OC ; I1SC ; V1SC ; I1OC :
1:17
1:18
The two nonzero base values turn out for all the regime parameters.
Then, there is a problem, how reasonably we may express these parameters for
1
2
, subsequent YL2
and regime changes in the relative form.
the initial regime YL2
It is possible to note that regime changes, as the length of segments on all the
load straight lines have different lengths for usually used Euclidean geometry.
1.3
For the illustration of the assigned task, we consider two simple active two-poles
with a load conductivity YL1 in Fig. 1.10. The equation of the load straight line of
the rst active two-pole in Fig. 1.10a is given by
I1 V0 V1 y01 y01 V0 y01 V1 ;
where the conductivity y01 corresponds to the internal resistance of the voltage
source V0 and SC current I1SC y01 V0 .
Then, we use normalized expression (1.2) which contains the two normalized
values
I1 V1
;
:
I1SC V0
(a)
V0
y01
(b)
I1
V1
YL1
V0
V1N
y01
I1
y1N
V1
YL1
Fig. 1.10 a Active two-pole without a voltage stabilization. b Stabilization of a load voltage
12
Introduction
The regimes of two similar circuits with running values of currents I1 ; ~I1 and
~1 will be equivalent or equal to each other if the normalized values of
voltages V1 ; V
currents and voltages (1.3) take place; that is,
~I1
I1
SC ;
SC
~I
I1
1
~1
V1 V
:
~0
V0 V
The equation of the load straight line of the second active two-pole in Fig. 1.10b
is given by
I1
y0N y1N
V0 V1 ;
y0N y1N
1
:
V0
I1SC 1 y1N =y0N
1:19
~1
V1 V
;
~0
V0 V
y1N ~y1N
:
y0N ~y0N
1:20
If regimes differ, how may we express this difference in the convenient view? It
is not clear, how may we work with the set of these six different values. It would be
convenient to work with one value which characterizes this difference.
If the access to the voltage source is missing, what then must we choose as the
normalizing value? It is possible to normalize by the maximum load current
I1M y0N V0 V1 , when the linear regulator is completely closed. Then, we have
I1
y1N =y0N
:
I1M 1 y1N =y0N
1:21
;
I1M ~I1M
y1N ~y1N
:
y0N ~y0N
We again obtain the two normalized values. But there is a contradiction with
condition (1.20) for currents.
13
y01
N
V0
y1N
V1
V1N
In Eq. (1.21) we pass from the value y1N to the voltage V1N of the linear
regulator. Then, we obtain the normalized expression of the regulator
I1
V1N
1
:
I1M
V0 V1
The equivalent circuit in Fig. 1.11 corresponds to this expression.
Even for such a simple circuit there is an uncertainty, how correctly or reasonably we may represent the normalized expression of regime?
The problem becomes complicated even more for a case of two and more loads
with the voltage stabilization. For this, we consider Fig. 1.12.
In the case of two loads (without conductivity yN ), the two equations turns
out to be
y0N y1N
V0 V1 y0N I2 ;
y1N
y0N y2N
V0 V2 y0N I1 :
I2
y2N
I1
y1N
y0N
V1
y2N
V0
yN
V2
YL2
YL1
14
Introduction
On the other hand, the normalization by the maximum load currents I1M
y0N V0 V1 ; I2M y0N V0 V2 leads to the reciprocal components I2 =I1M ; I1 =I2M .
These components raise a number of normalized values.
Therefore, the shown examples of the formal normalization do not allow comparing the regimes of different systems.
1.4
1.4.1
Active Two-Port
Volt Characteristics of an Active Two-Port
Let us consider the features of an active two-port with two load conductivities
YL1 ; YL2 in Fig. 1.13a. The interaction between load voltages is observed at change
of loads. Therefore, the variable slope of the load volt characteristics or load straight
lines V2 V1 ; YL1 ; V2 V1 ; YL2 takes place in Fig. 1.13b; the values YL1 ; YL2 are
parameters.
Therefore, such base values as the load SC currents and OC voltages are
changed. In turn, the corresponding variations of scales (1.3) complicate analysis.
Hence, similar to Sect. 1.2.4, it is necessary to obtain such base values that do
not depend on variable loads.
1.4.2
Let us lead the base statements of the traditional approach to the recalculation of the
load currents [3]. Expression for the currents I1 ; I2 of Fig. 1.14a has the view
"
I1
I2
"
Y11 Y12
Y12 Y22
# "
# "
#
I1SC
V1
SC :
V2
I2
(a)
(b)
1:22
V2
YL1
V1
A
YL1
YL2
YL1
1
V2
YL2
YL2
0V
V1
15
(a)
I2
I1
1
V1
V2
RL1
(b)
RL2
I1
21
RL1
I2
G
Y
RL1
RL2
21
RL2
SC currents I1SC ; I2SC are considered as the initial values I10 ; I20 of the load currents for
the zero load resistances R0L1 0; R0L2 0.
Now, we introduce the rst load resistances R1L1 ; R1L2 as the rst increments
10
DR10
L1 ; DRL2 concerning the initial zero values; that is,
0
10
R1L1 DR10
L1 RL1 DRL1 ;
0
10
R1L2 DR10
L2 RL2 DRL2 :
Therefore, the relationships take place for these subsequent or the rst values of
currents and voltages
1
V11 DR10
L1 I1 ;
1
V22 DR10
L2 I2 :
1
10 1
0
I11 Y11 DR10
L1 I1 Y12 DRL2 I2 I1
1
10 1
10 1
0
I2 Y12 DRL1 I1 Y22 DRL2 I2 I2 :
Then, we get
1
10 1
0
1 Y11 DR10
L1 I1 Y12 DRL2 I2 I1
10 1
10 1
Y12 DRL1 I1 1 Y22 DRL2 I2 I20 :
1:23
I11
1:24
16
Introduction
Let us carry out the analysis of these relationships. Let the second load resis20
tances R2L1 ; R2L2 be given as the second increments DR20
L1 ; DRL2 concerning the
initial zero values; that is,
0
20
R2L1 DR20
L1 RL1 DRL1 ;
0
20
R2L2 DR20
L2 RL2 DRL2 :
Then,
0
0
I10 DR20
L2 Y22 I1 Y12 I2
;
20
D
0
0
I 0 DR20
L1 Y11 I2 Y12 I1
I22 2
;
D20
I12
1:25
21
We try to introduce the intermediate changes DR21
L1 ; DRL2 thus
1
21
10
20
R2L1 DR21
L1 RL1 DRL1 DRL1 DRL1 ;
1
21
10
20
R2L2 DR21
L2 RL2 DRL2 DRL2 DRL2 :
Therefore, the denominator D20 will contain both intermediate changes DR21
L1 ;
1
1
DR21
,
and
the
rst
values
of
resistances
R
;
R
.
Thus,
the
structure
of
denomiL2
L1
L2
nator (1.25) shows that group properties are not carried out for intermediate changes
of load resistances. Therefore, subsequent changes must be counted concerning
21
initial zero values. However, if we want to use the changes DR21
L1 ; DRL2 concerning
the values R1L1 ; R1L2 , the two-port with conductivity parameters G is obtained in
Fig. 1.14b. Thus, the recalculation of parameters of a new two-port is required.
Let us introduce variations of currents as increments DI110 ; DI210 concerning the
initial values; that is,
I11 DI110 I10 ;
10
10
10
10 0
10 0
1 Y11 DR10
L1 DI1 Y12 DRL2 DI2 Y11 DRL1 I1 Y12 DRL2 I2
10
10
10
10 0
10 0
Y12 DR10
L1 DI1 1 Y22 DRL2 DI2 Y12 DRL1 I1 Y22 DRL2 I2 :
17
The solution of this system gives the increments of currents via the increments of
load resistances
2
10
10
Y11 DR10
Y12 DR10
L1 Y11 Y22 Y12 DRL1 DRL2 0
L2 0
I
I ;
1
D10
D10 2
2
10
10
Y22 DR10
Y12 DR10
L2 Y11 Y22 Y12 DRL1 DRL2 0
L1 0
I
I ;
2
D10
D10 1
DI110
DI210
1:26
1.5
Nonlinear Characteristics
Besides the considered linear voltampere characteristics of linear circuits, nonlinear characteristics represent a practical interest too. The above-mentioned
dependence of power via load voltage (1.6) is such nonlinear characteristic. For a
passive two-port network, the dependence of efciency via load voltage is also a
nonlinear function.
Besides linear active two-pole networks, a wide class of active nonlinear
two-poles is known. Solar cells, quasi-resonant voltage converters possess a typical
convex characteristic. Power-source and power-load elements with a concave
characteristic are also known. Further, we will consider features of such nonlinear
characteristics.
1.5.1
Let us consider a two-port in Fig. 1.15. In this case, the power transfer ratio KP or
efciency g is a nonlinear function.
As it is known, the system of equation of this circuit has the view
"
I0
I1
"
Y00 Y10
Y10 Y11
# "
#
V0
;
V1
1:27
18
V0, I0
Introduction
V1, I1
y10
y0
y1
YL
YIN
where Y parameters
Y00 y10 y0 ;
Y11 y10 y1 ;
Y10 y10 :
r
Y00
DY ;
Y11
YL:C
r
Y11
DY :
Y00
1:28
V0
I0
"
a11 a12
a21 a22
# "
V1
I1
3
Y11 1
" #
6 Y10 Y10 7 V1
7
6
4 DY Y00 5 I :
1
Y10 Y10
2
1:29
3
V0
4 I0 5 chc
shc
YIN:C
3
YL:C
V1 p
DY 7
shc 6
7:
6
4
chc
I1 5
p
DY
1:30
1:31
19
We have the relative values YIN =YIN:C ; YL =YL:C . However, expression (1.31) is
not a pure relative because contains the value thc. This value is diverse for
two-ports with different losses.
In turn, the power transfer ratio has the following forms:
8
< KP PP1 ch2 c sh2 c chc shc YYIN YYIN:C
0
IN:C
IN
h
i
: K ch2 c sh2 c chc shc YL YL:C 1 :
P
YL:C
YL
1:32
1:33
So, the relative values YIN =YIN:C ; YL =YL:C determine the deviation from the
admittance matching by some way. However, expressions (1.32) are not pure relative because they contain the value thc. If we try to use the value KPM , as the scale
value, we cannot obtain the pure relative form of expression (1.32).
Let us use the following voltage transfer ratio:
KG
V1
Y11 V1
;
V1OC Y10 V0
which corresponds to the voltage ratio of the Thvenin equivalent circuit. Using
(1.30), we get
KP KG KG
1 KG
;
A KG
A ch2 c:
1:34
An example of efciency (1.34) is shown in Fig. 1.16 for the various values
A. The maximum values of KPM correspond to the various values KG .
Fig. 1.16 Efciency of
two-ports for different losses
0.8
KPM
0.6
A=1.05
KP
0.4
KPM
A=1.25
0.2
0.0
0.0
0.2
0.4
0.6
0.8
KG
1.0
20
Introduction
The solid arrows demonstrate the correspondence of the characteristic points. So,
we have the three characteristic regimes. Then, the problem of scales, relative
expressions, and correspondence of running points (dash arrow) arises.
1.5.2
1.5.3
It is known the load resonant converters, for example, with the zero-current
switching [17]. These converters regulate their output by changing the switching
period TS . Simulation of ORCAD model shows that its load curve is almost
(a)
IL IL
P P
SC
IL
PM
SC
IL
PM
IM
PM
PM
V
0
VM V OC
OC
VL, VL
0
VM V OC
OC
VL,VL
(b)
I
M
I
I
0 V
M
V
P
V
Fig. 1.17 a Voltampere characteristic and power-volt characteristic of solar cells. b Voltampere
characteristic of power-load element
21
rectangular with maximum load power point PM and similar to Fig. 1.17a. So, we
have the same problems to determine the regime parameters.
1.5.4
The power-source and power-load elements are known from [4, 18]. Their load
characteristics have the typical form as shown in Fig. 1.17b. Analysis of the
power-load element shows the two-valued voltage of a limited capacity voltage
~ But for all that, the voltampere characteristic of this
source; that is, points M; M.
power-load element has one-valued representation.
On the other hand, taking into account the losses of real power-source and
power-load elements, we get the two-valued voltampere characteristic of these
elements. So, we have some characteristic regimes, and the same problems to
determine the regime parameters.
1.6
1.6.1
In power supply systems with limited capacity voltage sources, the limitation of
load power appears. An independent power supply system with a solar array and
rechargeable battery may be the example of one.
We consider DC transformer with switched tapped secondary windings as a
voltage regulator VR1 connected to a limited capacity voltage source V0 in
Fig. 1.18a. The regulator denes a transformation ratio
n1
V1
:
V
1:35
1:36
We get the relative expression which coincides with different circuits. But,
similar to Sect. 1.2.2, the relative regime parameters V 1 ; n1 have the different
quantities. For example, the maximum value V 1M 0:5 corresponds to n1M 1.
22
Fig. 1.18 a Power supply
system with limited capacity
voltage sources. b Example of
its regulation characteristic
Introduction
VR1
(a)
Ri
V1
V0
VR2
n1
V2
R
n2
(b)
0.50
V1
V0
0.25
0.00
0
n1
Therefore, it is necessary to stipulate the parameter to set the regime value in the
form of a number.
Also, there is a question of the next switching step for the secondary winding;
whether the regular or irregular step will be for the load voltage or for transformation ratio relatively to maximum permissible values. It is determined by how to
set the value changes; that is, by increments, ratio and so on. The situation becomes
complicated even more, if such a power supply system contains two loads with
individual voltage regulators shown by dash lines in Fig. 1.18a; the interference of
these loads takes place.
1.6.2
23
Buck Converter
V0
V0
D
D;
1 RRL
1 r2
1:37
1
1 r2
1:38
D:
VL
1
VL
D2
:
D1
1:39
V0
1 r2
L
V0
24
Introduction
^L1 V
^L2 :
V
In turn, the regime changes are expressed by the ratio or increment of values
^L1 ; V
^L2 .
V
If an initial expression of type (1.37) is more complex, it will be also difcult to
constitute relative expression (1.39) directly and reasonably. This case is considered
in the next section.
1.6.3
Boost Converter
;
V0 1 r2 n2
1:40
1
where n 1D
1 is the inverse relative pause width.
(a)
V0
(b)
2=0.05
VL
V0 2
2=0.1
1
0
1
25
^n rn:
Then
^L
V
^n
1 ^n2
Although the relative expression turns out, for the characteristic value n 1 (the
transistor is switched off) the new value ^n r depends on the parameter of the
converter.
References
1. Alexander, C.K., Sadiku, M.N.O.: Fundamentals of Electric Circuits, 5th edn. McGrawHill,
New York (2009)
2. Bessonov L.A.: Teoreticheskie Osnovy Elektrotekhniki. Elektricheskie tsepi, Izd.9. (Basic
electrical engineering theory: Electric circuits, 9th edn.). Vyshaia shkola, Moskva (1996)
3. Bogatyrev, O.M.: General method of solving of problems for linear circuit at changeable
resistances of branches. Elektrichestvo. 9, 6769 (1955)
4. Emadi, A., Khaligh, A., Rivetta, C.H., Williamson, G.A.: Constant power loads and negative
impedance instability in automotive systems: denition, modeling, stability, and control of
power electronic converters and motor drives. IEEE Trans. Veh. Technol. 55(4), 11121125
(2006)
5. Erickson, R.W., Maksimovic, D.: Fundamentals of power electronics. Springer, Berlin (2001)
6. Glisson, T.H.: Introduction to circuit analysis and design. Springer, Berlin (2011)
7. Irwin, J.D., Nelms, R.M.: Basic Engineering Circuit Analysis, 10th edn. Wiley, Hoboken
(2011)
8. Ishibashi, K., Kimura, Y., Nivano, M.: An extensively valid and stable method for derivation
of all parameters of a solar cell from a single current-voltage characteristic. J. Appl. Phys. 103
(9), 094507/1094507/6 (2008)
9. IvanovTsyganov, A,I.: Electropreobrazovatelnye ustroistva RES. (Electro-converting
equipments of radio-electronic systems RES). Vyshaia shkola, Moskva (1991)
10. Kazimierczuk, M.K.: Pulsewidth Modulated DCDC Power Converters. Wiley, New York
(2008)
11. Leger, A.S., Nwankpa, C.O.: Analog and hybrid computation approaches for static power
flow. Paper presented at the 40th Annual Hawaii international conference on system sciences
(2007)
26
Introduction
12. Levron, Y., Shmilovitz, D.: On the maximum efciency of systems containing multiple
sources. IEEE Trans. Circuits Syst. I Regul. Pap. 57(8), 22322241 (2010)
13. Liu, X., Wang, P., Loh, P.C.: A hybrid AC/DC microgrid and its coordination control. IEEE
Trans. Smart Grid 2(2), 278286 (2011)
14. Nelatury, S.R.: Didactic uses of maximum power transfer theorem and guided problem
solving. Int. J. Electr. Eng. Educ. 51(3), 244260 (2014)
15. Penin, A., Sidorenko, A.: A convenient model for I-V characteristic of a solar cell generator as
an active twopole with selflimitation of current. World Acad. Sci. Eng. Technol. 3(4),
905-909 (2009). http://www.waset.org/publications/9926. Accessed 30Nov 2014
16. Penin, A., Sidorenko, A.: Determination of deviation from the maximum power regime of a
photovoltaic module. Moldavian J. Phys. Sci. 9(2), 191198 (2010). http://sfm.asm.md/
moldphys/2010/vol9/n2/index.html. Accessed 30 Nov 2014
17. Penin, A.: A quasi-resonance voltage converter with improved parameters. Elektrichestvo. 2,
5864 (2009)
18. Singer, S., Erickson, R.W.: Powersource element and its properties. IEEE Proc. Circ. Devices
Syst. 141(3), 220226 (1994)
19. Venikov, V.A.: Theory of similarity and simulation: with applications to problems in electrical
power engineering. Macdonald, London (1969)
Part I
Chapter 2
2.1
2.1.1
Afne Transformations
It is known that any linear circuit (an active two-pole A) relative to load terminals is
replaced by a voltage source V0 in series with an internal resistance Ri [1, 6]. Let us
consider two cases of the load.
Case 1
A variable voltage source VL is the load of an active two-pole A in Fig. 2.1. The
voltage VL (independent quantity) and load current IL are the parameters of operating or running regime.
At change of the load voltage from the short circuit SC VL 0 to open circuit
OC VLOC V0 , a load straight line is given by (1.1)
IL
V0 VL
VL
ILSC ;
Ri Ri
Ri
2:1
where ILSC is the SC current. This straight line is shown in Fig. 2.2.
The characteristic regime parameters ILSC ; V0 determine the different scales for
the axes of coordinates.
We want to represent a running regime parameter of the load by a certain
quantity, which would have an identical value for various actual regime parameters
such as voltage and current. To do this, we use a geometrical interpretation of
change or kinematics of a circuit regime.
As the load characteristic is dened by linear expression (2.1), a similar
expression in geometry denes an afne transformation, conformity, or mapping of
Springer International Publishing Switzerland 2016
A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4_2
29
30
Fig. 2.1 Active two-pole
with a load voltage source
the voltage axis to current axis VL ! IL [7]. The values ILSC ; Ri are parameters of
this afne transformation. The mechanism of this mapping is shown by parallel
lines with arrows for three voltage values or points VL1 ; VL2 ; VL3 .
An afne transformation is characterized by an invariant of three points
uVL1 ; VL2 ; VL3 uIL1 ; IL2 ; IL3 :
The invariant represents a certain expression, which only contains the chosen
values of voltage or corresponding values of current; the parameters of the afne
transformation do not enter in this expression.
To obtain the invariant expression, it is necessary to exclude the two parameters
from Eq. (2.1) by means of three equations. For arbitrary three voltage values
VL1 ; VL2 ; VL3 we have the system of equations
IL1 ILSC
VL1
;
Ri
IL2 ILSC
VL2
;
Ri
IL3 ILSC
VL3
:
Ri
2:2
VL2 VL1
:
Ri
Ri
2:3
31
Similarly, we have
IL1 IL3
VL3 VL1
:
Ri
Ri
2:4
IL2 IL3
VL3 VL2
:
Ri
Ri
2:5
Let us exclude the parameter Ri . To do this, we use expressions (2.3) and (2.5).
Then, the required invariant or afne ratio is obtained as follows:
IL1 IL2 VL2 VL1
:
IL2 IL3 VL3 VL2
2:6
:
IL1 IL2 VL2 VL1
2:7
;
0 V0
0 ILSC
IL1 0
V0 VL1
:
V0 0
ILSC 0
2:8
0 IL1
IL1
;
0 ILSC ILSC
2:9
32
VL1 V0 0
V0 VL1
:
V0 0
2:10
Further, we consider the sense of the value n1 . We obtain the quantity that has an
identical value for current and voltage. For the current, this value n1 is simply the
normalized value (at the expense of a particular choice of the base points).
Therefore, expression (2.8) coincides with Eq. (2.1) for the normalized values
IL1
V1
1 L:
SC
V0
IL
2:11
VL1 VL2
IL1 IL21 :
Ri
Ri
2:12
The obtained transformation with a parameter IL21 translates the initial regime IL1
point into the subsequent regime IL2 point; that is, IL1 ! IL2 shown in Fig. 2.5.
33
Further, transformation (2.12) with a parameter IL32 translates the point IL2 ! IL3 ;
that is,
IL3 IL2 IL32 :
For example, let IL32 be equal IL21 , as it is shown in Fig. 2.5. It now follows that
IL3 IL1 IL21 IL32 IL1 IL31 :
Therefore, we have the resultant transformation IL1 ! IL3 with a parameter IL31 .
Thus, a set of transformations (2.12) is a group.
Let an initial regime be given by two values I11 ; I21 . These values correspond to a
segment I21 in Fig. 2.6.
34
Fig. 2.7 Signal V21 and bias VL voltage is the load of an active two-pole
I22 I21 I 21 :
2:13
2:14
Thus, this invariant is the Euclidean distance between two points. Therefore,
geometry of this transformation group is geometry of a Euclidean straight line.
Also, we can consider the dual variant of segment and its movement in Fig. 2.8.
Let an initial regime be given by two values I11 ; I12 or segment I12 I11 . We use the
translation with a parameter I21 . Then, the subsequent points
I21 I11 I21 ;
35
n21 :
ILSC ILSC V0 V0
2:15
This regime change corresponds to a relative change in percent. Note that the
invariants of transformation (2.12) coincide with the known principle of superposition for linear circuits.
Remark Let us introduce a regime change by ratio of currents. Using expression
(2.11), we get
IL2
IL1
IL2 V0 VL2
:
ILSC ILSC IL1 V0 VL1
We can note that the ratio of currents does not contain a transformation
parameter. But, this ratio of voltages contains the transformation parameter V0 .
Hence, the condition of invariant is not executed.
Case 2
A variable conductivity YL is a load of the simplest circuit in Fig. 2.9; an internal
resistance of this active two-pole Ri 0.The equation of this circuit
IL V0 YL :
2:16
The conductivity YL (independent quantity) and load current are the parameters
of a running regime.
36
For this circuit, the regime has only an absolute value; it is impossible to state a
relative expression in view of the absence of a scale.
Now, we consider changes of regime. The initial conductivity value equals YL1
and subsequent one equals YL2 . Then, we have the two values of currents
IL1 V0 YL1 ;
IL2 V0 YL2 :
2:17
m21 :
IL1 YL1
2:18
2:19
This transformation with a parameter m21 translates a point of an initial regime IL1
to a point of a subsequent regime IL2 ; that is, IL1 ! IL2 . The transformation is a group
transformation because
IL3 m32 IL2 m32 m21 IL1 m31 IL1 :
This group is the dilation group. For example, let m32 be equal m21 , as shown in
Fig. 2.10; a usual distance between currents is increased.
Let an initial regime be given by two values I11 ; I21 . These values correspond to
segment I1 I2 in Fig. 2.11. If we apply transformation (2.19) to the initial points
I11 ; I21 or to the segment, we obtain the subsequent points
37
2:20
2.1.2
Projective Transformations
V0 VL
VL
ILSC :
Ri Ri
Ri
2:21
38
The bunch of straight lines with a parameter RL corresponds to this straight line.
The bunch center is the point 0. The equation of this bunch is given as
IL
1
VL :
RL
2:22
Further, it is possible to calibrate the load straight line in the load resistance
values [10, 13, 15]. The internal area RL [ 0 at the load change corresponds to
regime of energy consumption by the load. If to continue the calibration for the
negative values RL \0, the regime passes into the external area, which physically
means return of energy to the voltage source V0 .
Therefore, at the innitely remote point RL Ri , the calibrations of the load
straight line will coincide for the areas VL [ V0 ; VL \0. So, this straight line is
closed; that is typical property of projective geometry.
The load resistance value determines the nonhomogeneous coordinate for a point
of load straight line. In turn, the two values VL ; IL are the homogeneous coordinates
because RL qVL qIL , where q is any nonzero real number. Homogeneous
coordinates have nite values. Further, the equation VL RL has the characteristic
fractionally linear view
VL V0
RL
:
Ri RL
2:23
Of course, similar expressions take place for currents and resistances for different
branches of active circuits and are used for measurement of circuit parameters [2].
That gives the solid grounds for considering the map RL ! VL as a projective
transformation of projective geometry [4, 5, 7]. In general, a projective transformation of points of one straight line into points of another line is set by a center S of
projection or three pairs of respective points in Fig. 2.14. Therefore, the innitely
remote point RL 1 corresponds to the nite point VL V0 . Also, fractionally
linear expressions of type (2.23) as projective transformations are used for
39
V0
V0
2
VL
VL
RL
RL
Ri
mL
VL
21
mL
mL
RL
mL
measuring instruments [8, 9]. In addition, a class of innite ideal points of projective geometry is introduced to represent innite resistances [3].
As the pairs of respective points, it is convenient to use the points of the
characteristic regimes, which can be easily determined at a qualitative level; that is,
the short circuit, open circuit, and maximum load power. In turn, the point of a
running regime is the fourth point.
We want to represent the running regime of load in the relative form regarding to
these three characteristic or base points. To do this, we may use a cross-ratio of four
points.
The projective transformations preserve a cross-ratio of four points. For values
R1L ; VL1 ; IL1 of initial or running regime, the cross-ratio mL has the view
m1L 0 R1L Ri 1
m1L 0 VL1
m1L
ILSC
IL1
R1L 0 Ri 0
R1L
;
:
R1L 1 Ri 1 Ri
V0
0
V0
V1 0
VL1
V0 1L
V02
;
2
VL V0 2 V0 V0 VL1
ILSC
I 1 I SC
0 L 1 L
2
IL 0
ILSC
SC
2 IL
SC
IL
2 0
ILSC IL1
:
IL1
2:24
2:25
2:26
R2L
;
R1L
2:27
40
2
1
m21
L 0 VL VL V0
VL2
VL1
;
V0 VL2 V0 VL1
2:28
SC 2 1
m21
L IL IL IL 0
2:29
R1L
1
21 :
2
RL mL
2:30
For the next regime change R2L ! R3L , the group property is given by
2
32 21 1
31 1
m3L m32
L mL mL mL mL mL mL :
Let us express the subsequent voltage value VL2 by the initial value VL1 and
regime change value m21
L . Using (2.28), we get the recalculation formula
V1
21
L
VL2
V0 mL
V1
:
V0
L
m21 1 1
V0
2:31
This formula allows nding a subsequent voltage value by an initial voltage and
transformation parameter m21
L . Also, this expression can be obtained from (2.23).
For an initial R1L and subsequent R2L value, we get the following system of the
equations:
8
1
< V 1 V0 RL 1
L
Ri R
: V 2 V0 RL 2
L
Ri R
2
41
Fig. 2.15 Identical regime changes for the different initial regimes
In the theory of the projective transformations, the xed points play an important
role. For their nding, Eq. (2.31) is solved as VL1 VL2 . It turns out the two real
roots, VL 0; VL V0 , which dene the hyperbolic transformation of hyperbolic
(Lobachevski) geometry. Physically, the xed point means such a regime when a
variable VL does not depend on the initial or subsequent value RL . It is evident for
SC; OC regimes.
If the roots of equation coincide, the xed point denes the parabolic transformation and, respectively, parabolic (Euclidean) geometry. If the roots are imaginary, geometry is elliptic (Riemannian).
In geometry, it is established that these three kinds of transformations (projective, afne, and Euclidean) exhaust possible variants of group transformations,
which underlie the denition of the metrics of a straight line. Thus, the geometrical
approach allows validating regimes determination, and both denitions of a regime
and its change are coordinated by structure of expressions and ensure the performance of group properties.
Reasoning from such a geometrical interpretation, it is possible to give the
following denition [12, 13]:
a circuit regime is a coordinate of point on load straight lines and axes of
coordinates;
a regime change is a movement of point on all the straight lines, which denes a
segment of corresponding length.
In this connection, it is possible to accept the following requirement (likewise to
metric space axioms):
independence or invariance of regimes and their changes from variables (regime
parameters) as type R; V; I;
the additive postulate of regimes changes;
assignment of equal regime change for various initial regimes.
42
2.2
2.2.1
Let us consider the Thvenin equivalent circuit with the variable internal resistance
Ri in Fig. 2.16.
In this case, a bunch of load straight lines with a parameter Ri is obtained at a
center G in Fig. 2.17. The unied equation of this bunch is given by
IL
1
VL V0 :
Ri
2:32
43
m21
L CI C2 C1 G BI B2 B1 G
0 R2L R1L 1
R2L
:
R1L
2:33
VLC2 0
VLC1 0
C1
:
C2
VL V0 VL V0
2:34
R1i
:
R2i
2:35
VLB2 0
VLC2 0
C2
:
B2
VL V0 VL V0
2:36
In expressions (2.34) and (2.36) the identical base points, which are the centers
of the two bunches of straight lines RL ,Ri , are used.
44
Case 3
If the regime is changed as point C1 ! C2 ! B2 , it is possible to speak about the
general or compound change
21
m21 m21
i mL
R1i R2L
:
R2i R1L
2:37
Then
21
VLB2 0
VLC2 0
VLB2 V0 VLC2 V0
C2
VL 0
VLC1 0
:
VLC2 V0 VLC1 V0
Finally, we obtain
m21
VLB2 0
VLC1 0
0 VLB2 VLC1 V0 :
VLB2 V0 VLC1 V0
2:38
21
L
VLB2
V0 m
V C1
:
V0
L
m21 1 1
2:39
V0
The compound regime and voltage changes, as point moving, are shown in
Fig. 2.18.
We may note that values RL , Ri have not any scales.
2.2.2
Let us consider the Norton equivalent circuit with the variable internal resistance or
conductance Yi and load conductance YL in Fig. 2.19.
45
2:40
The coordinate of the center G corresponding to I0 ILSC does not depend on the
value Yi . Physically, it means that the current across this element is equal to zero.
The element Yi can accept the two base or characteristic values, as 0; 1. The third
characteristic value is not present for Yi .
Let the relative regimes be considered for this case.
Case 1
Let the internal conductance Yi be equal to Yi1 and the load conductance varies from
YL1 to YL2 . In this case, a point of initial regime C1 ! C2 . If Yi is equal to Yi2 , a point
of initial regime B1 ! B2 . Therefore, the regime change, which is determined by
the load change (an own change), is expressed similarly to (2.33)
m21
L CV C2 C1 G BV B2 B1 G
0 YL2 YL1 1
YL2
:
YL1
2:41
46
ILC2 0
ILC1 0
:
ILC2 V0 ILC1 I0
2:42
Yi1
:
Yi2
2:43
ILB2 0 ILC2 0
:
ILB2 I0 ILC2 I0
2:44
Case 3
If regime is changed as C1 ! C2 ! B2 , the compound change is
21
m21 m21
i mL
Yi1 YL2
:
Yi2 YL1
2:45
Then
m21
ILB2 0 ILC1 0
0 ILB2 ILC1 I0 :
ILB2 I0 ILC1 I0
2:46
21
L
ILB2
I0 m
I C1
:
I0
L
m21 1 1
I0
The compound regime and current changes are shown in Fig. 2.21.
2:47
47
2.3
In the above Sect. 2.1, the examined regime parameters of the type V; I; R are
expressed among themselves by linear and fractionally linear expressions (2.1),
(2.23) and have projective properties. In turn, such an important energy characteristic, as a load-power, represents quadratic expression (1.6) and determines a
parabola in Fig. 1.6. This quadratic curve has similar projective properties that
permit to compare the regime of different circuits, to determine the deviation from
the power matching [11, 12, 14]. Let us consider these properties in detail.
To do this, we use the circuit in Fig. 2.12 and rewrite Eq. (2.21) of load straight
line in the following relative form
IL
VL
I 1
1 KV ;
V0
ILSC
2:48
where KV is the voltage transfer ratio. Also, we rewrite Eq. (1.6) of load power in
the similar relative form
P
PL
KV KV 2 ;
PSC
0
2:49
where PSC
0 is the maximum power of the voltage source V0 for SC regime.
Therefore, the task of equal regimes does not cause a problem; that is simply
corresponding equality of values KV ; I; P. But a deeper analysis will be useful,
which allows generalizing the justication of the equality of regimes and will be
used for considering a more complex case, the efciency of two-ports.
48
2.3.1
R1L
:
Ri
2:50
Ri
1:
Ri
This special case of the fourth point Ri determines the property of the harmonic
conjugacy of four points, which determines the symmetry of points Ri , Ri relatively to base points 0, .
49
(a)
(b)
For an arbitrary running load RL , we obtain, at once, the corresponding conjugate load resistance equals RL because cross-ratio (2.50)
0 RL RL 1 1:
2:51
So, we have the symmetry of points RL , RL relatively to the base points 0,
too. The points Ri ; Ri ; R1L ; R1L and so on pass into each other, as it is shown by
arrows. Physically, this symmetry corresponds to mapping of the region of power
consumption by a load on the region of return. Then, the mapping of points of
parabola of the region P [ 0 onto the region P\0 is realized from a point F. In this
case, the points 0, 1 of the axis KV are xed. Therefore, we obtain the following
condition:
0 KV RL KV RL 1
KV RL 0 KV RL 0
1:
KV RL 1 KV RL 1
2:52
50
2:53
The point F is formed due to the intersection of the tangential FX; FY at the xed
points. This point F is called a pole, and a straight line, passing through the xed
points 0, 1 is a polar 0T. The indicated symmetry is obtained relatively to the polar.
We may pass to the projective system of coordinates YFX. In this coordinate
system, a polar is considered as the innitely remote straight line. Therefore, our
initial parabola will be already a hyperbola, and the coordinate axes FX, FY are
asymptotes in Fig. 2.24. The point P
L M has a nite value.
In this case, a point on hyperbola is assigned as the rotation of radius-vector RF F
from the initial position at the point PLM .
The non-Euclidean distance R1 P
M is determined by a hyperbolic arc length of a
hyperbola; it will be later on shown in Sect. 4.4.
2.3.2
Also, the symmetry of points R1L ; R2L , as the points of equal power, is manifested by
arrows relatively to the point Ri or relatively to the straight line PLM P
L M in
Figs. 2.25 and 2.26. The points Ri ; Ri are the xed points. This symmetry
corresponds to points KV1 , KV2 also.
Using (2.49), we get the following condition:
KV1 KV2 1:
2:54
51
(a)
(b)
52
is assigned as the rotation of radius-vector RT T from the initial position at the point
0 or point 1. In the Cartesian coordinate system, this radius-vector RT T is the
parallel line to the axis KV .
Similar to (2.50), we introduce the other value of running regime
m1L Ri R1L 0 Ri
R1L Ri 0 Ri R1L Ri
1
:
R1L Ri 0 Ri
RL Ri
2:55
R1L Ri R2L Ri
1:
R1L Ri R2L Ri
2.3.3
2:56
points P
M ; Q; PM ; F onto the polar PM PM of the pole T. Let us consider this
harmonic conjugacy in detail.
For the pole T, the following correspondences take place. We believe the harmonic conjugate points 0; 1 relatively to the base points Q; T of the polar 0T. Then,
these points correspond to four points 0; 1; KV Q 0:5; KV T 1 of the axis
KV . The mutual mapping of the points 0; 1 relatively to xed points KV T; KV Q
is shown by arrows in Fig. 2.27.
Fig. 2.27 Mutual mapping of points relatively to the xed points onto the straight line TR1
53
In turn, there are such harmonic conjugate points R1L ; R2L relatively to the base
points A; T of running straight line TR1 . The harmonic conjugate points
KV1 ; KV2 ; 0:5; 1 of the axis KV correspond to these points of the line TR1 .
The mutual mapping of points KV1 ; KV2 relatively to the xed or base points
1; 0:5 is shown by arrows in Fig. 2.27 too. Therefore, we can constitute the
cross-ratio
0:5 KV1 KV2 1
1:
KV1 1 KV2 1 KV2 0:5
From this, we get expression (2.54) that conrms the accepted geometrical model.
Similarly, for the pole F, the following correspondences take place. We believe
54
References
1. Alexander, C.K., Sadiku, M.N.O.: Fundamentals of Electric Circuits, 5th edn. McGraw-Hill,
New York (2009)
2. Bhattacharyya, S.P., Keel, L.H., Mohsenizadeh, D.N.: Linear Systems: A Measurement Based
Approach. Springer, India (2014)
3. Bryant, R.E., Tygar, J.D., Huang, L.P.: Geometric characterization of series-parallel variable
resistor networks. Circ. Syst. I Fundam. Theory Appl. IEEE Trans. 41(11), 686698 (1994)
4. Frank, J.A.: Schaums Outline of Theory and Problems of Projective Geometry. McGraw-Hill,
New York (1967)
5. Glagolev, N.A.: Proektivnaia geometria (Projective Geometry). Nauka, Moskva (1963)
6. Irwin, J.D., Nelms, R.M.: Basic Engineering Circuit Analysis, 10th edn. Wiley, Hoboken
(2011)
7. Kagan, V.F.: Osnovania geometrii, Chasti II (Geometry Basics. Part II). Gostekhizdat, Moskva
(1956)
8. Mazin, V.D.: Method for raising the precision of measuring instruments and transducers.
Meas. Tech. 23(6), 479480 (1980)
9. Mazin, V.D.: Error of measurement in the compound-ratio method. Meas. Tech. 26(8), 628
629 (1983)
10. Penin, A.: Projectively-afne properties of resistive two-ports with a variable load.
Tehnicheskaia elektrodinamika 2, 3842 (1991)
11. Penin, A.: Denition of deviation from the matching regime for two-port circuit.
Electrichestvo 4, 3240 (1994)
12. Penin, A.: Fractionally linear relations in the problems of analysis of resistive circuits with
variable parameters. Electrichestvo 11, 3244 (1999)
13. Penin, A.: Determination of regimes of the equivalent generator based on projective geometry:
the generalized equivalent generator. World Acad. Sci. Eng. Technol. 2(10), 703711 (2008).
http://www.waset.org/publications/10252. Accessed 30 Nov 2014
14. Penin, A.: Projective geometry method in the theory of electric circuits with variable
parameters of elements. Int. J. Electron. Commun. Electr. Eng. 3(2), 1834 (2013). https://
sites.google.com/site/ijeceejournal/volume-3-issue-2. Accessed 30 November 2014
15. Penin, A., Sidorenko, A.: Determination of deviation from the maximum power regime of a
photovoltaic module. Moldavian J. Phys. Sci. 9(2), 191198 (2010). http://sfm.asm.md/
moldphys/2010/vol9/n2/index.html. Accessed 30 November 2014
Chapter 3
3.1
Introduction
55
56
which turn out identical to the load and element changes. Such invariant expressions permit to obtain convenient formulas of recalculation of the load current.
3.2
3.2.1
Let us consider an active two-pole circuit with a variable series resistance r0N and
base load resistance RL in Fig. 3.1. This circuit has a practical importance for a
voltage regulation.
At change of the load resistance from the short circuit SC to open circuit OC for
the specied series resistance r0N, a load straight line is given by
IL
VLOC VL
VL
ILSC ;
Ri
Ri
Ri
3:1
where ILSC is the SC current; the OC voltage VLOC and the internal resistance Ri are
the parameters of the Thvenin equivalent circuit in Fig. 3.2.
For our two-pole circuit, we have
VLOC V0
3:2
r1N IL
r0N
V0
rN
;
rN r0N
VL
rN
RL
Ri
IL
VL
OC
VL
RL
57
IL
1
Ri
SC1
IL
IL
r0 N
2
Ri
RL
SC2
r0 N
OC1
VL
V L VL
OC2
VL
IL
Ri r1N
ILSC
r0N rN
;
r0N rN
VLOC
V0
:
Ri
r0N 1 rr1NN r1N
3:3
3:4
1
2
Let the resistance r0N vary from r0N
to r0N
. Then, we get the two load straight
lines in Fig. 3.3 with the following parameters of the Thvenin equivalent circuit
VLOC1 ; VLOC2 ;
R1i ; R2i ;
ILSC1 ; ILSC2 :
Expression (3.1) allows calculating the load current for the given load resistance.
But, the recalculation of all the parameters of the equivalent circuit is necessary that
denes the disadvantage of this known Thvenin equivalent circuit.
3.2.2
According to Fig. 3.3, these load straight lines intersect into a point G. Physically, it
means that regime parameters do not depend on the value r0N; that is, the current
across this element is equal to zero at the expense of the load value.
In this case, the point G will be a bunch center of the load straight lines with the
parameter r0N.
Let us dene this bunch center. At once, it is obvious that the current across the
resistance r0N will be equal to zero if the voltage VN = V0. Then, we get a circuit in
Fig. 3.4 for the calculation of the load parameters.
58
IL
V0 r1N
G
rN
VL
+
-
R L <0
Fig. 3.4 Part of the circuit with zero current across the variable resistance r0N
V0
;
rN
3:5
r1N rN
V0 [ V0 :
rN
3:6
VLG
r1N rN :
ILG
3:7
Thus, an equation of straight line, passing through a point I1G ; V1G , has the form
IL ILG
VLG VL
:
Ri
Ri
3:8
So, the values I1G ; V1G , and Ri are the parameters of the generalized
Thvenin/Helmholtz equivalent generator in Fig. 3.5.
We note that besides a base energy source of one kind (a voltage source VLG )
there is an additional energy source of another kind (a current source ILG ) that it is
possible to consider as a corresponding theorem.
It is natural, when the value ILG 0, we obtain the known Thvenin/Helmholtz
equivalent generator. In this case VLG VLOC . Using the generalized generator, we
must recalculate the internal resistance only. It is the advantage of the offered
generator.
Let us show how the internal resistance Ri and changeable resistance r0N,
respectively influence on properties of the generalized equivalent generator in
Fig. 3.5. The corresponding family of the load straight lines is shown in Fig. 3.6.
59
Ri
IL
VL
IL
VL
RL
IL
r0 N
RL
Ri =0
r0 N
Ri
0
r 0 N =
0
Ri
VL
VL
r 0N
I
R i =
IL
rN Ri r1N
:
rN Ri r1N
3:9
V
r0N
rN r1N
;
rN r1N
3:10
I
r0N
rN ;
3:11
0
r0N
1;
3:12
which corresponds to the beam G0 and denes the zero-order source when the
current and voltage of the load are always equal to zero for all the load values.
The generalized equivalent generator that displays the zero-order generator is
presented in Fig. 3.7. VL = 0 because the internal resistance voltage V0i = VG
L.
60
IL = 0
Vi
VL
VL = 0
Ri
IL
RL
So, a variable element and internal resistance can have these three specied
characteristic values. These values are dened at a qualitative level. This brings up
the problem of determination in the relative or normalized form of the value r0N
regarding of these characteristic values. In this case, it is possibly to dene a kind of
active two-pole as an energy source and to compare the different circuits. Therefore,
the obvious value r0N = 0 is not characteristic one concerning the load.
Let us view the possible load characteristic values. Both the traditional values
RL = 0, RL = , and RG
L will be the characteristic values according to Fig. 3.6.
Physical sense of these values is clear.
3.2.3
Let us consider a common change of load RL and variable resistance r0N of the
circuit in Fig. 3.1. The corresponding load straight lines are shown in Fig. 3.8.
In this gure, we get the two bunches with parameters RL and r0N. Let an initial
1
2
and subsequent value is r0N
. Similarly, an
value of the variable element be r0N
initial value of the load equals R1L and subsequent one is R2L .
Fig. 3.8 Common change of
load RL and variable
resistance r0N
IL R =0
L
1
r 0N
F2
CI
r 0N
F1
BI
B2
RL
RL
C2
C1
R i = RL
B1
0
r 0N
A2
IL
RL=
CV
r0N
A1
r 0N
BV
VL
VL
G
RL
61
Let us consider the straight line of the initial load R1L . The three straight lines
I
0
V
with the characteristic values r0N
; r0N
; r0N
and the two lines with the parameters
1
2
1
r0N , r0N intersect this line RL . The points A1, 0, B1, C1, F1 are points of this
intersection. In turn, the points A2, 0, B2, C2, F2 are points of intersection of the line
R2L . Therefore, a projective map (conformity) of one line R1L on the other line R2L
takes place. This conformity is set by a projection center G.
1
. The three straight lines with the
Similarly, we consider the straight line r0N
characteristic values of load RL and two lines with parameters R1L , R2L intersect this
1
line r0N
. The points G, CV, C1, C2, CI are points of this intersection.
2
In turn, the points G, BV, B1, B2, BI are points of intersection of the line r0N
.
1
2
Therefore, a projective map (conformity) of one line r0N
on the other line r0N
takes place. This conformity is set by the projection center 0.
The above conformities of straight line points represent projective geometry
transformations. As shown earlier, it is convenient to use projective geometry for
analysis of circuits with variable elements. The projective transformation is also set
by three pairs of respective points. As pairs of these points, it is convenient to use
the points corresponding to the characteristic values of the load and variable element. The projective transformations preserve a cross ratio of four points; otherwise, a cross ratio is an invariant of these transformations. Further, we will show the
use of such invariants.
Case 1 Denition of the relative operating regime at the load change
Let the series resistance r0N be equal r10N and the load resistance varies from R1L to
R2L . In this case, a point of initial regime C1 C2, as it is shown in Fig. 3.8. If the
2
series resistor is equal to r0N
, a point of initial regime B1 B2. The given points CI,
C2, C1, CV, G correspond to the points BI, B2, B1, BV, G.
Using (2.24), we may constitute the cross ratio for the initial regime in the form
m1L CI C1 CV G BI B1 BV G
0 R1L 1 RG
L
R1L
R1L
:
RG
L
3:13
The points CI, G (and BI, G) are the base points, and CV (and BV) is a unit one.
Likewise, we have the base values RL 0; RG
L , and a unit RL = .
This determination of relative regime does not depend on r0N. Therefore, we
must use the load voltage for the cross ratio calculation. In this case, the base points
CI, BI give the common base point VL = 0.
Similarly to (2.23), the equation VL(RL) from (3.8) has the fractionally linear
view; it is possible by formalized method to express the cross ratio right now
62
Fig. 3.9 Mutual conformity
of the load voltage, resistance,
and cross ratio
CI
C2
VL
C1
CV
C1
C2
VL
21
RL
VL
VL
mL
RL
mL1
m L1
RL
VLC1 0
VLCV 0
VLC1 VLG VLCV VLG
V B1 0
VLBV 0
VLG B1L
:
VL VLG VLBV VLG
VLB1
VLBV
VL
mL1
0
CV
RL
3:14
In this case, we have the base values VL 0; VLG , and a unit value VLCV .
The corresponding values of this cross ratio are shown in Fig. 3.9.
So, it is possible to consider cross ratio (3.13) and (3.14) as a projective coordinate of the initial or running regime points VLC1 , R1L . This coordinate is expressed
by invariant (identical) manner by various regime parameters.
The cross ratio for the subsequent regime
m2L CI C2 CV G BI B2 BV G
0 R2L 1 RG
L
R2L
R2L
:
RG
L
VLC2 0
VLCV 0
CV
C2
G
VL VL VL VLG
V B2 0
VLBV 0
VLG B2L
:
VL VLG VLBV VLG
VLB2
VLBV
3:15
3:16
Similarly to (2.33) and (2.34), the regime change has the view
m21
L CI C2 C1 G BI B2 B1 G
0 R2L R1L RG
L
R2L
R1L
m2L m1L ;
R2L RG
R1L RG
L
L
3:17
63
VLC2
VLC1
VLC2 VLG VLC1 VLG
V B2 0
VLB1 0
VLG B2L
:
VL VLG VLB1 VLG
C2 C1 G
m21
L 0 VL VL VL
VLB2
VLB1
3:18
This change is expressed by invariant manner through various regime parameters. Therefore, usually used regime changes by increments (as formal) are
eliminated.
In turn, the values VLG , RG
L are the scales for normalizing the voltage and
resistance values. Then, expressions (3.17) and (3.18) represent the relative
regimes. It permits to compare or set the regime of different circuits with various
parameters.
Let us remind properties of a cross ratio. If the components VLC1 , VLC2 of
expression (3.18) are interchanged, we get
m12
L
1
:
m21
L
3:19
3:20
Let us obtain the subsequent voltage value from expression (3.18). Then, we
have
VLC2
3:21
C1 0
A1 0
:
C1 F1 A1 F1
3:22
The points 0, F1 are chosen as the base points; that will be explained later. In
turn, A1 is a unit point.
64
The cross ratio for the points 0, C2, A2, F2 of the line R2L has the same value
m1i 0 C2 A2 F2
C2 0
A2 0
:
C2 F2 A2 F2
3:23
VLC1 0
V A1 0
A1L
:
C1
G
VL VL VL VLG
3:24
VLB1 0
VLA1 0
:
G A1
B1
VL VL VL VLG
3:25
The distance between the points of the initial and subsequent regimes on the
straight line R1L
2
1
m21
i mi mi
VLB1 0
VLC1 0
C1
0 VLB1 VLC1 VLG :
G
B1
VL VL VL VLG
3:26
VLB2 0
VLC2 0
:
G C2
B2
VL VL VL VLG
3:27
For a given load, the equation VL(Ri) = VL(r0N) from expression (3.8) has
fractionally linear view; it is possible, by the formalized method, to express cross
ratio (3.24) and (3.26) for the resistance Ri and r0N
m1i 0 VLC1 VLA1 VLG R0i R1i RIi RVi
1
RG
L Ri 1 0
R1i RG
1 RG
R1 RG
L
L
i 1 L;
1
10
Ri 0
Ri
0
1
I
V
m1i 0 VLC1 VLA1 VLG r0N
r0N
r0N
r0N
1
I
V
1 r0N
r0N
r0N
1
I
I
V
r0N
1
r0N
1 r0N
r0N
:
I rV
1 rV
1 rV
r0N
r0N
r0N
0N
0N
0N
B1 C1 G
0 2 1 V
m21
i 0 VL VL VL Ri Ri Ri Ri
2 1
RG
L Ri Ri 0
R2i RG
R1i RG
L
L
;
R2i
R1i
3:28
3:29
3:30
65
B1 C1 G
0
2
1
V
m21
i 0 VL VL VL r0N r0N r0N r0N
2
1 V
1 r0N
r0N
r0N
2
1
1
V
r0N
1
r0N
1 r0N
r0N
:
2 rV
1 rV
2 rV
r0N
r0N
r0N
0N
0N
0N
3:31
3:32
VLB2 0
VLC1 0
0 VLB2 VLC1 VLG :
VLB2 VLG VLC1 VLG
3:33
3.2.4
3:34
Example
Let us consider a circuit with given values in Fig. 3.10. Dimensions of these values
are not indicated for simplifying.
1
0:5.
Let the initial value of the resistance r0N be equal to r0N
Parameters (3.2)(3.4) of the Thvenin equivalent circuit
VLOC1 V0
rN
5
9:0909;
10
1
5 0:5
rN r0N
R1i r1N
ILSC1
1
r0N
rN
0:5 5
1:4545;
1
1
0:5 5
r0N rN
VLOC1
6:25:
R1i
66
r1N IL
r0N
V0
VL
1
rN 5
10
RL
VLOC2 10
5
8:3333;
51
15
1:8333;
15
R2i 1
ILSC2 4:5454:
ILCI 6:25;
VLBV 8:3333;
ILBI 4:5454:
V0 10
2;
5
rN
VLG
r1N rN
15
10 12:
V0
5
rN
IL
1
r 0N
6
F2
RL
CI
r 0N
BI
C3
B2
F1
C1
B1
r 0N=
4
A1
r 0N
RL
C2
r 0N
CV
6
BV
VL
10
12
VL
RL
A2
IL
67
Ri
VL= 0
VL
12
IL
12
RL
VLG
6:
ILG
51
0:8333;
51
rN 5;
V
r0N
I
r0N
R0i RG
L 6;
0
r0N
1:
ILC1 ILG
VLG VLC1
12 6
2:1251:
2
1
1:4545
Ri
VLC1
6
2:8233:
C1
2:1251
IL
R2L 1:4256:
68
2
If the series resistor is equal to r0N
, the initial regime point B1 B2. Then,
VLB1 5:0524;
VLB2 3:6455:
R1L
2:8233
0:32:
2:8233
6
R1L RG
L
6 12 9:0909 12
V B1 0
VLBV 0
m1L B1L
VL VLG VLBV VLG
5:0524 0
8:3333 0
0:7272 2:2726 0:32:
5:0524 12 8:3333 12
m1L
R2L
1:4256
0:192:
G
2
RL RL 1:4256 6
R2L
R1L
0:192 0:32 0:6:
R2L RG
R1L RG
L
L
4:5:
C1
G
21
6 0:6 1 12
9:6
VL mL 1 VL
We consider the load resistance R3L = 0.7163 that corresponds to the point C3 and
cross ratio
m3L CI C3 CV G
RL
0:7163
0:1066:
0:7163
6
R3L RG
L
69
12 4:5 0:5555
30
3:0:
4:5 0:5555 1 12 10
Case 2 Recalculation of the load voltage at the series resistance r0N change
1
2
!r0N
. In this
Let the load resistance be equal to R1L and the series resistance r0N
case, the initial regime point C1 B1.
Regime change (3.31)
2
1
V
m21
i 1 r0N r0N r0N
1
V
r0N
r0N
0:5 0:8333
V 1 0:8333 0:7272:
2
r0N r0N
We consider the load voltage of the initial regime VLC1 6. Then, subsequent
voltage value (3.32)
VLB1
12 6 0:7272
5:0523:
6 0:7272 1 12
Similarly, let the load resistance be equal to R2L ; the point of the initial regime
C2 B2. We consider the load voltage of the initial regime VLC2 4:5. Then, the
subsequent voltage value
VLB2
C2
G
21
4:5 0:7272 1 12
VL mi 1 VL
Case 3 Recalculation of the load voltage at the common change of the load
RL and resistance r0N
Let the regime change be given as C1 C2 B2.
Common regime change (3.33)
21
m21 m21
i mL 0:7272 0:6 0:4363:
C1
G
21
6 0:4363 1 12
VL m 1 VL
70
3.3
3.3.1
Let us consider an active two-pole circuit with a base load conductivity YL and
variable auxiliary load or shunt regulating conductivity yN in Fig. 3.13. This circuit
has a practical importance for a current regulation.
At change of the load conductivity from the short circuit SC to open circuit OC
for the specied shunt conductivity yN, a load straight line is given by expression
(3.1)
IL Yi VLOC Yi VL ILSC Yi VL ;
3:35
where VLOC is the OC voltage; the internal conductivity Yi and SC current ILSC are the
parameters of the Norton equivalent circuit in Fig. 3.14.
For our two-pole circuit we have
VLOC V0
Yi
y0N
;
yN y0N
y0N yN
y1N ;
y0N yN y1N
ILSC Yi VLOC V0
y0N y1N
:
y1N y0N yN
3:37
3:38
y1N IL
y0N
V0
3:36
VL
yN
YL
IL
VL
SC
IL
Yi
YL
Fig. 3.14 Norton equivalent circuit with the variable internal conductivity and current source
71
IL
IL
YL
SC1
IL
SC2
IL
OC2
VL
OC1
VL
VL
VL
Y i y2
N
Y i y1
N
Let the conductivity yN varies from y1N to y2N . Then, we get the two load straight
lines in Fig. 3.15 with the following parameter of the Norton equivalent circuit
VLOC1 ; VLOC2 ;
Yi1 ; Yi2 ;
ILSC1 ; ILSC2 :
Expression (3.35) allows calculating the load current for the given load resistance. But, the recalculation of all parameters of the equivalent circuit is necessary
that denes the disadvantage of this known Norton equivalent circuit.
3.3.2
According to Fig. 3.15, these load straight lines intersect into a point G. Physically,
it means that regime parameters do not depend on the value yN; that is, the current
through this element is equal to zero at the expense of the load value.
In this case, the point G will be a bunch center of load straight lines with the
parameter yN.
Let us dene this bunch center. At once, it is visible that the current across yN
will be equal to zero if the voltage VN = 0. Then, we get a circuit in Fig. 3.16 for the
calculation of the load parameters.
y1N I L
y0N
V0
VN=0
VL
Y L <0
Fig. 3.16 Circuit with the zero voltage of the variable conductance yN
72
3:39
ILG
:
y1N
y0N
V0 :
y1N
3:40
ILG
y1N :
VLG
3:41
Thus, an equation of a straight line, passing through a point I1G ; V1G , has the form
IL ILG Yi VLG VL :
3:42
So, the values ILG ; VLG , and Yi are the parameters of the generalized
Norton/Mayer equivalent generator in Fig. 3.17.
We note that besides a base energy source of one kind (a current source IG
L)
there is an additional energy source of another kind (a voltage source VG
L ) that it is
possible to consider as a corresponding theorem.
It is natural, when the value VG
L = 0, we obtain the known Norton/Mayer
SC
equivalent generator. In this case IG
L = IL . Using the generalized generator, we must
VL
IL
VL
IL
Yi
YL
73
IL
IL
yN
I
Y i=0
YL1
VL
0
0
y N =
Yi
yN
Yi
VL
yN
Y i =
V
recalculate the internal conductivity value only. It is the advantage of the offered
generator.
Let us demonstrate how the internal conductivity Yi and respectively the
changeable conductivity yN influence on properties of the generalized equivalent
generator in Fig. 3.17. The corresponding family of the load straight lines is shown
in Fig. 3.18.
Further, we use expression (3.37).
The conductivities Yi, yN have the following characteristic values:
YiI 0;
yIN y0N ;
3:43
3:44
y0N 1;
3:45
which corresponds to the beam G0 and denes the zero-order source when the
current and voltage of the load are always equal to zero for all load values.
The generalized equivalent generator that exhibits the zero-order generator is
presented in Fig. 3.19. The load voltage VL = 0 because V0i = VG
L.
So, a variable element and internal conductivity can have these three specied
characteristic values. This brings up the problem of determination in the relative or
normalized form of the conductivity value yN regarding of these characteristic
values. Therefore, the obvious value yN = 0 is not characteristic ones concerning the
load.
74
IL= 0
VL
0
IL
Yi
VL= 0
0
Vi
YL
Let us view the load characteristic values. Both the traditional values YL = 0,
YL = , and YLG will be the characteristic values according to Fig. 3.18.
3.3.3
IL
G
YL
YL=
IL
F2
YL
F1
YL
I
yN
G
CI
C2
BI
G
VL
B2
C1
B1
BV
A1
V
yN
A2
yN
2
yN
YL=0
CV
yN
VL
75
m1L CV C1 CI G
3:46
The cross ratio for the points BV ; B1 ; BI ; G of the line y2N has the same value
m1L BV B1 BI G
B1 BV BI BV
:
B1 G
BI G
3:47
The points CV, G (and BV, G) are the base points, and CI (and BI) is a unit one.
Correspondingly, we have the base values yL = 0, yG
L , and a unit yL = . This
determination of relative regime does not depend on YN. Therefore, we must use the
load current for the cross ratio calculation. In this case, the base points CV, BV give
the common base point IL = 0.
Then, the cross ratio is expressed by current components
m1L 0 ILC1 ILCI ILG
ILC1 0
ILCI 0
:
ILC1 ILG ILCI ILG
3:48
The corresponding values of this cross ratio are shown in Fig. 3.21.
Similarly, the subsequent regime cross ratio
m2L 0 ILC2 ILCI ILG
ILC2 0
ILCI 0
:
ILC2 ILG ILCI ILG
3:49
CI
C2
C1
CI
C2
C1
IL
IL
IL
mL
YL
IL
IL
mL
YL
21
mL
mL
YL
CV
YL
76
YL2
YL1 0
:
YL2 YLG YL1 YLG
C2 C1 G
m21
L 0 IL IL IL
ILC2 0
ILC1 0
:
ILC2 ILG ILC1 ILG
3:50
3:51
This change is expressed by invariant manner through various regime parameters. Therefore, usually used regime changes by increments (as formal), are
eliminated.
In turn, the values ILG , YLG are the scales for normalizing the current and conductivity values. Then, expressions (3.50) and (3.51) represent the relative regimes.
It permits to compare or set the regime of the different circuits with various
parameters.
Also, the group property takes place
2
32 21 1
31 1
m3L m32
L mL mL mL mL mL mL :
3:52
Let us obtain the subsequent current value from expression (3.51). Then, we
have
ILC2
3:53
C1 0
A1 0
:
C1 F1 A1 F1
3:54
The points 0, F1 are chosen as the base points; that will be explained later. In
turn, the point A1 is a unit point.
The cross ratio for the points 0; C2 ; A2 ; F2 of the line YL2 has the same value
m1i 0 C2 A2 F2
C2 0
A2 0
:
C2 F2 A2 F2
3:55
77
ILC1 0
ILA1 0
:
ILC1 ILG ILA1 ILG
3:56
ILB1 0
ILA1 0
:
G A1
B1
IL IL IL ILG
3:57
The distance between the points of the initial and subsequent regimes on the
straight line YL1
2
1
m21
i mi mi
ILB1 0
ILC1 0
0 ILB1 ILC1 ILG :
G C1
B1
IL IL IL ILG
3:58
ILB2 0
ILC2 0
:
ILB2 ILG ILC2 ILG
3:59
For a given load, the expression IL = (Yi) = IL = (yN) has fractionally linear view;
it is possible, by the formalized method, to express cross ratio (3.56) and (3.58) for
the conductivity Yi and yN
m1i 0 ILC1 ILA1 ILG Yi0 Yi1 YiV YiI
YLG Yi1 1 0
:
10
Yi1 0
Yi1
:
y1N yIN yVN yIN y1N yIN
B1 C1 G
2 1 I
m21
i 0 I1 I1 I1 0 Yi Yi Yi
B1 C1 G
0 2 1 I
m21
i 0 IL IL IL yN yN yN yN
:
y2N yIN y1N yIN y2N yIN
3:60
3:61
3:62
3:63
78
3:64
ILB2 0
ILC1 0
C1
0 ILB2 ILC1 ILG :
G
B2
IL IL IL ILG
3:65
3.3.4
3:66
Example
y0N
1
yN y0N
y1N y0N
10
1
9:0909;
0:1 1
0:1 1
2 0:7096;
y1N
0:1 1 2
y1N y0N y1N
Yi1 VLOC1 0:7096 9:0909 6:4516:
y1N IL
y0N
V0
10
1 y
N
VL
YL
79
G
YL
IL
IL
F2
10
6
4
2
C2
BI
C1
B2
B1
4
CV
BV
0
4
YL
CI
VL
yN
YL
F1
6
2
yN
10
VL
yN
A1
yN
A2
y N =
VLOC2 10
ILSC2 4:2561:
ILCI 6:4516;
VLBV 3:7055;
ILBI 4:2561:
VLG
y0N
1
V0 10 5:
2
y1N
80
VL
0
Yi
IL
VL=0
5
-
YL
ILG
2:
VLG
YiV 1;
Yi0 YLG 2;
y0N 1:
ILG ILC1
10 3:5
4:160:
5
1
0:7096
Yi
ILC1
3:5
0:8413:
VLC1 4:16
YL2 1:635:
81
If the shunt conductivity is equal to y2N = 1.6987, the initial regime point
B1 B2. Then, we obtain
ILB1 1:8;
ILB2 2:5:
YL1
YL1
0:8413
0:296:
YLG 0:8413 2
3:5 10 6:4516 10
0:5384 1:8181 0:296:
m1L
ILC2 0
ILCI 0
CI
C2
G
IL IL IL ILG
YL2
YL2
YL1
0:45 0:296 1:5202:
G 1
YL YL YLG
10 3:5 1:5207
4:5:
3:5 0:5207 10
y1N yIN
0:1 1
0:407:
82
G
3:5
0:407 1 10
ILC1 m21
1
I
L
i
Case 3 Recalculation of the load current at the common change of the load
YL and conductivity yN
Let the common change of regime be given as C1 C2 B2.
Common regime change (3.65)
21
m21 m21
i mL 0:407 1:5207 0:6187:
3.4
Let us consider an active two-pole A with a load conductivity YL1 and variable
auxiliary conductivity YL2 in Fig. 3.25. For convenience of a mathematical
description, the variable element YL2 is taken out from the two-pole contour.
This circuit (as a two-port network) is described by the following system of the
Y parameter equations
"
I1
I2
Y11
Y12
YL2
I1
A
V1
YL1
Fig. 3.25 Active two-pole with a load conductivity YL1 and variable element YL2
3:67
83
I2SC;SC Y20 V0 ;
3.4.1
3:68
2
Y12
Y12 Y20
V1 Y10
V0 :
YL2 Y22
YL2 Y22
3:69
The load straight line with the parameter YL2 is shown in Fig. 3.26.
Let us consider the short circuit current ISC
1 . In this case V1 = 0 and
I1SC
Y12 Y20
Y10
V0 :
YL2 Y22
3:70
1 SC
I ;
Yi 1
3:71
I1
YL1=
SC
YL1
I1
I1
YL1 = 0
0
V1
OC
V1
YL2
84
2
Y12
YL2 Y11 DY
1
Ri
YL2 Y22
YL2 Y22
3:72
is the internal conductivity (internal resistance Ri) of the circuit relatively to the load
YL1; DY is the determinate of the matrix Y parameters.
Taking into account the entered parameters, Eq. (3.69) becomes as
I1 Yi V1 I1SC :
3:73
3:74
3:75
3.4.2
Let us study features of load straight line (3.69). This expression represents a bunch
G
of the load straight lines with the parameter YL2. To nd the coordinates VG
1 , I1 of
the bunch center G of these lines, it is convenient to use the extreme parameter
values; that is, YL2 = 0, YL2 = . These lines are shown in Fig. 3.27.
In this case, expression (3.69) gives the following system of equations
(
Y2
I1 0 Y11 Y1222 V1 Y10
I1 1 Y11 V1 Y10 V0 :
Y12 Y20
Y22
V0
3:76
85
I1
G
Y L1
YL1
I1
SC
I1
I1
OC
V1
G
V1
V1
YL2=
YL2
V1
YL2= 0
3:77
3:78
G
The obtained values VG
1 , I1 , and the internal conductivity Yi allow to represent
Eq. (3.69) or (3.74) in the other kind
3:79
G
Thus, we obtain an equation of a straight line passing through a point IG
1 , V1 . So,
G
G
the values I1 , V1 , and Yi are the parameters of the generalized Norton/Mayer
equivalent circuit in Fig. 3.17 of Sect. 3.2. In turn, this equation corresponds to
expression (3.42).
G
Let us note the bunch center G corresponds to such a voltage VG
1 and current I1
G
of the load YL1 = YL1 when the current of the element YL2 is equal to zero.
According to this condition, from (3.67), it is also possible to nd values (3.77) and
G
(3.78) of VG
1 , I1 . Then, the corresponding negative load conductivity
G
YL1
I1G
Y11 Y20 Y10 Y21
:
Y20
V1G
3:80
In the above case, the bunch center G is in the second quadrant of the coordinate
G
system and so VG
1 < 0, I1 > 0. It is natural to consider a case when the bunch center
G is in the fourth quadrant shown in Fig. 3.28.
86
Fig. 3.28 Bunch center is in
the fourth quadrant
I1
YL1
YL2 =
YL2 =0
V1
I1
V1
G
R L1
I1G
:
V1G
3:81
Y20
V0 :
Y12
Y11 Y20
I1G Y10
V0 :
Y12
3:82
3:83
1
V1 V1G :
Ri
3:84
G
So, the values IG
1 , V1 , and Ri are the parameters of the generalized
Thvenin/Helmholtz equivalent generator in Fig. 3.5 of Sect. 3.1. In turn, Eq. (3.84)
corresponds to expression (3.8).
The position of center G (in the second or fourth quadrant) is dened by the kind
or type of an active two-pole as an energy source. If an active two-pole shows more
properties of current source, the case of Fig. 3.27 takes place. If it shows more
properties of voltage source, we have the case of Fig. 3.28.
Let us demonstrate how the internal conductivity Yi and respectively the conductivity YL2 influence on properties of the generalized equivalent generator.
87
Y22 Yi DY
:
Y11 Yi
3:85
I
YL2
DY
;
Y11
3:86
V
YL2
Y22 ;
3:87
;
G
Y20
V1
Y20 Y12
Y22
;
Y10
Yi0
0
YL2
3:88
3.4.3
Let us consider the electric circuit with the base load conductivity YL1 and auxiliary
load conductivity YL2 in Fig. 3.29.
This circuit may be considered as an active two-port A network relatively to the
specied loads in Fig. 3.30.
Taking into account the specied directions of currents, this network is described
by (3.67)
2
3
" #
" #
I1SC;SC
I1
V1
Y11 Y12
5
4
Y12 Y22
I2
V2
I2SC;SC
" # " #
V1
3
1:2
0:2
0:2 0:95
V2
2
88
Fig. 3.29 Active two-pole
with the load conductivity YL1
and variable element YL2
I1
y1N
1.25
y0N
V0
5
y2N
2.5
I0
yN
0.833
0.625
V1
y1
0.25
I2
y2
0.25
YL1
V2
YL2
V1
A
YL1
V2
YL2
Y11 y1 y1N
Let the initial value of the conductivity YL2 be equal to Y1L2 = 0.5. Parameters
(3.70)(3.72) of the Norton equivalent circuit
I1SC1
I1CI
89
Y12 Y20
0:2 0:4
Y10 1
5 3:2758:
V0 0:6
0:5 0:95
YL2 Y22
1
YL2
Y11 DY 0:5 1:2 1:1
1:1724;
1
0:5 0:95
YL2 Y22
1
V1CV 1 I1SC1 2:7941:
Yi
Yi1
V1OC1
Let the subsequent value of the conductivity YL2 be equal to Y2L2 = 2. The
corresponding parameters of the Norton equivalent circuit
I1SC2 I1BI 3:1355;
Yi2 1:1864;
Y L1
G
I1
G
I1
YL1=
F2
Y L2
15
2
Y L1=1
CI
3.2758
3.135
BI
C2
1.5073
1.4341
0.979
B2
V1
C1
B1
Y L1=0.5
BV
Y L1=0
CV
V1
10
A1
Y L2
0
Y L2
V
Y L2
Y L2
90
I1G
Y Y20 Y10 Y21
11
y1N y1 1:5:
Y20
I1G
DY
1:1
0:9166;
1:2
Y11
V
YL2
Y22 0:95;
Y Y20
0
1:5; YL2
Y22 12
y2N y2 1:0833:
Y10
I
YL2
YiV 1;
G
Yi0 YL1
So, the case of the generalized Norton/Mayer equivalent generator takes place.
Therefore, we will use the corresponding relationships of Sect. 3.2.
Case 1 Recalculation of the load current at the load change
Let the conductivity YL2 be equal to Y1L2 and load conductivity varies from Y1L1 = 0.5
to Y2L1 = 1. In this case, the initial regime point C1 C2; the load current of the
initial regime IC1
1 = 0.979.
Cross ratio (3.46) for the initial regime
m1L CV C1 CI G
1
YL1
0:5
0:25:
1 YG
0:5
1:5
YL1
L
2
YL1
1
0:4:
2 YG
1
1:5
YL1
L
15 0:979 1:6
1:5073:
0:979 0:6 15
91
1
0
0
YL2
YL2
Y V YL2
L2
V
I
1
I
YL2 YL2 YL2 YL2
2
0
V
0
YL2
YL2
YL2
YL2
V
I
2 YI
YL2
YL2
YL2
L2
2 1:0833
4 1:0571 4 0:2642:
2 0:9166
C2
G
21
1:5073 0:9459 1 15
I1 mi 1 I1
Case 3 Recalculation of the load current at the common change of the load
YL1 and conductivity YL2
Let the common change of regime be given as C1 C2 B2.
The common regime change
21
m21 m21
i mL 0:9459 1:6 1:5134:
3.5
92
2
1
3
4
V0
VOUT
12
I2
11
V1
V2
10
8
through a match circuit 7 (general case) and a bias voltage source 8 are connected to
one of the lateral resistance 4.
In turn, the negative resistance 6 is determined by a negative-impedance converter with an operation amplier 9. Resistor 10 is an initial positive resistance,
corresponding to the negative resistance 6; resistances 11, 12 are a feedback circuit.
The appliance in Fig. 3.32 work as follows. Let the bias voltage 8 be equal to
zero at rst. The power supply voltage source 1 sets the load current I1 and voltage
V2. The operation amplier 9 gains this voltage V2; the output voltage VOUT
species the current I2. The voltage for the inverse input of the amplier 9 is equal
to V2 and the same current, as I2, flows across the resistor 10. Therefore, the
operation amplier 9 with the resistor 10 is an energy source. This energy source, as
the negative resistance 6, sets the voltage V2, current I2; the ratio V2/I2 is equal to
the resistance 10 value. If the resistance 10 is chosen by necessary way, the load
current will be constant. In turn, the voltage V0 determines the value of the load
current.
For convenience of use of expression (3.86), we may consider a three-port in
Fig. 3.33, which is equivalent to the structure in Fig. 3.32 and the circuit in
Fig. 3.29. In particular, conductivities y0N, y1N correspond to the longitudinal
resistances; conductivities yN, y1 correspond to the lateral resistances of the line.
Also, we have the load conductivity YL1, negative conductivity YL2, and bias
voltage VBS.
For example, let us consider a circuit of ORCAD model in Fig. 3.34. The
negative resistance, by expression (3.86), is equal to 6.63488 k.
The results of modeling for various values of the load resistance RL1 and voltages V0, VBS are given in Tables 3.1 and 3.2.
93
y0N
I1
y1N
V1
V0
yN
y1
YL1
V2
y2N
I2
YL2
y2
VBS
Fig. 3.33 Equivalent circuit to the structure of stabilization of the load current
I
R1
R6
V
V
8k
6.35V
8k
I0
R3
R7
20k
40k
1.5k
R
V
R11
I
8
V+
+
R2
V2
6.63488k
R12
V-
2k
.
4
L1
20k
OUT
R4
R10
100k
LM358
30V
20k
2V
BS
We consider Table 3.1. It is visible, that the output voltage VOUT of the operational amplier reaches the maximum value at the load resistance of 3.0 k.
Also, we note that the voltage source 1 does not give but consumes energy in the
given load values because the negative current I0.
94
Table 3.1 V0 = 3 V,
VBS = 0 V
RL1 (k)
I1 (mA)
I0 (mA)
V2 (V)
I2 (mA)
VOUT (V)
0.001
0.451
0.076
5.022
0.757
20.17
1.5
0.4508
0.177
6.15
0.9273
24.7
3.0
0.4506
0.2784
7.276
1.097
29.21
RL1 (k)
I1 (mA)
I0 (mA)
V2 (V)
I2 (mA)
VOUT (V)
0.001
0.4502
0.345
4.175
0.328
10.75
1.5
0.450
0.242
5.3
0.497
15.26
6.1
0.4496
0.0067
8.749
1.017
29.1
Let us consider Table 3.2. The output voltage VOUT of the operational amplier
reaches the maximum value at the load resistance of 6.1 k. It is the advantage of
utilization of a bias voltage source. Also, the voltage source 1 gives energy practically for all load values. Therefore, the use of a bias voltage source improves an
energy effectiveness of power supply voltage source.
References
1. Alexander, C.K., Sadiku, M.N.O.: Fundamentals of Electric Circuits, 5th edn. McGrawHill,
New York (2009)
2. Chatzarakis, G.E.: A new method for nding the Thvenin and Norton equivalent circuits.
Engineering 2(05), 328336 (2010)
3. Gluskin, E., Patlakh, A.: An ideal independent source as an equivalent 1-port. arXiv preprint
arXiv:1108.4592 (2011)
4. Hashemian, R.: Hybrid equivalent circuit, and alternative to Thvenin and Norton equivalents,
its properties and application. In: Midwest Symposium on Circuits and Systems, MWSCAS
(2009)
5. Hosoya, M.: Derivation of the equivalent circuit of a multi-terminal network given by
generalization of Helmholtz-Thevenins theorem. Bull. Coll. Sci. Univ. Ryukyus. 84, 13
(2007)
6. Irwin, J.D., Nelms, R.M.: Basic Engineering Circuit Analysis, 10th edn. Wiley, Hoboken
(2011)
7. Johnson, D.: Origins of the equivalent circuit concept: the voltage-source equivalent. Proc.
IEEE 91(4), 636640 (2003)
8. Karris, S.T.: Circuit Analysis I with MATLAB Computing and Simulink/SimPower Systems
Modeling. Orchard Publications, Fremont (2004)
9. Penin, A.: Utilization of the projective coordinates in the linear electrical network with the
variable regimes. Buletinul Academiei de Stiinte a Republicii Moldova, Fizica si tehnica. 2,
6471 (1992)
References
95
10. Penin, A.: Characteristics of modied equivalent generator of active two-terminal network
with variable resistor. Electrichestvo 4, 5559 (1995)
11. Penin, A.: Linear-fractional relations in the problems of analysis of resistive circuits with
variable parameters. Electrichestvo 11, 3244 (1999)
12. Penin, A.: Generalized Thvenin/Helmholtz and Norton/Mayer theorems of electric circuits
with variable resistances. WSEAS Trans. Circ. Syst. 13, 104116 (2014). http://www.wseas.
org/cms.action?id=7648. Accessed 30 Nov 2014
13. Penin, A., Sidorenko, A.: Transmission of measuring signals and power supply of remote
sensors. In: Bonca, J., Kruchinin, S. (eds.) Nanotechnology in the Security Systems. NATO
Science for Peace and Security Series C: Environmental Security, pp. 267281. Springer,
Dordrecht (2014)
14. Vandewalle, J.: Shortcuts in circuits and systems education with a case study of the
Thvenin/Helmholtz and Norton/Mayer equivalents. In: International Symposium on the
Circuits and Systems (ISCAS), IEEE (2012)
Chapter 4
Two-Port Circuits
4.1
4.1.1
Let us consider a two-port TP1 circuit in Fig. 4.1. A variable voltage source V1 is
the load of this two-port. The voltage V1 (independent quantity) and current I1 are
the parameters of operating or running load regime. Therefore, we use the approach
of Sect. 2.1.1.
As it is known [1, 2], the system of equation of this two-port has the view
I0 Y00 V0 Y10 V1
I1 Y10 V0 Y11 V1 :
4:1
Y11 y10 y1 ;
Y10 y10 :
We determine the characteristic value of regime parameters for the short circuit
SC V1SC 0 and open circuit OC I1OC 0. Then,
V1SC 0;
V1OC
I1SC Y10 V0 ;
Y10
V0 ;
Y11
I1OC 0;
I0SC Y00 V0 ;
I0OC
DY
V0 :
Y11
4:2
4:3
97
98
4 Two-Port Circuits
I0
TP1
I1
V1
y10
V0
y0
y1
4:4
: I I SC I1SC V :
1
1
V OC 1
1
Equation (4.4) gives the following input and output load straight lines in Fig. 4.2.
Our circuit represents an active two-pole relatively to load. By denition, the
internal conductivity of the active two-pole is Yi I1SC =V1OC . Therefore, the second
equation of system (4.4) corresponds to Eq. (2.1) of Chap. 2. Then, we may use the
results of this chapter and determine the running regime parameter by an identical
value for the various actual regime parameters (voltages, currents) and for different
sections of circuit (input and output).
As the load characteristics are dened by linear expression (4.4), an afne
transformation or mapping V1 ! I1 ; V1 ! I0 ; I1 ! I0 takes place [3, 9].
Let an initial regime be given by values V11 ; I11 ; I01 . Then, afne ratio (2.6) or (2.8)
has the form
n11
V1OC V11
I11 I1OC
I01 I0OC
4:5
I1, I0
SC
I0
SC
I1
I0
I0
I0(V1)
I1
I1
OC
I0
I1(V1)
V1
V1
OC
V1
V1
99
4:6
Let a subsequent regime be set by values V12 ; I12 ; I02 . Then, the afne ratio is
n21
V1OC V12
I12 I1OC
I02 I0OC
SC
SC
OC
SC
OC
V1 V1
I1 I1
I0 I0OC
4:7
V11 V12
I12 I11
I02 I01
:
V1OC V1SC I1SC I1OC I0SC I0OC
4:8
So, we obtain the identical values for the voltage and currents at the output and
input.
Using (4.8), we get the recalculation formulas
SC
OC
I12 I11 n21
1 I1 I1 ;
SC
OC
I02 I01 n21
1 I0 I0 :
4:9
Let us consider the practical case. Let the voltage V1 be changed once again. We
get regime change (4.8) and currents (4.9). The structure of expression (4.8) shows
that errors of voltage measurement are reduced. Therefore, recalculation formula
(4.9) gives a more great accuracy, than primary Eq. (4.4), which contains the actual
voltage value V1. Also, the invariant value n1 can represent the interest for remote
voltage measurement that will be shown in this chapter further.
4.1.2
Let us now consider the cascade connection of two-ports TP1 and TP2 in Fig. 4.3.
A variable voltage source V2 is the load of this connection.
At change of the load voltage V2, we get the family of load straight lines in
Fig. 4.4.
Similarly, we have afne transformations or mapping of regime parameters for
various sections; that is,
V2 ! I2 ;
V 2 ! I1 ;
V2 ! V1 ;
V2 ! I0 ;
I2 ! I1 ! I0 :
100
4 Two-Port Circuits
TP1
V0, I0
V1, I1
TP 2
I2
V2
y10
y0
y1
I2, I1, I0
SC
I0
I0
SC
I0
I1
I0(V1)
SC
I2
OC
I0
I2
I1(V1)
OC
I2
I1
I2(V1)
V2
n12
V2OC V21
I21 I2OC
SC
OC
SC
V2 V2
I2 I2OC
V1OC V11
I11 I1OC
SC
OC
SC
V1 V1
I1 I1OC
I01 I0OC
:
I0SC I0OC
V2
OC
V2
V2
4:10
V21 V22
I22 I21
V11 V12
I12 I11
I02 I01
:
I0SC I0OC
4:11
101
So, we obtain the identical values for the voltages and currents of different
sections. From here, we get the voltage recalculation formula
OC
SC
V12 V11 n21
2 V1 V1
4:12
Example Let us consider a circuit with given values in Fig. 4.5a. The initial and
subsequent regimes correspond to the load voltages V21 1, V22 3. The calculated
currents, voltages of all the sections and their conformity are shown by the diagram
in Fig. 4.5b.
Afne ratio (4.10) for the initial regime
n12
V2OC V21
3:902 1
0:743;
OC
SC
3:902 0
V2 V2
n12
I21 I2OC
10:43 0
0:743;
n12
V1OC V11
4:878 3:86
0:743;
n12
I11 I1OC
15:3 8:78
0:743;
n12
I01 I0OC
34:56 30:49
0:743:
SC
OC
35:97 30:49
I0 I0
(a) V0, I0
(b)
SC
TP1
y10=4
TP 2
V1, I1
y0=1
y0=1
y1=2.4
y1=1
I2
y10=4
V2
OC
V2
V2
V2
V2
3.902
SC
OC
I2
I2
I2
I2
14.03
10.43
3.241
V2
I2
17.55
15.3
10.8
8.78
I1
3.509
3.86
4.562
4.878
V1
35.97
34.56
31.75
30.49
I0
n2
n2
0.7437
0.2317
n2
102
4 Two-Port Circuits
V21 V22
13
0:513:
4.2
4.2.1
I0
I1
Y00
Y10
Y10
V0
Y11
V1
4:13
At change of the load conductivity YL1, we get the load straight lines in Fig. 4.7.
In turn, bunches of straight lines with parameters YL1, YIN1 correspond to these
load lines. It is also possible to calibrate the load straight lines in the conductivity
values. The point of maximum load power corresponds to values V1i 0:5V1OC ,
i
Yi . The value Yi is the internal conductivity of circuit relatively to load YL1.
YL1
In previous Sect. 4.1.1 it was shown that the conformity of load straight lines is
an afne transformation for voltage and current. Now, we consider the conformity
YL1 ! YIN1 [5, 9]. For this purpose, we demonstrate the relationship YIN1 YL1 by
the transmission a parameters.
I0
TP1
V0
I1
V1
y10
y0
YIN1
y1
YL1
103
I 1, I 0
SC
I0(V0)
YIN1
SC
YIN1
I0
I1(V1)
YIN1
SC
YL1
I0
SC
I1
YL1
OC
I1
OC
YIN1
I0
YL1
V1
OC
V1
V0
V1, V0
V0
I0
" Y11
a12
V1
YD10Y
a22
I1
Y
a
11
a21
10
1
Y10
Y00
Y10
#
V1
:
I1
4:14
V1
I1
a22
a21
a12
V0
a11
I0
4:15
V0
I0
YIN1:C
#
" YpL1:C
V1
chc shc
DY
1
pI
shc chc
D
4:16
r
Y00
DY ;
Y11
YL1:C
r
Y11
DY :
Y00
4:17
104
4 Two-Port Circuits
In turn, the relation YIN1 YL1 or admittance transformation has the fractionally
linear view by (4.14)
YIN1
I0
a22 YL1 a21
:
V0 a12 YL1 a11
4:18
4:19
4:20
1
YL1
1 Yi 1
Yi
1 :
1
Yi 0
YL1 0
YL1
The conformity of all the parameters is given by the diagram in Fig. 4.8. Then,
we can constitute the cross ratio for these parameters.
For the input conductivity
SC
1
i
OC
m1L1 YIN1
YIN1
YIN1
YIN1
4:21
1
SC
i
SC
SC
1
YIN1
YIN1
YIN1
YIN1
YIN1
YIN1
OC i
OC 1
OC ;
1
YIN1 YIN1 YIN1 YIN1 YIN1 YIN1
i
SC OC
is the center of segment YIN1
YIN1 ; that is,
where the point YIN1
i
OC
YIN1
YIN1
SC
OC
YIN1
YIN1
:
2
V1 = 0
SC
SC
I1
YL1 =
SC
SC
V1
1
I1
V1
I1
I1
1
YL1
1
YIN1
YIN1
mL1
V1
YL1
i
YL1
2
OC
V1
V1
OC
I1
OC
YL1
OC
I1 =0
YL1 =0
YIN1
YIN1
YIN1
YIN1
mL1
mL1
105
Also, we have
m1L1 0 V11
V1OC OC
V1
V1 OC 1 1 ;
2
V1 V1
4:22
I1SC
I SC I 1
0 1 1 1 ;
2
I1
4:23
;
I01 I0OC I0i I0OC I01 I0OC
4:24
where the input current I0i is the center of segment I0SC I0OC ; that is,
I0i I0OC
I0SC I0OC
:
2
1
2
We consider the load change YL1
! YL1
. This change denes the segment
2
1
2
1
YL1 YL1 and segment YIN1 YIN1 on the line I0 V0 . It may be noted that length of
segments is different for usually used Euclidean geometry. However, if the mapping
is viewed as a projective transformation, an invariant or cross ratio is performed,
which denes the same length of segments.
Similar to (2.27)(2.29), this cross ratio has the forms
2
1
m21
L1 1 YL1 YL1 0
1
YL1
;
2
YL1
SC
2
1
OC
m21
L1 YIN1 YIN1 YIN1 YIN1
2
SC
SC
YIN1
YIN1
Y 1 YIN1
IN1
m2L1 m1L1 ;
OC
OC
2
1
YIN1 YIN1 YIN1 YIN1
2
1
OC
m21
L1 0 V1 V1 V1
SC 2 1
m21
L1 I1 I1 I1 0
4:25
4:26
V12 0
V11 0
;
V12 V1OC V11 V1OC
4:27
4:28
4:29
SC 2 1 OC
m21
L1 I0 I0 I0 I0
106
4 Two-Port Circuits
SC
OC 21 1
YIN1
YIN1
mL1 mL1
:
1
1 m21
L1 mL1
4:30
V1OC
L1
I11 1
I02
4.2.2
4:31
I1SC I11
:
SC 21
m21
L1 I1 mL1
4:32
1
I0SC I0OC m21
L1 mL1
:
1
1 m21
L1 mL1
4:33
There are different versions or variants of cross ratio depending on the choice of the
base points and a unit point. Let us consider expression (4.19). This expression sets
the conformity YL1 ! YIN1 , including negative values, as it is shown in Fig. 4.9 for
the values YL1:C YIN1:C 1.
Version 1
OC
SC
If the base points are YL1
0; YL1
1 and a unit point is YL1 YL1:C 1, the
cross ratio
1
~ 1L1 1 YL1
1 0
m
1
:
1
YL1
1
1
1
YIN1
thc
1 thc
1
1 Y 1 thc
1
YIN1
1 IN1 :
1 thc 1
thc
YIN1 thc
YIN1 thc 1 thc
-1
-th
1
YL1
-1
th
1
YIN1
-1/th
1/th
YL1
YIN1
107
1
2
1
2
We express the regime change for YL1
! YL1
and YIN1
! YIN1
as follows
1
Y 2 Y 1 thc
thc IN1 IN1
1
1 Y 2 thc 1 Y 1 thc YL1
2 IN1 1 IN1
2 :
YIN1 thc
YIN1 thc
YL1
~ 2L1 m
~ 1L1
m21
L1 m
It may be noted that the expression for the regime change and the expressions for
the conductivity changes at the input and output are expressed analytically in
different ways, but their numerical values are equal. In this matter, we can choose
the base points and a unit point in such a way as to these expressions are the same
view. This case is presented by the following version 2.
Version 2
We use the characteristic values 1; 1 as the base points. Then, the regime change
2
1
YL1
YL1
2
1
m21 1 YL1
YL1
1
Y 2 1 YL1
1 1 1YL12 YL11
1 L1
:
2 1
1 1
Y 2 Y 1
YL1
YL1
1 L1 2 L11
1YL1 YL1
2
1
YL1
YL1
:
2
1
1 YL1 YL1
4:34
21
1 YL1
:
21
1 YL1
4:35
Similarly, the same regime change and the conductivity change are resulted at
the input
2
1
m21 1 YIN1
YIN1
1
21
YIN1
2
1 YIN1
;
2
1 YIN1
2
1
YIN1
YIN1
:
2
1
1 YIN1 YIN1
4:36
4:37
Now, we constitute the cross ratio for the initial regime. It is possible to use the
OC
OC
0 as a unit point. This point corresponds to the point YIN1
thc in
point YL1
Fig. 4.9.
108
4 Two-Port Circuits
-1/th
1/th
1
m (YIN1)
-1
-th
th
YL1
YIN1
m (YL1)
Then
1
m1 YL1 1 YL1
0 1
1
1 YL1
;
1
1 YL1
1
m1 YIN1 1 YIN1
thc 1
1
1 YIN1
1 thc
m1 YL1 :
1
1
thc
1 YIN1
These cross ratio are shown by dash arrows on the superposed axis in Fig. 4.10.
In turn, we can consider the movement of point from the position YL1 to position
YIN1 . This movement determines the segment YL1 YIN1 . The movement of this
segment for different initial values YL1 is shown by arrows in Fig. 4.10.
Then, the points 1 are xed. It is possible to make the cross ratio for the points
YL1 ; YIN1 relatively the xed points. This cross ratio determines the length of
segment YL1 YIN1 and has the view
m YL1 ; YIN1 1 YL1 YIN1 1:
It is known that the property of such a cross ratio; its value does not depend on
running values YL1 ; YIN1 . For simplication, we set YL1 0; YIN1 thc. Therefore,
the cross ratio
m YL1 ; YIN1
1 thc
KPM :
1 thc
109
I1,I0
I0(V0)
I1(V1)
SC
YIN1
YIN1
I0
YL1
I1
Y
Y
OC
YIN1
0
0
V0
chc
p
shc
I0 DY
V1
shc
V1
p
:
chc
I1 DY
V0
V1,V0
4:38
4:39
This value is the length of radius-vector Y. On the other hand, the trajectory of
rotation is a hyperbola in the Euclidean plane.
Proposed interpretation corresponds to Lorenzs transformation for mechanics of
relative motion [4]. In turn, expression (4.19) corresponds to the rule of addition of
relativistic velocities.
4.2.3
Let us now consider the cascade connection of two-ports TP1 and TP2 in Fig. 4.12.
A variable conductivity YL2 is the load of this connection.
At change of the load conductivity YL2, we get the family of load straight lines in
Fig. 4.13. In this case, we have a projective transformation or mapping of load
straight lines and conductivities YL2 ! YL1 ! YIN1 for various sections. This
mapping is shown conditionally by dash-dot lines. According to (4.18), the resultant relationship YIN1 YL2 has the fractionally linear view.
In turn, the resultant matrix of a parameters is equal to product of the matrices
for the rst and second two-ports. Such a group property conrms projective
transformations.
110
4 Two-Port Circuits
V,I
0
TP1
TP 2
V ,I
2 2
V 1 , I1
y10
y1
y0
Y
L2
Y =Y
IN1
L1
I2, I1, I0
IN2
I1(V1)
I0(V0)
SC
I2(V2)
I0
SC
I1
SC
I0
SC
I2
V1
OC
I2
I0
I1
1
YL2
2
I2
OC
YL2
I1
OC
V1
0
OC
V2
V0
V2, V1, V0
Therefore, by (4.20) and (4.21), we constitute the cross ratio m1L2 for initial
values of regime parameters relatively to the load YL2 .
The conformity of all the parameters is given in Fig. 4.14.
Then, we get
SC
1
OC
m1L2 YL2
YL2
Yi YL2
1
SC
SC
YL2
YL2
Yi YL2
Yi
1 ;
OC
OC
1
YL2 YL2
Yi YL2
YL2
SC
1
i
OC
m1L2 YL1
YL1
YL1
YL1
1
SC
i
SC
YL1
YL1
YL1
YL1
OC i
OC ;
1
YL1 YL1
YL1 YL1
SC
1
i
OC
m1L2 YIN1
YIN1
YIN1
YIN1
1
SC
i
SC
SC
1
YIN1
YIN1
YIN1
YIN1
YIN1
YIN1
OC i
OC 1
OC :
1
YIN1 YIN1 YIN1 YIN1 YIN1 YIN1
4:40
111
1
SC
YL2
SC
V2
YL2 =
V2
SC
I2
SC
V1
SC
I1
SC
YL1
SC
YL2
V2
V2
I2
YL2
I2
I1
V1
I1
YL1
YIN1
YIN1
mL2
I2
OC
V1
OC
I1
V2
I2 =0
V1
YL1
OC
V1
I1
YL1
V2
YL2
OC
I2
V1
OC
YL2 =0
I1
OC
YL1
OC
YL1
YIN1
YIN1
YIN1
YIN1
mL2
mL2
SC
OC
YIN1
YIN1
:
2
Using (4.22)(4.24), we may constitute the same cross ratio for all the currents
and voltages. In particular, for the middle section with the parameters V1 ; I1
m1L2 V1SC V11 V1i V1OC
I11 I1SC
I11 I1OC
I1i I1SC
I1i I1OC
4:41
I1SC I11
I11 I1OC
where
V1i V1OC
V1SC V1OC
;
2
I1i I1OC
I1SC I1OC
:
2
1
2
We consider the load change YL2
! YL2
. This change denes the segments
1
2
1
2
1
YL2 , YL1 YL1 , and YIN1 YIN1 . The cross ratio denes the same length of these
segments
2
YL2
112
4 Two-Port Circuits
SC
2
1
OC
m21
L2 YL2 YL2 YL2 YL2
SC
2
1
OC
YL1
YL1
YL1
YL1
2
SC
1
SC
1
YL2
YL2
YL2
YL2
YL2
2 Y OC
1 Y OC
2
YL2
YL2
YL2
L2
L2
2
SC
1
SC
YL1
YL1
YL1
YL1
OC 1
OC
2
YL1 YL1
YL1 YL1
SC
2
1
OC
YIN1
YIN1
YIN1
YIN1
4:42
2
SC
SC
YIN1
YIN1
Y 1 YIN1
IN1
:
OC
OC
2
1
YIN1 YIN1 YIN1 YIN1
Using (4.27)(4.29), we may constitute the same cross ratio for all the currents
and voltages. In particular, for the middle section with the parameters V1 ; I1 , we get
SC
m21
V12 V11 V1OC
L2 V1
I1SC
I12
I11
I1OC
4:43
SC
1 OC 21
YL1
NL1
YL1 mL2
;
1 m21
1 NL1
L2
1
NL1
SC
1
YL1
YL1
OC :
1
YL1 YL1
4:44
1
V1SC V1OC m21
L2 mL2
;
21
1
1 mL2 mL2
I12
1
I1SC I1OC m21
L2 mL2
:
21
1
1 mL2 mL2
4:45
V0, I0
113
V1, I1
y10=4
TP 2
y10=4
y0=1
y0=1
y1=2.4
y1=1
YIN1
V2, I2
YL2
YL1=YIN 2
Fig. 4.15 Example of the cascaded circuit with the load conductivity
V0
I0
YIN1
a11 a12
V1
1:6 0:25
V1
;
4 1:25
a21 a22
I1
I1
I0
a22 YL1 a21
1:25 YL1 4
:
V1
I1
1:25
4
V0
0:25
I0
1:6
4:46
V1
I1
V2
I2
1:25
2:25
1:25
2:25
YIN2 YL1
V2
0:25
;
I2
1:25
V1
0:25
;
I1
1:25
4:47
I1
1:25 YL2 2:25
:
V0
I0
1:6
4
1:25
0:25
2:25
1:25
YIN1
V2
0:25
2:5625
I2
1:25
7:8125
V2
0:7125
;
I2
2:5625
2
YL2
1:08:
4:48
114
4 Two-Port Circuits
2
YIN1
3:175:
1
V0 34:56;
I01 YIN1
2
I02 YIN1
V0 31:75:
By (4.46), we obtain
V11 3:86;
V12 4:562;
I11 15:3;
I12 10:8;
1
YL1
15:3 3:86 3:963;
2
YL1
10:8 4:562 2:367:
Now, by (4.47)
V21 1;
I21 10:43;
1
YL1
10:43;
V22 3;
I12 3:241;
2
YL1
1:08:
0:344;
10:43 0
3:597 0
10:43
3:964 5
3:138 5
SC
1
i
OC
0:344;
YL1
YL1
YL1
YL1
SC
1
OC
m1L2 YL2
YL2
Yi YL2
m1L2
m1L2
SC
YL2
SC
OC
YL2
YL2
YL2
YL2
10.43
3.597
1.08
YL2
OC
V2
V2
V2
V2
V2
1.951
3.902
3.509
3.86
4.193
4.562
4.878
V1
17.55
15.3
13.16
10.8
8.78
I1
SC
OC
YL1
YL1
YL1
YL1
YL1
3.964
3.138
2.367
1.8
3.596
3.456
3.323
3.175
3.049
0.344
3.323
V2
YL1
YIN1
mL2
15:3 8:78
I11 I1OC
1
YL2
10:43
9:66;
2
1:08
YL1
2
SC
1
SC
YL1
YL1
YL1
YL1
2 Y OC
1 Y OC
YL1
YL1
L1
L1
2:367 5
3:964 5
4:64 0:478 9:66;
SC 2
1
OC
m21
L2 YL1 YL1 YL1 YL1
m21
L2
10:8 8:78
15:3 8:78
SC
2 1 OC
m21
L2 V1 V1 V1 V1
m21
L2
SC
1 OC 21
YL1
NL1
YL1 mL2 5 0:478 1:8 9:66
2:367;
1 m21
1 0:478 9:66
1 NL1
L2
1
NL1
1
Y1SC YL1
0:478:
1 Y OC
YL1
1
115
116
4 Two-Port Circuits
4.3
1
V1SC V1OC m21
3:509 4:878 3:323 19:72
L2 mL2
4:56;
1
1 3:323
4:323
1 m21
m
L2 L2
1
I1SC I1OC m21
17:55 8:78 3:323
L2 mL2
10:8:
21
1
1 3:323
1 mL2 mL2
Different sensors of physical values are used for monitoring of technical or natural
objects. For these, usually remote devices, it is necessary to transmit the measuring
signals, for example, over wire lines. Usually, digital signals are used for this
transmission. But such disadvantages as low noise stability, complicated and
expensive equipment, separate wire lines of power supply and communication take
place in digital systems. Therefore, researches and elaborations of easy-to-use
systems for transmitting signals are important.
Invariant properties of two-port networks allow transmitting the measuring
signals, using even the joint or combined wire line for communication and power
supply [5, 12]. As it was shown, the value of afne and cross ratio does not depend
on two-port network (wire line) parameters, accuracy of measuring devices, and
influence of noises. The parameters of this communication wire line, for example,
joint with a power supply line, are dened by both own parameters and current
consumed by other devices.
4.3.1
Let us consider the transfer of signals for the following simple examples. More
complex case of transmitting over a joint wire line for communication and power
supply is presented in [10].
Example 1 The value of cross ratio mL1 is accepted as a transmitted analog signal
VS; that is, VS mL1 . The structure of the appliance, designed to this transfer, is
shown in Fig. 4.17a.
In a short time t1 ; t2 ; . . .t5 , when the parameters of this circuit do not change, the
four samples of load conductivity are transmitting by connecting the respective
SC
1
i
OC
; YL1
; YL1
; YL1
. The connection is realized by a multichannel
conductivities YL1
switching and multi-output generator Gen with a pulse period TK. The conductivity
1
1
is the signal or information sample. Therefore, its value YL1
VS is calculated by
YL1
a unit F for the running signal value VS.
(a)
117
Gen
V0 , I0
OC
F
SC
SC
YL1
VS
-1
VS
OC
I0 , I0, I0, I0
(b)
I0
SC
I0
I0
I0
OC
I0
0
t1 t2 t3 t4 t5
t1+TK
1
SC
SC
YL1
YL1
Y i YL1
L1
VS :
OC
OC
1
i
YL1 YL1
YL1 YL1
SC
i
OC
YL1
VS NL1
YL1
FVS ;
i
1 VS NL1
i
NL1
SC
i
YL1
YL1
:
OC
i
YL1 YL1
In particular case
1
i
m1L1 1 YL1
YL1
0
i
YL1
VS ;
1
YL1
1
YL1
i
YL1
:
VS
118
4 Two-Port Circuits
Gen
V0, I0
OC
I1
SC
V1
F
SC
I0 , I0, I0
VS
-1
OC
V1
V1
VS
At the input of the circuit, the cross ratio or the signal is calculated by a unit F 1
similar to (4.24). For this purpose, the measured input currents I0SC ; I01 ; I0i ; I0OC ,
shown in Fig. 4.17b, are used; that is,
VS m1L1 I0SC I01 I0i I0OC
The structure of this expression shows that measuring errors of currents mutually
are reduced. The process of connection and calculation is repeated every period TK.
If during this time the parameters of two-port network changed, the values of input
currents changed too (for example, the currents decreased in Fig. 4.17b). But the
cross ratio value remains the same.
Example 2 The value of afne ratio n1 is accepted as a transmitted analog signal VS;
that is, VS n1 . The structure of the appliance is shown in Fig. 4.18.
In a short time, the three samples of load voltages are transmitted by connecting the
respective voltage sources V1OC ; V11 ; V1SC . The voltage V11 is the information sample.
Therefore, its value V11 VS is calculated by a unit F for the running signal value VS.
We use expression of afne ratio (2.6) or (4.5)
n11
V1OC V11
VS :
V1OC V1SC
From this,
V11 V1OC VS V1OC V1SC FVS :
119
I01 I0OC
F 1 I01 :
I0SC I0OC
The structure of this expression shows that measuring errors of currents mutually
are reduced. The process of connection and calculation is repeated by period TK.
Also, we may use afne ratio (2.7)
VS
4.3.2
V1OC VS V1SC
:
VS 1
Let us consider the circuit in Fig. 4.19. The action of this appliance is similar to the
system in Fig. 4.17.
1
of a resistive sensor by input
We need to calculate the conductivity value YL1
SC
i
OC
currents. The given conductivities YL1 ; YL1 ; YL1 are reference conductivities.
Using the known cross ratio
SC
1
i
OC
YL1
YL1
YL1
m1L1 YL1
1
SC
i
SC
YL1
YL1
YL1
YL1
;
1 Y OC
i Y OC
YL1
YL1
L1
L1
Gen
V0 , I0
OC
SC
YL1(I0)
OC
I0 , I0, I0, I0
SC
YL1
YL1
120
4 Two-Port Circuits
we nd the value
1
YL1
SC
i
OC 1
YL1
NL1
YL1
mL1
1
YL1
m1L1 ;
i m1
1 NL1
L1
i
NL1
SC
i
YL1
YL1
:
i Y OC
YL1
L1
On the other hand, the cross ratio, by the input currents I0SC ; I01 ; I0i ; I0OC , has the view
m1L1 I0SC I01 I0i I0OC
1
Using this expression, we obtain the value YL1
I0 .
4.4
We use the results of Sects. 1.5.1 and 2.3. Let us consider a more complex case of a
quadratic curve, as the efciency of a two-port in accordance with expression (1.34)
KP KG
1 KG
:
A KG
4:49
This expression represents a hyperbola in Fig. 4.20 for all area of load changes.
The positive load consumes energy; the maximum power transfer ratio (1.33)
p p
KP A A 12 :
4:50
p
AA 1:
4:51
In turn, the negative load returns energy and we get the corresponding maximum
values
p p
KP A A 12 ;
KG A
p
AA 1:
4:52
Next, we must determine all the characteristic points [6, 11]. Obviously, there
are points B1 ; 0; KP ; 1; A1 ; KP by Fig. 4.20b. These points correspond to points
T 1; 0; KG ; 1; A; KG of the axis KG .
121
It is possible to take up the points 0; 1 as the base points and the point KG as a
unit point. But visibly, the other characteristic points have to be dened relatively to
these basic points and not depend on the parameter A of comparable two-ports.
Therefore, we must, at rst, dene the possible systems of all the characteristic
points. To do this, we will study a regime symmetry of efciency using the regime
symmetry for the load power of Sect. 2.3.
4.4.1
In Fig. 4.20b, the pole S and polar TA1 determine the mapping or symmetry of the
region of power delivery by the voltage source V0 (above of the polar) on the region
of power consumption of this voltage source (below of the polar). The point KP
passes into the point KP . Points B1 ; A1 are the xed base points and KG B1 1,
KG A1 A. A hyperbola point is assigned by the hyperbolic rotation of
radius-vector RS S.
KP
(a)
KP
KP
F
1 A
KG
Q K+ K
GF
G
KG
(b)
KG
KP
KP
B1 M
KP
RS
KG
A1
KG
122
4 Two-Port Circuits
Y00
mIN(0)
KG
KG 1
KG
YIN.C
YIN.C YIN
1 1/mIN(0) 0
mIN
DIN
In turn, the pole T and polar SM determine the mapping or symmetry of the
hyperbola relatively to the straight line KP KP or to the points of maximum
efciency.
For the pole T, the following correspondences take place.
We believe harmonic conjugate points T; 0; Q; 1 onto the line TQ or the axis KG.
Therefore, these points correspond to the points KG T 1; 0, KG Q 0:5; 1.
The mutual mapping of the points 0; 1 relatively to Q is shown by arrow in
Fig. 4.21.
Similarly, for the pole S, the following correspondences take place.
We believe harmonic conjugate points S; KP ; M; KP , which correspond to the
points A; KG , 1; KG of the axis KG . The mutual mapping of the points KG ; KG
relatively to A; 1 is shown by arrow in Fig. 4.21.
Let us use the points A; 1 as the base points and KG as a unit point. Therefore,
a running regime point is expressed by cross ratio
A KG
mIN 1 KG KG A p
AA 1
YIN
1 YIN:C YIN 0
:
YIN:C
4:53
This value is the deviation of KG from the point KG . The correspondence of the
values KG , YIN ; mIN is shown in Fig. 4.21.
Now, it is necessary to check the cross ratios for the points 0, 1
r
A
Y00
mIN 0
;
A 1 YIN:C
r
A1
1
:
mIN 1
A
mIN 0
4:54
The modules of these values are equal to each other if we use the hyperbolic metric
123
HIN
LnmIN
:
HIN 0 LnmIN 0
4:55
A KG
A
A1
DIN =2
YIN
YIN:C
Y00
YIN:C
DIN
4:56
The rst member of these equations is the relative deviation but this deviation is
determined by a value DIN and by parameters of a circuit (the second member).
If it is necessary to set any equal deviation by KG for different circuits, we get
p
KG A AA 1
4.4.2
A
A1
DIN =2
4:57
In Fig. 4.22, the pole F and polar TQ, as the axis KG, determine the mapping or
symmetry of the region of power consumption by the load (above of the polar) on
Fig. 4.22 Symmetry of the
efciency relatively to the
axis KG
KP
KP
T
KP
A
KG
Q KG KGF
A1
124
4 Two-Port Circuits
KG KGF 1
YL.C
Y11
KG KG
YL.C YL
DL
the region of power return of this load (below of the polar). The point KP passes
into the point KP . The Points 0, 1 are the xed base points.
In turn, the pole T and polar FQ are a complementary system and determine the
mapping or symmetry of the hyperbola relatively to the straight line KP KP or to
the points of maximum efciency.
For the pole T, the following correspondences take place.
We believe harmonic conjugate points T; 0; Q; 1 onto the line TQ. The mutual
mapping of the points T 1; Q relatively to the points 0, 1 is shown by dash
arrow in Fig. 4.23.
Similarly, for the pole F, the following correspondences take place.
We believe harmonic conjugate points F; KP ; Q; KP , which correspond to the
points KGF ; KG , 0:5; KG of the axis KG. The mutual mapping of the points KG ; KG
relatively to 0:5; KGF A=2A 1 is shown by arrow in Fig. 4.23.
Let us use the point KG as a unit point. Therefore, a running regime point is
expressed by cross ratio
r
1 KG
A
mL 0 KG KG 1
A1
KG
4:58
YL
1 YL:C YL 0
:
YL:C
The correspondence of the values KG, YL ; mL is shown in Fig. 4.23.
Now, it is necessary to check the cross ratio for the rest characteristic points, the
points 1; A
r
A
Y11
;
A1
YL:C
mL 1
mL A
r
A1
1
:
A
mL 1
4:59
125
The negative values show that these points disposed into external area of the
base points. We map these points into internal area and obtain the harmonic
conjugate points 0:5; KG F shown by dash arrows.
Similarly to (4.54) and (4.55), we may at once write the following expressions
for deviation DL relatively to the load
1 KG
KG
r
DL =2
A
A
;
A1
A1
(a)
YL
YL:C
Y11
YL:C
DL
4:60
0.8
DL
0.25
0.5
0.6
1
0
KP 0.4
0.2
A=1.05
0.5 0.25
DL
A=1.25
0.0
0.0
0.2
0.4
0.6
0.8
1.0
KG
(b) 0.8
DL
0.25
0.5
0.6
0.25
1
A=1.05
0.5
KP 0.4
DIN
0.2
0.0
0.0
0.2
0.4
0.6
KG
0.8
1.0
126
4 Two-Port Circuits
1
q
A1 A DL =2
A A1
4:61
As an example, the calibration of the hyperboles for two parameters A by deviation DL of (4.61) is given in Fig. 4.24a. The comparison of calibration by DIN of
(4.57) and DL is shown in Fig. 4.24b.
4.5
4.5.1
At rst, we develop the results of Sect. 4.4. Using Eq. (4.13), we pass on to the
input and output power of a two-port
P0 I0 V0 Y00 V0 2 Y10 V0 V1
P1 I1 V1 Y10 V0 V1 Y11 V1 2 :
4:62
4:63
Y10
V0 :
Y11
Y10 2
V0 2 :
Y11
4:64
P0M
:
PGM
4:65
127
4:66
As it was told, expression (4.49) represents the hyperbola in Fig. 4.20 with the
center S. This center coordinate corresponds to the value
KPS 2A 1:
4:67
p
AA 1;
KP KPS 2
p
AA 1:
4:68
Therefore,
KP KP 2KPS :
4.5.2
Let us consider two regulable converters or voltage regulators VR1; VR2 with a
common load R1 in Fig. 4.25a. Two-ports TP1; TP2 with the corresponding
effectiveness parameters A1 ; A2 are losses of these regulators are shown in
Fig. 4.25b.
Taking into account the designations of the input and output powers, we form
the following system of the equations
P0 P01 P02
P1 P11 P12 :
We may believe that the corresponding maximum input powers are equal to
P01M P02M 1:
Then, using (4.66), the efciency of this modular connection takes the view
KP
:
P0
P01 P02
4:69
This equation KP P01 ; P02 represents a two-sheeted hyperboloid with the center
S. Therefore, for this surface, we get characteristic regime curves besides characteristic regime points [7]. The center coordinate corresponds to the value
128
Fig. 4.25 a Parallel
connection of two regulable
converters with losses, b its
equivalent circuit
4 Two-Port Circuits
(a)
(b)
KPS
4A1 A2
1:
A1 A2
4:70
2A1 A2
:
A1 A2
4:71
This value determines the total effectiveness parameter of the given modular
connection. In turn, equalities (4.68) take place, where the values KP ; KP are the
total maximum efciency.
We must note that expression (4.71), as the parallel connection of two resistances R1 ; R2 , corresponds to the harmonic conjugacy of four points or three
129
segments. Let us consider this property in details with the aid of Fig. 4.26. The
points A1 ; A2 are harmonic conjugate relatively to the base points 0; A. Also, the
points R1 ; R2 are harmonic conjugate relatively to the base points 0; 2R.
Therefore, we write the following cross ratio
0 A1 A2 A
A1 0 A2 0
1:
A1 A A2 A
:
A1 A A2 A
From here, expression (4.71) follows.
Analogously, we have
0 R1 R2 2R
R1 0
R2 0
1:
R1 2R R2 2R
Further,
R1
R2
:
R1 2R R2 2R
From here, the known formula follows
R
R1 R2
:
R1 R2
130
4 Two-Port Circuits
P01
P11
A1
A12
P0
V0
P02
4.5.3
P12
A2
P1 b1 A1 b1 2 b2 A2 b2 2 2Ab1 b2
;
P0
2 b1 b2
A1 A2 A12 2
1:
A1 A2 2A12
A1 A2 A12 2
:
A1 A2 2A12
4:72
Value (4.72) determines the total effectiveness parameter of the given module
connection.
We may also introduce the values
2
1 A1 A2 A12 ;
A
A2 A12
2 A1 A2 A12 :
A
A1 A12
131
These values determine the effectiveness parameters for every output terminal.
Therefore, expression (4.71) carries out too
2
1A
2A
A
2 :
A1 A
4.6
4.6.1
Usually, the efciency is used for evaluation of the power supply system effectiveness. But, there are some problems shown in [11].
For example, the efciency value still tells nothing about the effectiveness of an
appliance. Therefore, this efciency value is compared or confronted with the load
power or some reference regime. But, such proposed indices are insufcient to
compare the effectiveness of diverse systems or at change in system parameters.
After all, each power supply is characterized by the power opportunities too; for
example, maximum powers, currents and so on. Therefore, normalized regime
parameters are used. But, for example, the normalized load power does not give
information on the regime effectiveness.
4.6.2
4:73
2
POC
0 th c
A1
:
A
132
Fig. 4.28 Load power via the
input power for various
parameters A. a Cartesian
coordinates, b projective
coordinates
4 Two-Port Circuits
P1
(a)
1
A<0
A=0
A3<1
A=1
B1
P0
B2
2
th
A1>1
A2>A1
C2
C1
P1
(b)
1
A4<A3
A=0
A3<1
A=1
A1>1
P0
0
B1
C1
A4
S
One more case corresponds to A 1; the parabola degenerates to the straight line
P0 1. In turn, the load returns power for the region P1 \0.
Change of losses
Let us consider the projective coordinates in Fig. 4.28b. The bunch of our parabolas
represents the bunch of ellipses passing through the other common point P1 1.
133
The tangential innitely remote straight line corresponds to this point P1 1 and
the point P0 1.
So, we have the two tangential straight lines, which intersect into a point S. The
point S is called a pole; the straight line, corresponding to A 1, is a polar.
These pole and polar determine the symmetry or mapping of points of ellipses of
the region P0 [ 1 onto the region P0 \1. In this case, we have the symmetry of the
points C1, B1 relatively to the xed or base points Q, S. Analogously, the symmetry
of the points C2, B2 takes place.
Thus, these symmetrical points and base points are the harmonic conjugate
points. Similar to (4.72), we get
C1 Q B1 Q C2 Q B2 Q
1:
C1 S B1 S
C2 S B2 S
From here
C2 Q C1 Q B2 Q B1 Q
:
C2 S C1 S
B2 S B1 S
The obtained relationships are the cross ratios of the comparable pairs C1 ; C2 and
B1 ; B2 relatively to the common base points Q, S; that is,
Q C2 C1 S
C2 Q C1 Q
m21
A P1 Q B2 B1 S:
C2 S C1 S
4:74
If we have executed all the calculation, the nal expression follows [11]
m21
A P1
r
A1 chc1
:
A2 chc2
4:75
134
4 Two-Port Circuits
(a)
P1
1
A=0
A=
P0
1
P0
A=1
A1
P1
(b)
A=0
P1(0)
P1(1)
1
P1
A=1
1
0
P0
P0
A1
A=
P1()
4:76
According to (4.73),
P11 1 P0 A1 1 P0 2 ; P1 1 P0 1 P0 ; P1 0 1 P0 ; P1 1
1:
135
Therefore,
m1P1
P11 1 P0
A1 :
P0 1 P0 1 P0
Analogously, for the value A2, we get the load power P21 and
m2P1
P21 1 P0
A2 :
P0 1 P0 1 P0
m2P1
m1P1
P21
1P0
P11
1P0
1
1
A2
:
A1
4:77
4:78
4.6.3
P1 1 P0
1 P0 2
A
:
P0
P0
P0
4:79
2
POC
0 th c
A1
:
A
136
4 Two-Port Circuits
(a)
KP
(b)
KP
S 1
A2
A=1
A=0
C2
A=0
C1
B1
A=1
P0
A2
0
B1
1
B2
P0
A1>1 A2>A1
A2
C1
A1>1
Fig. 4.30 Efciency via the input power for various parameters A. a Cartesian coordinates,
b projective coordinates
straight lines, which determine the point S of intersection; the corresponding efciency KP 1. If A = 0, we obtain the hyperbola too.
Change of losses
Let us consider the projective coordinates in Fig. 4.30b. The bunch of our hyperbolas represents the bunch of ellipses passing through the other common point
KP 1. The tangential straight lines correspond to this point KP 1 and the
point PSC
0 1. Therefore, for A 1, the hyperbola degenerates into the innitely
remote straight line 1.
The above point S is the pole too; the straight line, corresponding to A 1, is
the polar. In this case, we have the symmetry of the points C1, B1 relatively to the
innitely remote straight line 1 or the base points Q; S. Analogously, the symmetry
of the points C2 ; B2 takes place.
(a)
137
KP
S
A=0
A=1
(b)
1
0
P0
=
P0
KP
KP(0)
A=1
A=0
KP(1)
1
KP
P0
P0
A=
A1
A1
KP( )
Fig. 4.31 Efciency changes for a given value of the input power. a Projective coordinates,
b Cartesian coordinates
shc1
:
shc2
4:80
The obtained expression differs from (4.75). From here, it follows that the form
of losses change is determined by the initial dependences (4.73) and (4.79).
Change of efciency
Let us consider a given value of the power P0 in Fig. 4.31a. This vertical straight
line intersects all the hyperbolas with the characteristic effectiveness parameters
A = 0, A = 1, A 1 and a running value A1 . So, we get the corresponding
efciency KP 0, Kp1 , KP 1, KP 1 in Fig. 4.31b.
138
4 Two-Port Circuits
Next, we may form the following cross ratio for the value A1,
m1KP KP 0 KP1 KP 1 KP 1:
According to (4.79)
KP 1 1 P0 ;
KP 0
1 P0
;
P0
KP 1 1:
KP2
1P0
KP1
1P0
1
1
KP1
1 A2 sh2 c2
:
1 A1 sh2 c1
References
1. Alexander, C.K., Sadiku, M.N.O.: Fundamentals of Electric Circuits, 5th edn. McGraw-Hill,
New York (2009)
2. Irwin, J.D., Nelms, R.M.: Basic Engineering Circuit Analysis, 10th edn. Wiley, Hoboken
(2011)
3. Kagan, V.F.: Osnovania geometrii, Chasti II (Geometry Basics. Part II). Gostekhizdat, Moskva
(1956)
4. Kittel, C.K., Knight, W.D., Ruderman, M.A.: Mechanics. Berkeley Physics Course, vol. 1.
McGraw-Hill, New York (1962)
5. Penin, A.: Projectivelyafne properties of resistive two-ports with a variable load.
Tehnicheskaia elektrodinamika 2, 3741 (1991)
6. Penin, A.: Denition of deviation from the matching regime for two-port circuit.
Electrichestvo 4, 3240 (1994)
7. Penin, A.: The relative regime of four-port connected in parallel: geometrical approach.
Electrichestvo 2, 4957 (1997)
8. Penin, A.: Fractionally linear relations in the problems of analysis of resistive circuits with
variable parameters. Electrichestvo 11, 3244 (1999)
9. Penin, A.: The invariant properties of two-port circuits. World Acad. Sci. Eng. Technol. 3(4),
893899 (2009). http://www.waset.org/publications/5517. Accessed 30 Nov 2014
10. Penin, A., Sidorenko, A.: Method for signal transmission through the direct current line. MD
Patent 536, 12 Dec 2011
References
139
11. Penin, A.: Denition of normalized energy efciency indices of resistive two-port networks.
Probl. Reg. Energ. 2(22), 2037 (2013). http://journal.ie.asm.md/assets/les/m71_2_246.pdf.
Accessed 30 Nov 2015
12. Penin, A., Sidorenko, A.: Transmission of measuring signals and power supply of remote
sensors. In: Bonca, J., Kruchinin, S. (eds.) Nanotechnology in the Security Systems. NATO
Science for Peace and Security Series C: Environmental Security, pp. 267281. Springer,
Dordrecht (2014)
Chapter 5
5.1
Introduction
The paralleling of lower power voltage sources (converter modules) offers the
well-known advantages over a single, high-power source. The base problem of such
a power supply system is the load-current sharing among the paralleled modules.
Various approaches of current distribution are known [3]. In the simplest droop
method equalizing resistors are used [4, 5, 11], including lossless passive elements
[10]. Usually, the equality of module parameters is provided; that is, open circuit
voltages and internal resistances. Therefore, the distribution of currents means the
equality of these currents.
On the other hand, scatter of module parameters, possible cases of use of primary voltage sources with different capacity determines the non-uniformity distribution of currents.
Therefore, it is natural to understand the uniform loading of sources in the
relative sense when the actual loading corresponds to the capacity of the source.
The analysis of this power supply system by the method of projective geometry has
led to introduction of some concepts for the quantitative representation of operating
regimes [68].
5.2
Initial Relationships
Let us consider voltage sources V1 ; V2 presented in Fig. 5.1. Resistances Ri1 ; Ri2 are
internal resistances of these voltage sources; equalizing resistors Re1 ; Re2 provide
the distribution of currents for a given load resistance R0 .
141
142
RL1
R i1
R e2
R i2
1.0
12.5
V2
0.2
RL2
0.857
0.5
1.25
V1
I1
R e1
I2
I0
0.833
R0
12.0
6.507
5:1
The possible normalized parameters of a load regime (or source loading) of the
rst voltage source look like
m1
RL1
;
Ri1
J1
I1
:
IM1
5:2
Here, the maximum current of the voltage source corresponds to its short circuit
current
IM1
V1
:
Ri1
5:3
IM1 I1
;
IM1 I1
IM1
;
I1
and so on.
These expressions are used in different areas of electrical engineering, radio
engineering, and power. Let us note that all these equations represent fractionally
linear expressions and can be interpreted as projective transformations which
possess an invariant. Therefore, all the above-mentioned expressions are equivalent.
Further we will use expressions (5.2).
Since
RL1
V1 Ri1 I1 V1
Ri1 ;
I1
I1
143
then
m1
RL1
V1
IM1
1
1
1 1:
J1
Ri1 Ri1 I1
I1
5:4
IM1
:
m1 1
5:5
RL2 IM2
1
1 1;
J2
Ri2
I2
I2
5.3
5.3.1
J2
I2
V2
; IM2
;
IM2
Ri2
IM2
:
m2 1
5:6
5:7
IM1
V1
;
m1 1 m1 1Ri1
I2
V2
;
m2 1Ri2
then
V1 V2 Ri1 Re1
V1
V2
Ri2 Re2
:
m1 1Ri1
m2 1Ri2
am1 d a 1
;
m1 d
5:8
144
where
V1
V2 Re2
V2
V1 Re1
; d
;
V1 V2 V1 V2 Ri2
V1 V2 V1 V2 Ri1
V1 Re1
V2 Re2
a d1
:
V1 V2 Ri1 V1 V2 Ri2
5:9
IM1
1
;
m1 1IM1 m1 1
J2
IM2
1
:
m2 1IM2 m2 1
d1
1
J1
:
a1
a1
5:10
The plots of dependences (5.8) and (5.10) are presented in Fig. 5.2.
Expression (5.8) corresponds to a hyperbole, and (5.10) corresponds to a straight
line. The desirable operating regime corresponds to straight lines on these plots; that
is, m2 m1 ; J2 J1 . The crossing of this straight line with the hyperbole plot
gives the two points m1 ; m2 of the equal loading of sources. The working area,
when load consumes energy, corresponds to the rst point m1 . The second point
corresponds to the condition when the voltage sources relatively equally consume
energy. Let us determine the points m1 ; m2 . In this case, expression (5.8) leads to
the quadratic equation
m2 d am d a 1 0:
Its solution gives the two roots
m1 d a 1;
m2 1:
5:11
J 2 1:
5:12
1
;
d a2
For the second point J 2 , the currents I2 ; I1 ! 1. Though this case physically is
not feasible, but its mathematical description allows introducing some necessary
characteristics of a circuit.
145
m2
(a)
m
(1)
(1)
-1
(2)
-1
-a
(b)
(1)
m1
J2
J
(1)
(1)
0
1
d+1
-1
a-1
5.3.2
(1)
J1
The points of equal loading are the xed points of the projective transformation
m1 ! m2 ; J1 ! J2 [1, 2], as it is shown in Fig. 5.3.
Let us consider in detail this geometrical interpretation of transformation (5.8) for
different initial values of the quantities m1 ; m2 at loading change. These quantities
m1 , m2
J1 , J2
-1
J2
J1
m2 m1
(1)
J1
m
(1)
m1
J2 0
m2
146
dene a line segment, its length (in the usual sense of Euclidean geometry) or
degree of difference of source loading is decreased at its approach to the xed
points. It is obvious that this length for the different circuits will be various.
Thus, it is possible to enter two concepts; the rst of them denes a circuit: how
much the loadings of sources can differ. The second concept denes deviations of
actual loadings from the xed point in the relative form.
In this case, it is possible to compare running regimes of the different circuits.
For introduction of such characteristics we use a number of concepts of projective geometry, applied in Chap. 2. We constitute the following cross-ratio:
m2 m2 m1 m1
m2 m2 m1 m2
;
m2 m1 m1 m1
5:13
where the points m2 ; m1 are the base ones. Also, it is known that the cross-ratio,
concerning the xed points, does not depend on running points m1 ; m2 . Therefore,
we accept m1 1 for simplication of calculations.So, by (5.8)
m2 1
a 1 d a 1
a;
1d
then
m2 1 m2
m2 1 m1
a 1
a1
KL :
a d a 1 d 1
m2 m2 1 1 m1
5:14
V2 1 Re2 =Ri2
:
V1 1 Re1 =Ri1
5:15
m2 1
m1 1
:
m2 m1 m1 m1
5:16
147
m2
KL m
1 1K
m1 m1
L
1
m
m1 KL1K
L
5:17
1 KL m1 m1 1 KL m1
:
1 KL m1 KL m1
Dependence (5.17) for different values of KL is presented in Fig. 5.4. The bunch of
hyperboles is obtained for KL 6 1. In turn, if KL 1, these hyperboles degenerate
into the straight line m2 m1 .
Let us analyze expression (5.15). We consider V1 V2 .
Then
KL
1 Re2 =Ri2
:
1 Re1 =Ri1
:
Ri2 Ri1
Therefore, it is quite possible to put Re2 Re1 0. Thus, if voltage sources have
identical open circuit voltages, they are equally loaded, and it is independent on
their capacity.
Generally, expression (5.14) allows comparing the factors of non-uniformity of
loading of different circuits; to determine the values of equalizing resistors for
necessary value of this factor.
m2
(1)
(1)
KL>1
KL=1
KL<1
-1
(2)
-1
-a
(1)
m1
148
5.3.3
KL m1
:
1 KL
1 m2 m1 1
m1 1
m1
m1 KL1KL
m1 1
m
m1 KL1K
L
m1 1
1
m
m1 KL1K
L
KL
;
1 KL
m2 1
m1 1
m2 1
1
:
m2 1 m 1 m1 1
m1 1 m1 KL m1 1 KL
1KL
The rst member of this expression represents the normalized value and prompts
how to write the similar value in the second member.
Therefore,
KL m 1 1
1K
m2 1
1 1
L m
:
1
m1 1 mm11 11 1K
5:18
:
1
1
a 1J
a1
KL
J
J KL
5:19
149
It should be noted that expressions (5.18) and (5.19) set certainly deviation of
running parameters of loading from the equal loading regime by the normalized
values. But, it is not enough for comparison of deviations for circuits with different
parameters KL . As an example of the most simple relation (5.19), we will show,
why it turns out.
~ L , but with the
Let us consider two circuits with the different parameters KL ; K
1
identical value J 1. The characteristics of circuits are presented in Fig. 5.5.
Loading regimes may be considered identical if conformity of the characteristic
regime points takes place (are shown by arrows in Fig. 5.5) at change of the load. It
follows from the similarity principle [9]. Then, a projective transformation takes
place and it is set by the center at the point 0 and by three pairs of the characteristic
~ B;
~ D.
~ The points D; D
~ coincide among themselves and
regime points A, B, D, and A;
correspond to the xed point J 1 . The point of running regime C should correspond
~ For such a projective transformation, the cross-ratio is carried out
to the point C.
~C
~D
~ B:
~
A C D B A
~ B
~ are the base points, and points D; D
~ are unit ones.
Here, the points A, B and A;
Therefore, this cross-ratio can be accepted as the equal deviation of the running
~ from unit points.
points C; C
Further, we map the points A, C, D, B onto the axis of current J1 . Then, we
obtain the deviation for the rst source
(1)
J2/J
1
C
0
B
B
1-K
-K
1-K
-K
(1-K)
(1)
J1/J
150
D1 0 J1 J 1 1 KL J 1
J1 0
J 1 0
J1
KL :
1
1
1
J1 1 KL J 1
J1 1 KL J
J 1 KL J
5:20
J2
KL :
J2 1 KL J 1
Thus, the deviations include parameters of a circuit and are not simply the normalized values J1 =J 1 ; J2 =J 1 .
We can represent the deviations in the other form
D1
0 J1 J 1
D2
1
d1
J1
a1
a 1J1
:
J1 d 1 1 d 1 d 1J1 1
0 J2 J 1
1
d1
a 1J2
:
d 1J2 1
5:21
5:22
Taking into account conformity (5.4) and (5.6) between various denitions of
parameters of loading regime, the deviations are expressed in the invariant form
through the corresponding cross-ratio for the variables m1 ; m2
D1 0 J1 J 1
D2 0 J2 J 1
1
d 1
1
d1
1 m1 m1 d
1a
:
m2 d
m1 d
1a
;
m1 d
m1 d
5:23
The values of deviations for the characteristic points and running points are presented in Fig. 5.6.
In particular, the deviation D2 for the second xed point, m2 1, is equal to
the parameter KL ; that is,
D2
a1
KL :
d1
5:24
(2)
(1)
151
J1
1
d+1
J2
0
(2)
m1
(2)
(1)
LnK
m,m
1
2
r2
r1
(1)
(2)
Ln1 Ln2
r1
J1, J2
m2
(da+1)
(1)
r2
1, 2
r1, r2
r1, r2
It turns out that such a deviation depends on a circuit parameter. We can exclude
this dependence as follows. We introduce the hyperbolic metrics or distances
r1 Ln D1 ;
r2 Ln D2 :
5:25
r 2 Ln D2 Ln KL :
5:26
Only one nonzero and nite value of the distance r 2 is obtained. Therefore, we can
use this value as the scale and introduce the normalized values
r1
r1
Ln D1
;
2
Ln
KL
r
r2
r2
Ln D2
:
2
Ln
KL
r
5:27
Example 1 We use the specic elements in Fig. 5.1. Maximum currents of sources
(5.3)
IM1 10;
IM2 12:
m2 13:4; J2 0:0694:
152
29:8m1 5:2
;
m1 34
J2 1:215J1 0:0347:
J 1 0:1613:
D2 1:3986:
r2 0:3354:
0:2105
1:0795;
0:195
r2
0:3354
1:72:
0:195
(2)
(1)
J1
0.6113
(2)
(1)
5.2
0.0857
J2
0.0694
m1
m2
10.66
13.4
1/35
34
m,m
1
2
(2)
0.8228
1.2343
1.3986
r1
r2
0.2105
0.3354
(1)
(2)
-0.195
(1)
r1
1
-1.0795
J1, J2
1, 2
r1, r2
r2
-1.72
r1, r2
5.4
5.4.1
153
Let us write expressions which associate parameters of source loadings in the form
m2 m1 and J2 J1 with the variable equalizing resistor Re1 for a given load. From
(5.1), (5.5) and (5.7) it follows:
V2
V1
V2
R0
Ri2 Re2 R0 :
m1 1Ri1
m2 1Ri2
Then, we obtain
m2 m1
a m1 d a 1
;
m1 d
5:28
where
Re2 R0
V1 R0
; d
1;
Ri2
V2 Ri1
Re2 R0 V1 R0
ad1
:
Ri2
V2 Ri1
a
5:29
d1
1
J1
;
a1
a1
5:30
The plots of dependences (5.28) and (5.30) are presented in Fig. 5.8. Expression
(5.28) corresponds to a hyperbole and (5.30) corresponds to a straight line. The
desirable operating regime corresponds to straight lines on these plots; that is,
m2 m1 ; J2 J1 . The crossing of this straight line with the hyperbole gives the
two points m1 ; m2 of the equal loading of sources. The working area when load
consumes energy corresponds to the rst point m1 . The second point corresponds
to the condition when the voltage sources relatively equally consume energy.
Let us nd the xed points m1 ; m2 . In this case, expression (5.28) leads to the
quadratic equation
m2 d am d a 1 0:
154
(a)
m2
(1)
(1)
m
1
m2
-1 0
-1
(2)
(b)
(1)
m1
m1
J2
1
1+
J
(1)
(1)
0
J
(1)
1
1+
J1
m2 1:
5:31
J 2 1:
5:32
1
;
da2
5.4.2
The points of equal loading are xed points of the projective transformations, for
example, of points m11 ! m12 ; m21 ! m22 , and J11 ! J21 ; J12 ! J22 , shown in Fig. 5.9.
-1
155
m1
(1)
J1
1
+1
m2
+1
m2
m1
(1)
J2
m2
J1
J1
1
J2
m1
J2
These transformations are another kind than those for variable load case (5.8) and
(5.10).
Similarly, it is possible to introduce two concepts; one of them denes a circuit:
how much the source loadings can differ. The second concept denes deviations of
actual loadings from the xed point in the relative form.
For introduction of such characteristics, we use also cross-ratio (5.13)
m2 m2 m1 m1
m2 m2 m1 m2
:
m2 m1 m1 m1
m2 m2 1 1 m1
m2 1 m2
a1
K:
1
d1
m2 1 m
5:33
5:34
156
m2 1
m1 1
:
m2 m1 m1 m1
5:35
From here
1
m2
Km
1 1K
m1 m1
1
m
m1 K 1K
5:36
1 Km1 m1 1 Km1
;
1 Km1 K m1
1
m
Therefore, d K 1K
.
Dependence (5.36) for different values of K is presented in Fig. 5.10. We obtain
the bunch of hyperboles with the common point (1).
If K K2 1, then
m2
m1 1
1
2 m1 m
1
m1 m 21
dm1 m1
; a d:
m1 d
The points m1 ; m2 are symmetric points concerning m1 and carry the special name
in projective geometry as harmonic conjugate points shown in Fig. 5.11. In particular, the point m1 d corresponds to the point m2 1 and vice versa. Also, the
center S2 will be on the straight line of equal loading. The cross-ratio for harmonic
conjugate points
m2 m1 m2 m1 1:
m2
(1)
(1)
K1 < -1,
1
2
3
S1
>
K2 = -1, 2=
K3 > -1, 3<
S2
S3
m
(1)
m1
-1
m1
157
(1)
m1
m2
m2
5.4.3
1
m1 1 mm11 11 1K
5:37
J2
J1
1K
:
1
K
J 1
J K
5:38
It should be noted, that expressions (5.37) and (5.38) set certainly deviation of running
parameters of loading from the equal loading regime by the normalized values. But it
is not enough for comparison of deviations for circuits with different parameters K. As
an example of the most simple relation (5.38), we will show, why it turns out.
~ but with the
Let us consider two circuits with the different parameters K; K,
1
identical value J 1. The characteristics of circuits are presented in Fig. 5.12;
we are reminding K\0.
As it was noted above, the loading regimes may be considered identical if
conformity of the characteristic regime points takes place (are shown by arrows) at
change of the load. Then, a projective transformation takes place and it is set by the
center at the point 0 and by three pairs of the characteristic regime points A, B, D,
~ B;
~ D.
~ The points D; D
~ coincide among themselves and correspond to the
and A;
1
~ For such
xed point J . The point of running regime C corresponds to the point C.
a projective transformation, the cross-ratio is carried out
~C
~D
~B
~ :
A C D B A
158
J2/J
(1)
1-K
A
-K
1-K
A
-K
D
1
D
C
C
0
1
B
B
(1-K)
J1/J
(1)
Therefore, this cross-ratio can be accepted as the equal deviation of the running
points C, C from unit points. Further, we map the points A; C; D; B onto the axis
of currents J1 . Then, similarly to (5.20), we obtain the deviations for the rst and
second sources at once
D1 0 J1 J 1 1 KJ 1
J1
K;
J1 1 KJ 1
5:39
D2 0 J2 J 1 1 KJ 1
J2
K:
J2 1 KJ 1
5:40
Thus, the deviations include the parameters of a circuit and are not simply the
normalized values J1 =J 1 ; J2 =J 1 .
With the aim to show the presented reasons for introduction of such deviations,
we will consider the special case K 1. Then, expressions (5.38) and (5.39)
became
J2 =J 1 2 J1 =J 1 ;
D1
J1 =J 1
2 J2 =J 1
1
:
1
1
D2
2 J1 =J
J2 =J
5:41
0 J1 J 1
1
d1
0 J2 J 1
1
d1
J1
a 1
:
J1 d 1 1 d 1
5:42
J2
a 1
:
J2 d 1 1 d 1
5:43
159
Similarly to (5.23), the deviations are expressed in the invariant form through the
cross-ratio for the variables m1 ; m2
D1
D2
0 J1 J 1
1
d1
1
d1
0 J2 J
1 m1 m1 d
1a
:
m2 d
1a
;
m1 d
5:44
5:45
m2 d
1 d
2
1:
2
^ d m
^ d
m
5:46
Then, we get
^ 2 1 2d:
m
Next, using (5.44), we nd
^ 2 1 a 1 a K:
D
^ 2 d 1 d
m
This value is also obtained from the similar condition to (5.46)
^ 2 D2 1
0 D
^ 2
D
D2
1:
Finally, we get
r 2 Ln K:
5:47
160
Using this value as the scale similar to (5.27), we have at once the same normalized
expressions
r1
r1
;
r 2
r2
r2
:
r 2
5:48
The values of deviations for the characteristic points and running points are presented in Fig. 5.13.
If K 1, then D2 1. Using (5.41), we get r1 r2 . In this case, there is
^ 2 1 D1 . From here, the physical sense of
no need to use the scale, because D
value K 1 follows: deviations of sources are equal by quantity, but are
opposite by a sign (for the entire equalizing resistor values that can be useful for
practice).
For this case, the deviations for the characteristic and running points are presented in Fig. 5.14.
Example 2 We consider the circuit in Fig. 5.1 and rewrite the required date of
Example 1.
Maximum currents of sources (5.3)
IM1 10;
IM2 12:
J1 0:0857; m2 13:4;
(2)
m = 1
m1
(2)
<0
(1)
J2 0:0694:
m2
m1, m2
( 2 ) >0
1, 2
(2)
r1
r2
r1, r2
r1
r2
r1, r2
-1
(2)
=1
m1
r1
(1)
1
0
m2
2=1/1
r2=r1
m1, m2
1, 2
r1, r2
6:707m1 12:13
;
m1 4:422
J2 0:703J1 0:13:
J 1 0:0761:
D2 0:858 1=1:165:
r2 Ln D2 0:1526:
The deviation for the second source turns out less than for the rst one.
Scale (5.47)
r 2 Ln 1:421 0:351:
Normalized distances (5.48)
r1
0:211
0:1526
0:601; r2
0:4347:
0:351
0:351
m1
4.422
6.707 10.66
(1)
m2
m1, m2
13.4
12.13
1, 2
1.235
0.858
0.351 0.211
-0.1526
r1, r2
r2
r1, r2
-1.421
1.421
r1
0.601
-0.4347
161
162
m1 1
V1 R0
5:5647
1;
2
V2 Ri1
Re2 R0
:
Ri2
From here
Ri1
V1 2R0
1:0325;
V2 m1 1
Ri2 2
Re2 R0
1:2052:
m1 1
where R0 ; Re2 ; V1 , and V2 are equal to the given initial values. Then, maximum
currents of sources (5.3)
IM1 12:1065;
IM2 9:9563:
d1
10:8825;
D1
m2 d
d1
13:67:
D2
1
0:0841;
m1 1
J2
1
0:0682:
m2 1
References
1. Frank, J.A.: Schaums outline of theory and problems of projective geometry. McGrawHill,
New York (1967)
2. Glagolev, N.A.: Proektivnaia geometria. (Projective geometry). Nauka, Moskva (1963)
3. Huang, Y., Tse, C.K.: Circuit theoretic classication of parallel connected DCDC
Converters. IEEE Trans. Circuits Syst. I Regul. Pap. 54(5), 10991108 (2007)
4. Irving, B.T., Jovanovi, M.M.: Analysis, design, and performance evaluation of droop
current-sharing method. In: The 15th Annual IEEE Applied Power Electronics Conference
(2000) http://www.deltartp.com/dpel/dpelconferencepapers/06_3.pdf. Accessed 30 Nov 2014
5. Kim, J.W., Choi, H.S., Cho, B.H.: A novel droop method for converter parallel operation.
IEEE Trans. Power Electron. 17(1), 2532 (2002) http://pearlx.snu.ac.kr/Publication/
IEEE0202.pdf. Accessed 30 Nov 2014
6. Penin, A.: Projectivelygeometric approach to parallel connection of electric energy sources.
Tehnicheskaia elektrodinamika 1, 3744 (1992)
References
163
7. Penin, A.: Analysis of paralleling limited capacity voltage sources by projective geometry
method. Sci. World J. (2014). doi:10.1155/2014/359893
8. Penin, A., Sidorenko, A.: Investigation of the effect of an equalizing resistor on the paralleling
voltage sources by projective geometry. Moldavian J. Phys. Sci. 11(12), 124131 (2012)
http://sfm.asm.md/moldphys/2012/vol11/n12/ins_14_penin.pdf Accessed 30 Nov 2014
9. Venikov, V.A.: Theory of similarity and simulation: with applications to problems in electrical
power engineering. Macdonald, London (1969)
10. Villarejo, J.A., De Jodar, E., Soto, F., Jimenez, J.: Multistage high power factor rectier with
passive lossless current sharing. In: The 7th WSEAS International Conference on Circuits,
Systems, Electronics, Control and Signal Processing (CSECS08) (2008)
11. Wang, J.B.: Primary droop currentsharing control of the parallel DC/DC converters system
considering output cable resistance. Adv. Power Electron. (2011). doi:10.1155/2011/713250,
http://www.hindawi.com/journals/ape/2011/713250/
Part II
Chapter 6
6.1
6.1.1
Let us give necessary relationships for an active two-port network in Fig. 6.1 with
changeable load voltage sourcesV1 ; V2 .
Taking into account the specied directions of the load currents, this network is
described by the following system of Y parameters equations [1, 5]:
"
I1
I2
Y11
Y12
6:1
where I1SC;SC ; I2SC;SC are the short circuit SC currents of both loads.
The inverse expression is
3 2
3
Y12
I1SC;SC I1
7
DY 7 6
7
4
5:
Y11 5
SC;SC
I2
I2
DY
6:2
I2 I2SC;SC
Y22
DY
I1 I1SC;SC
V1 ;
Y12
Y12
6:3
I2 I2SC;SC
Y12
DY
I1 I1SC;SC
V2 :
Y11
Y11
6:4
Y22
6
6 7 6 DY
4 54
Y12
V2
DY
V1
167
168
Y12
I1 I1SC;SC :
Y11
6:5
This straight line passes through the point SC; SC with the coordinates
I1SC;SC ; I2SC;SC or the point 0V . The value Y12 =Y11 corresponds to a slope angle.
Similarly, the axis V2 equation is dened by expression (6.3) as V1 0:
I2 I2SC;SC
169
Y22
I1 I1SC;SC :
Y12
6:6
The open circuit regime of both loads corresponds to the point OC; OC
with the coordinates I1OC;OC 0; I2OC;OC 0 or the point 0I .
Using (6.2), we get
V1OC
I
;
DY 1
DY 2
6:7
V2OC
I
:
DY 1
DY 2
6:8
We must map the point OC; OC onto the coordinate axes V1 ; V2 by parallel
lines to these axes.
Then, the coordinates or components I1OC;SC ; I2OC;SC correspond to the point V1OC
on the axis V1 . Using (6.7) and (6.1) as V2 0, we get
I1OC;SC
Y11 V1OC
I1SC;SC
Y11 Y22 SC;SC Y11 Y12 SC;SC
1
I
;
I1
DY
DY 2
Y12 Y22 SC;SC Y11 Y22 SC;SC
I
I
:
DY 1
DY 2
6:9
6:10
In turn, the coordinates I1SC;OC ; I2SC;OC correspond to the point V2OC on the axis
V2 . Using (6.8) and (6.1) as V1 0, we obtain
2
Y12
Y11 Y12 SC;SC
I
;
I SC;SC
DY 1
DY 2
Y12 Y22 SC;SC
Y11 Y22 SC;SC
I1
1
:
I2
DY
DY
6:11
6:12
Let an initial regime be given by values V11 ; V21 or by a point M 1 . Then, the
coordinates I11;SC ; I21;SC dene the point V11 on the axis V1 . Using (6.1) as V2 0,
we have
I11;SC Y11 V11 I1SC;SC ;
6:13
6:14
In turn, the coordinates I1SC;1 ; I2SC;1 dene the point V21 on the axis V2 .
Using (6.1) as V1 0,
170
6:15
6:16
Let us introduce the normalized coordinates for the point M 1 of this running
regime. The point OC; OC is the scale one for the coordinates V1 0V V2 . Then, the
normalized coordinates have the view
n11
V11 0V
V11
;
V1OC 0V V1OC
6:17
n12
V21 0V
V21
OC
:
OC
V2 0V V2
6:18
n12
I11;SC I1SC;SC
I1OC;SC
I1SC;SC
I1SC;1 I1SC;SC
I1SC;OC I1SC;SC
I21;SC I2SC;SC
I2OC;SC I2SC;SC
I2SC;1 I2SC;SC
I2SC;OC I2SC;SC
6:19
6:20
The obtained expressions are similar to expressions (2.6) and (2.9). Also, we
may obtain the other afne ratio similar to (2.7).
The presented approach will be used for analysis of inputoutput transformations
of a circuit with two inputs and two outputs.
Example 1 We consider the following circuit with given parameters in Fig. 6.3.
System of Eq. (6.1) has the view
"
I1
I2
171
2
3
" #
I1SC;SC
V1
5
4
Y12 Y22
V2
I2SC;SC
" # " #
V1
3
1:2 0:2
:
0:2 0:95
V2
2
Y11 Y12
y21N
1:2;
yR
y1N
0:2;
yR
Y22 y2 y2N
y22N
0:95:
yR
V1
V2
Y22
1
DY
Y12
0:95
1
1:1
0:2
Y12
Y11
2
4
I1SC;SC I1
I2SC;SC I2
#
"
3 I1
0:2
:
1:2
2 I2
3
5
I2
Y22
I1 I1SC;SC I2SC;SC 4:75I1 16:25:
Y12
172
DY 1
Y12 SC;SC
DY 1
V1OC
V2OC
1:1
1:1
DY 2
Y11 SC;SC 0:2
1:2
3
2 2:7272:
I2
1:1
1:1
DY
173
V11
1:958
OC 2:9545 0:6627;
V1
n12
V21
1:6494
OC 2:7272 0:6047:
V2
I11;SC I1SC;SC
I21;SC I2SC;SC
0:6627;
0:5454 3 2:5909 2
I1SC;1 I1SC;SC
I2SC;1 I2SC;SC
n12 SC;OC
I1
I1SC;SC I2SC;OC I2SC;SC
3:3298 3
0:433 2
0:6047:
3:5454 3 0:5909 2
6.1.2
Now, let us introduce the projective plane. For this purpose, we consider the most
simple active circuit, which contains two load conductivities YL1 ; YL2 in Fig. 6.5a.
If an internal resistance of voltage source V0 is equal to zero, the independent
change of load currents takes place at the independent change of the load conductivities. The point M of running regime can be set by the values of load conductivities MYL1 ; YL2 or by load currents MIL1 ; IL2 .
The family of load straight lines (there are parallel lines) coincides, for example,
with the rectangular Cartesian coordinates in the Euclidean plane in Fig. 6.5a. Then,
the calibrations of the coordinate axes, by the values of currents and conductivities,
coincide. It is obvious that the circuit does not possess the own scales. Therefore, it
is possible to express the regime only by the absolute or actual values of currents or
conductivities.
174
Fig. 6.5 a Simple active
circuit and its load straight
lines in the Cartesian
coordinates. b This load
straight lines in the projective
coordinates
(a)
(b)
We can represent the obtained family of load straight lines in the projective plane
[7, 10]. In this case, the innitely large currents and conductivities (innitely remote
points) are located in a nite domain in Fig. 6.5b and form the line of innity 1.
Therefore, the load straight lines represent two bunches of straight lines with the
centers in these innitely remote points.
Now, the internal resistance Ri of the voltage source V0 accepts a nite value in
Fig. 6.6. In this case, the dependent change of load currents takes place. There are
two bunches of load straight lines; the bunch centers are dened by the SC current
IM . Then, the straight line of the maximum current IM passes across these centers.
This straight line is similar to the line of innity 1.
The obtained deformed coordinate grid denes the projective plane. The axes
of coordinates can be calibrated by the values of corresponding currents or conductivities of the loads. In this case, there is an internal scale, that is, the value of
conductivity Yi 1=Ri or the current IM .
It is possible to accept that the load conductivities are equal, for example, to the
internal conductivity Yi and dene the point Mi of characteristic regime.
The projective coordinates are uniquely set by four points; there are three points
of the reference triangle G1 0G2 and scale point Mi [3, 4].
The point M of running regime can be set by the values of load conductivities
(non-uniform coordinates) MYL1 ; YL2 or by the load currents (homogeneous
coordinates) MIL1 ; IL2 ; VL . The sense of homogeneous coordinates consists that
they are proportional to the length of perpendiculars from a point to the sides of
reference triangle. The homogeneous coordinates are used for uncertainty elimination, when the point is on the line of innity 1. The presence of the fourth
175
(characteristic) point allows to introduce the cross-ratios mL1 ; mL2 and to set the
regime in the relative form.
6.1.3
Let us give necessary relationships for an active two-port network in Fig. 6.7 with
changeable load conductivities YL1 ; YL2 [9, 11]. Taking into account the specied
directions of the load currents, these active two-ports are described by (6.1)
"
I1
I2
" # " SC #
I1
V1
Y11 Y12
SC ;
Y12 Y22
V2
I2
6:21
V1
V2
176
Next, we use the circuit of Example 1. In our case, this circuit is shown in
Fig. 6.8. For convenience, we rewrite Y parameters
Y11 y1 y1N
Y12 y2N
y1N
;
yR
y21N
;
yR
Y22 y2 y2N
y1N
V0 ;
yR
y22N
:
yR
6:22
y2N
V0 :
yR
Taking into account the voltages V1 I1 =YL1 ; V2 I2 =YL2 , the equations of two
bunches of straight lines with parameters YL1 ; YL2 are obtained from system (6.21):
DY
V0 Y20 Y12 Y10 Y22 Y12 I2 I1 Y22
;
YL1
DY
:
V0 Y10 Y12 Y20 Y11 Y12 I1 I2 Y11
YL2
6:23
Bunches of these straight lines are presented in Fig. 6.9. The bunch center, a
point G2 , corresponds to the straight lines with the parameter YL1 . We would remind
that bunch center corresponds to such a regime of the load YL1 which does not
depend on its values. It is carried out for the current and voltage I1 0; V1 0 at
the expense of the second load YL2 parameters.
Then, the center G2 parameters will be as follows:
1 SC
I ;
Y12 1
y2
V0 y0N 1
;
y2N
V2G2
I2G2
Y22 SC
I I2SC
Y12 1
6:24
G2
YL2
G2
IL2
y2 y2N :
G2
VL2
177
6:25
V1G1
I1G1
Y11 SC
I I1SC
Y12 2
G1
YL1
G1
IL1
y1 y1N :
G1
VL1
6:26
6:27
178
m11 0 YL1
1
1
YL1
10
YL1
:
G1
1 Y G1
1 Y G1
1 YL1
YL1
YL1
L1
L1
6:28
1
G1
YL1
correspond to the base values. The point
There, the points YL1 0; YL1
YL1 1 is a unit point. Also, the values of m1 are shown in Fig. 6.9. For the point
1
G1
YL1
, the projective coordinate m1 1 denes the sense of the line of
YL1
innity G1 G2 .
The projective coordinate m12 is expressed by the same way
1
G2
m12 0 YL2
1 YL2
1
YL2
1
YL2
:
G2
YL2
6:29
m1 ;
n3 qn3
where q is a proportionality factor.
n2 qn2
m2 ;
n3 qn3
6:30
179
In turn, the homogeneous coordinates are dened by the ratio of the distances for
points M 1 ; SC to the sides of reference triangle
qn11
d11
I11
SC
;
SC
I1
d1
qn12
d12
I21
SC
;
SC
I2
d2
qn13
d13
:
dSC
3
6:31
6:32
s
1
1
l 3 2 2 ;
G1
G2
I1
I2
1 I1SC
I2SC
1
;
l3 I1G1 I2G2
1 l3 dSC
3 ;
I1G1 I2G2
I11
I21
1
I1G1 I2G2
6:33
l3 d13 :
6:34
I11
;
I1SC
q n12
I21
;
I2SC
q n13
I11
I21
1
:
SC
SC
G1
G2
I 1 l 3 d3
I2 l3 d3
l3 dSC
3
6:35
q n1
7 6
6
7 6
6
7 6
6
6qn 7 6
0
6 27 6
7 6
6
5 4
4
1
G1
q n3
I1 l3 dSC
3
I1SC
3 2
0
1
I2SC
1
G2
I2 l3 dSC
3
I1
7 6 7
7 6 7
7 6 7
6 7
0 7
7 6 I2 7 C I:
7 6 7
1 5 4 5
1
l3 dSC
3
6:36
180
We do not use the superscripts 1 and 2 for currents. In this equation, the
values I1 ; I2 ; 1 are the homogeneous Cartesian coordinates.
The inverse transformation is
3 2
q I1
I SC
6 q I 7 6 01
6 27 6
7 4 I SC
6
5
4
1
I1G1
q1
2
0
0
I2SC
I2SC
I2G2
3 2n 3
1
6
7 6 n2 7
1
76 7
7 C n:
5
5
4
SC
l3 d3
6:37
n3
qI1
I1SC m1
I SC
;
I SC
q1
1
m I2G2 m2 l3 dSC
3
I G1 1
1
6:38
qI2
I2SC m2
I2
I SC
:
I SC
q1
1
m I2G2 m2 l3 dSC
3
I G1 1
1
Thus, for preset values of the conductivities YL1 ; YL2 , we nd the coordinates
m1 ; m2 ; transformation (6.38) allows nding the currents I1 ; I2 .
It is possible to represent system of Eq. (6.38) by the normalized or relative
form:
I1
SC
I1G1 I1G1
m1
I1
I SC
1
I1G1
I1SC
I1G1
I2SC
I2G2
m1
I SC
I SC
m2 I1G1 I2G2 1
I1SC
I1G2
m1 1
m1
I2SC
I2G2
m2 1 1
6:39
I SC
2
m
I2
I2G2 2
:
SC
I SC
I2G2 I1G1
m1 1 2G2 m2 1 1
I1
I2
6:40
181
Also, system of Eq. (6.39) allows obtaining in the relative form the equations of
load characteristics I1 V1 ; V2 ; I2 V1 ; V2 which correspond to Eq. (6.21). For this
purpose, we express the non-uniform coordinates m1 ; m2 by currents and voltages:
m1
YL1
I1 =V1
I1 =I1G1
;
G1
G1
G1
G1
YL1 YL1
I1 =V1 I1 =V1
I1 =I1 V1 =V1G1
m2
I2 =I2G2
:
I2 =I2G2 V2 =V2G2
;
I1
V1G1 I1G1 V2G2 I1G1
I2
I2SC V1
I2SC V2
I2SC
1
:
I2G2
I2G2 V1G1
I2G2 V2G2 I2G2
I1
I1G1
6:41
6:42
I1G2 I1SC
;
V1G1
Y22
I2G1 I2SC
;
V2G2
Y12
I1SC
I2SC
:
V2G2
V1G1
6:43
182
Systems of Eqs. (6.39) and (6.40) are important relationships, as represent the
purely relative expressions. It allows dening, at rst, the coordinates m1 ; m2 (as
relative values) for different load conductivities, and then, the normalized currents.
Also, as (6.41), these expressions allow estimating at once a qualitative state of
such a circuit; how much a running regime is close to the characteristic values.
Then, the currents I1G2 ; I2G1 represent the scales. In this sense, initial systems of
Eqs. (6.21) and (6.22), containing the actual or absolute values of Y parameters,
currents and voltages, are uninformative, because they do not give a direct representation about the qualitative characteristics of a circuit.
In addition, it is possible to notice that it is difcult to obtain directly the relative
expressions of type (6.39)(6.43) from systems of Eqs. (6.21) and (6.22). In this
sense, the geometrical interpretation allows to solve this problem easily.
Example 2 We use the date of Example 1.
Equation (6.21) is given as
I1
I2
1:2
0:2
0:2
0:95
V1
V2
1
YL2
0:5;
I11 0:979;
I21 0:8247:
m12 0:3158:
l3 d13 0:8839;
l3 0:0907:
183
1
3
0
1
15 0:677
1
2
1
16:25 0:677
0
1
0:677
7
7
7
7:
7
5
60
C1 6
4 3
15
2
2
16:25
3
7
7:
5
0:677
2
0:25 16:25 0:3158 0:677 0:7658
2 0:3159
0:825;
I21
0:7658
15 3
16:25 2
Y11
1:2; Y22
0:95;
10
15
3
2
Y12
0:2:
15 10
I11
6.2
6.2.1
3
15
The popularized power supply system is shown in Fig. 6.11. Three and more loads
are connected to a common supply voltage source V0 or common node N by own
circuits (two-port TP1; TP2 and so on). The interference of loads is observed
because of the internal circuits by y0N ; yN of the voltage source.
Let us use the above approach for an active two-port network with changeable
conductivities of loads.
184
We now consider the concrete active multi-port network in Fig. 6.12. Similar to
(6.21), this network is described by the following system of equations [8, 11]:
2
I1
Y11
6 7 4
4 I2 5 Y12
Y13
I3
Y12
Y22
Y23
3 2 V 3 2 I SC 3
1
1
Y13
6 7 6 SC 7
5
Y23 4 V2 5 4 I2 5;
Y33
V3
I SC
3
where
Y11 y1 y1N
y1N
y1N
y2
; Y13 y3N
; Y22 y2 y2N 2N ;
yR
yR
yR
2
y2N
y
y3N
; Y33 y3 y3N 3N :
yR
yR
Y12 y2N
Y23
y21N
;
yR
6:44
185
y2N
V0 ;
yR
6:45
Y10
y0N
V0
V0 ;
Y12
y2N
G2
YL2
y2N y2 :
The point of intersection with the current axis I3 denes the third load:
y3
I3G3 V0 y0N 1
;
y3N
V3G3
y0N
V0 ;
y3N
G3
YL3
y3N y3 :
6:46
Similarly, the axis of the bunch YL2 corresponds to the condition I2 0 and is
located in the plane I1 ; I3 . The point of intersection with the current axis I1 denes
the rst load
186
I1G1
y1
V0 y0N 1
;
y1N
V1G1
y0N
V0 ;
y1N
G1
YL1
y1N y1 :
6:47
I11
;
I1SC
qn12
I21
;
I2SC
qn13
I31
;
I3SC
qn14
d14
:
dSC
4
6:48
1
;
d
1
;
4
l4 I1G1 I2G2 I3G2
l4 I1G1 I2G2 I3G2
s
1
1
1
l4 2 2 2 ;
G1
G2
G3
I1
I2
I3
where l4 is a normalizing factor.
The non-uniform projective coordinate m11 is set by the familiar cross-ratio:
1
G1
m11 0 YL1
1 YL1
1
1
YL1
10
YL1
:
G1
1 Y G1
1 Y G1
1 YL1
YL1
YL1
L1
L1
6:49
1
G1
YL1
correspond to the base values. The point
There, the points YL1 0; YL1
1
G1
YL1 1 is a unit point. For the point YL1 YL1
, the coordinate m1 1
denes the sense of the line of innity G1 G2 . The cross-ratio for m12 ; m13 is
expressed similarly. Therefore, we get the plane G1 G2 G3 of innity 1.
Also, the homogeneous projective coordinates n1 ; n2 ; n3 set the non-uniform
coordinates as follows:
n1 qn1
m1 ;
n4 qn4
n2 qn2
m2 ;
n4 qn4
n3 qn3
m3 :
n4 qn4
6:50
187
6:51
where
2
qn1
I1
6 qn 7
6I 7
6 27
6 27
qn 6
7; I 6 7;
4 qn3 5
4 I3 5
2
qn4
1
1
6
6
6
6
6
6
C 6
6
6
6
6
6
4
I1SC
I2SC
I1G1 l4 dSC
4
I2G2 l4 dSC
4
I3SC
1
G3
I3 l4 dSC
4
0
1
l4 dSC
4
7
7
7
7
7
7
7:
7
7
7
7
7
5
I2
I2G2
0
I3SC
0
0
I3SC
I3G3
l4 dSC
4
3
7
7
7:
7
5
6:52
qI1
SC
q1 I1G1
m1
I
1
qI2
qI3
I2
; I3
:
q1
q1
I1SC m1
I2SC
I2G2
m2
I3SC
I3G3
m3 l4 dSC
4
;
6:53
1
m
I1
I1G1 1
;
G1
m1 ; m2 ; m3
I1
I SC
2
m
I2
I2G2 2
;
G2
m1 ; m2 ; m3
I2
I SC
3
m
I3
I3G3 3
;
G3
m1 ; m2 ; m3
I3
6:54
188
where
1
SC
m1 ; m2 ; m3 I1G1
m1 1
I
1
1
I2SC
I2G2
m2 1
I3SC
I3G3
m3 1 1
Also, system of Eq. (6.54) allows obtaining in the relative form the equations of
load characteristics which correspond to Eq. (6.44). For this purpose, we express
the non-uniform coordinates m1 ; m2 ; m3 by the currents and voltages
m1
YL1
I1 =I1G1
;
G1
YL1 YL1
I1 =I1G1 V1 =V1G1
m2
I2 =I2G2
G2
I2 =I2 V2 =V2G2
m3
I3 =I3G3
G3
I3 =I3 V3 =V3G3
1
:
I3G3
I3G3 V1G1 I3G3 V2G2
I3G3 V3G3 I3G3
I1
I1G1
6:55
The obtained equations represent the purely relative expressions. The currents
I1G1 ; I2G2 ; I3G3 are the scales. It is obvious that relative expressions (6.55) are generalized to any number of loads by the formal way. If we use the absolute values of
the regime parameters of Eq. (6.54), then Y parameters are expressed by the
parameters of the characteristic regimes.
Example 3 We consider the circuit with given parameters in Fig. 6.14.
System of Eq. (6.44) is
2
I1
1:236
0:17
6 7 4
0:17 0:966
4 I2 5
0:282
0:188
I3
3 2 V 3 2 2:641 3
1
0:282
7
6 7 6
5
0:188 4 V2 5 4 1:761 5:
1:283
2:817
V3
G2
YL2
1:0833;
G1
1:5:
I1G1 15; YL1
G3
I3G3 14:844; YL3
1:583;
189
1
YL2
0:5;
I11 0:974;
1
YL3
1;
I21 0:82;
I31 1:61:
m12 0:316;
m13 0:387:
l4 d14 0:776;
l4 0:0907:
q n12 0:466;
q n14 1:476:
C1
2:641
6 0
6
4
0
0:176
0
0
1:761
0
0
2:817
0:108 0:19
3
0
0 7
7:
0 5
0:526
190
6.2.2
2:641 0:25
0:66
0:974:
0:176 0:25 0:108 0:316 0:19 0:387 0:526 0:67
Let us consider the general case of an active multi-port, for example, with three load
conductivities YL1 ; YL2 , and YL3 in Fig. 6.15 [1315]. The circuit is also described
by system of Eq. (6.44).
Therefore, the equations of three bunches of planes are obtained in the form
I1 ; I2 ; I3 ; YL1 0;
I1 ; I2 ; I3 ; YL2 0;
I1 ; I2 ; I3 ; YL3 0:
The equation of the axis of the bunch YL1 corresponds to the condition
I1 0; V1 0 and to equation I2 ; I3 0. Therefore, this axis is located in the
plane I2 ; I3 in Fig. 6.16a.
The points I2 YL1 ; I3 YL1 are the points of intersection with corresponding
axis. Similarly, we obtain the points I1 YL2 ; I3 YL2 of intersection of the bunch
axis YL2 and the points I1 YL3 ; I2 YL3 of intersection of the bunch axis YL3 .
On the other hand, the projective system of coordinates has to represent the
tetrahedron 0 G1 G2 G3 in Fig. 6.16b. In this case, we accept the plane, which passes
through three base points I1G1 ; I2G2 , and I3G3 on the axes of coordinates, as the plane
of innity 1. Therefore, the next conditions have to be satised
I1 YL2 I1 YL3 I1G1 ;
I3 YL1 I3 YL2 I3G3 :
191
(a)
(b)
V1 V1G1 :
Then,
I2 0; V2 0;
I3 0;
V3 0:
6:56
I2SC I3SC
:
Y12 Y13
6:57
192
Similarly, we consider
I2 I2G2 ;
V2 V2G2 :
Then,
8
< 0 Y12 V2G2 I1SC
I G2 Y22 V2G2 I2SC
: 2
0 Y23 V2G2 I3SC ;
V2G2
I1SC I3SC
:
Y12 Y23
6:58
I1SC I2SC
:
Y13 Y23
6:59
And we consider
I3 I3G3 ;
8
< 0 Y13 V3G3 I1SC
0 Y V G3 I SC
: G3 23 3 G32 SC
I3 Y33 V3 I3 ;
V3 V3G3 :
V3G3
V1 V1G1 :
193
V1G1
I4SC
I SC
N :
y41
yN1
V3 0; I3 0:
I5SC I6SC
:
yab yac
6:60
6:61
194
Therefore, we get
y5b
y6c
V5
V6 :
yab
yac
We determine the base values as
V1G1 UaG1
I1G1
;
y1a
G1
YL1
I1G1
;
V1G1
6:62
where
I4 V4 UaG1 y4a ;
Ia ya UaG1 :
Let the second load voltage be V2 V2G2 . For the others, we consider
V1 0; I1 0;
V3 0; I3 0:
I4SC I6SC
;
yab ybc
6:63
195
Therefore, we get
y4a
y6c
V4
V6 :
yab
ybc
6:64
I2G2
;
y2b
G2
YL2
I2G2
;
V2G2
6:65
where
I5 V5 UbG2 y5b ;
Ib yb UbG2 :
V2 0; I2 0:
Therefore,
UcG3
I4SC I5SC
:
yac
ybc
From here,
y4a
y5b
V4
V5 :
yac
ybc
6:66
I3G3
;
y3c
G3
YL3
I3G3
;
V3G3
6:67
where
I6 V6 UcG3 y6c ;
Ic yc UcG3 :
Usually, the voltage sources of distributed power supply system have equal
voltage values and different output powers. Let the voltage source V5 and load YL2 be
the most powerful elements. Therefore, the values of conductivities y5b ; ybc ; yab ; y4a
can be given independently, for example, based on the power effectiveness of
distributed network or current sharing among paralleled voltage sources.
Then, relative to these conductivities, using (6.64) and (6.66), we get the values
of the rest conductivities
196
y6c
ybc
y4a ;
yab
yac
ybc
yab
y4a
y6c :
y5b
y5b
6:68
1
I1G1
I SC
1
I1G1
I21
m11
I2SC
I2G2
m12
I2SC
I2G2
I3SC
I3G2
1
I1SC m11
m11 1
I2SC
I2G2
m12 1
qI21
SC
q1 I1G1
m1 1
1
I1
I31
I1SC m11
SC
I
m13 I1G1
I3SC
I3G3
qI31
SC
q1 I1G1
m1 1
1
I1
I3SC
I3G3
m13 1 1
;
6:69
I2SC m12
I2SC
I2G2
m12 1
I3SC
I3G3
m13 1 1
I3SC
I3G3
m13 1 1
I3SC m13
I2SC
I2G2
m12 1
;
:
Therefore,
2
qI11
I1SC
6 17
6 SC 7
6 qI 7
6I 7
6 27
1 6 2 7
qI 6
7 J 6
7 J1 ISC ;
6 qI 1 7
6 I SC 7
4 35
4 3 5
1
q1
1
m11
0
0
m12
6
6
6
J 6
0
0
6
4 1
1
m1 1 G2 m12 1
I1G1 1
I2
0
0
m13
1
I3G3
m13 1
3
0
07
7
7
0 7:
7
5
1
6:70
197
I2SC 1:246;
I3SC 0:6601:
Voltage source or input SC currents (6.61) are I5SC 2; I6SC 0:666 and current
is I4SC 1.
In turn, voltage value (6.60) is UaG1 8.
Currents (6.63) are I4 13=5 2:6; Ia 8=20 0:4.
Bunch centers (6.62), (6.65), and (6.67) are
I1G1 5:666;
V1G1 13:666;
I2G2 5:466;
V2G2 9:466;
I3G3 5:866;
V3G3 17:866;
G1
YL1
0:4146;
G2
YL2
0:5774;
G3
YL3
0:3283:
1
YL2
0:25;
I11 0:4232;
1
YL3
0:1666;
I21 0:5673;
I31 0:3350:
m12 0:3022;
m13 0:3366
l4 dSC
4 0:5073:
m11
m12
0:4553
0:3022;
1:5069
198
C1
0:8629
6 0
6
4 0
0:1523
0
0
1:246
0
0
0:6601
0:2279 0:1125
3
0
0 7
7:
0 5
0:5073
0:6636
1:246 0:3022
0:6601 0:3367
I21
0:5673; I31
0:335:
0:6636
0:6636
I11
qI11
0:3254
6 17
6 qI2 7 6
0
7 6
6
6 174
0
4 qI3 5
0:1191
q1
0
0:3022
0
0:1277
0
0
0:3367
0:1131
2 SC 3
3
I1
0
6 SC 7
6I 7
07
7 6 2 7:
7
05 6
4 I3SC 5
1
1
0:6636
0:3022 1:246
0:3367 0:6601
I21
0:5673; I31
0:335:
0:6636
0:6636
I11
6.3
The popularized power supply system is shown in Fig. 6.20. The low-dropout linear
regulators can be used as voltage stabilizers.
There are important features of such a circuit, that is, the interference of load
currents on the voltage stabilizers regimes takes place; the SC regime has no
physical sense and cannot be accepted as the characteristic regime.
199
This brings up the problem for the choice of characteristic regimes, justication
of the normalized expressions for parameters and equations of circuit, and comparison of regimes between the loads in a given circuit [12].
Let us introduce the projective coordinates using the results of Sect. 6.1.3.
Using Eqs. (6.21)(6.23), we get the following system of equations:
y0N yN y1N
y0N V0 V1 y0N yN I2 ;
y1N
y0N yN y2N
y0N V0 V2 y0N yN I1 :
I2
y2N
I1
6:71
200
YL1 which does not depend on y1N . It is carried out for the current I1 0 at the
expense of the second load current I2G2 . In this case, the voltage is VN V1 . Using
the rst equation of system (6.71), we obtain the characteristic value of the second
load current
I2G2 y0N V0 V1 y0N yN :
6:72
The characteristic value of the rst load current, which denes the bunch center
G1 of the straight lines with parameter y2N , is expressed similarly
I1G1 y0N V0 V2 y0N yN :
6:73
y0N yN
y0N V0 V1 y0N yN I1G1
y0N yN
I1G1 G2
I2 I1G1
I1G1
:
V2 V1
6:74
y0N yN
I2G2
:
I1G1 I2G2 V1 V2
6:75
1
1
; YL2
or
Let an initial regime corresponds to a point M 1 which is set by loads YL1
1
1
1
1
currents I1 YL1 V1 ; I2 YL2 V2 . The corresponding values of regulator conductivities are dened by system (6.71). However, using (6.72) and (6.73), we obtain
the more convenient relationships as
y11N I11
y0N yN
;
I2G2 I11 I21
y12N I21
y0N yN
:
I1G1 I11 I21
6:76
Also, this point M 1 is dened by the projective nonuniform m11 ; m12 and
homogeneous n11 ; n12 ; n13 coordinates which are set by the reference triangle G1 0 G2
and a unit point. The point 0 is the origin of coordinates and the straight line G1 G2
is the line of innity 1. As a unit point, we must also choose some characteristic
regime using the condition of stability of the voltages V1 ; V2 . As mentioned above,
the SC load current regime is not such one. However, the SC current regime of the
voltage source, when VN 0, allows nding such a characteristic regime. In this
case, the SC current of V0 is
201
I0M y0N V0 I1 I2 :
This expression corresponds to an equation of straight line. In Fig. 6.21, this line
intersects the coordinate axes in the points I1 I0M ; I2 I0M . The straight lines
SC
with the parameters ySC
1N ; y2N correspond to these points.
SC
Let us determine the values ySC
1N ; y2N . We assume the current I2 0 for the rst
and I1 0 for the second equation of (6.71). Then, the values are
ySC
1N y0N V0 =V1 ;
ySC
2N y0N V0 =V2 :
6:77
Now, using (6.77) and Eq. (6.71), we can dene the load currents
I1SC V1
I2SC
I1G1
;
V2 V1 I1G1 =I0M
I2G2
V2
:
V1 V2 I2G2 =I0M
6:78
y11N 0
ySC
1N 0
:
G1
SC
1
y1N y1N y1N yG1
1N
6:79
SC
The points y1N 0; y1N yG1
1N correspond to the base points. The point y1N is a
unit one. The values of m1 are shown in Fig. 6.21. The nonuniform projective
coordinate m12 is expressed similarly
G2
m12 0 y12N ySC
2N y2N
y12N
y12N
ySC
SC 2N G2 :
G2
y2N y2N y2N
6:80
These expressions demonstrate the conductivities y1N ; y2N in the relative form.
The homogeneous projective coordinates n1 ; n2 ; n3 set the nonuniform coordinates by (6.30). For convenience, we rewrite (6.30) and all the required formulas
qn1
m1 ;
qn3
qn2
m2 :
qn3
6:81
202
q n1
3 2
7 6 I1SC
6
7 6
6
7 6
6
6 q n2 7 6
0
7 6
6
7 6
6
5 4
4
1
q n3
I1G1 l3 dSC
3
I1
7 6 7
7 6 7
7 6 7
6 7
0 7
7 6 I2 7 C I:
7 6 7
1 5 4 5
1
l3 dSC
3
1
I2SC
1
I2G2 l3 dSC
3
6:82
0
0
I2SC
I2SC
I2G2
3 2n 3
1
6
7 6 n2 7
76 7
C1 n:
5 4 7
5
SC
l3 d3
6:83
n3
I2
qI I1
I1SC m1
SC
;
qI 1 I1
I2SC
SC
m
m
l
d
1
2
3 3
I1G1
I2G2
qI I2
I2SC m2
SC
:
qI 1 I1
I2SC
SC
m1 G2 m2 l3 d3
I1G1
I2
6:84
q n3 m 1
7 6 I1SC
6
7 6
6
7 6
6
6 q n3 m 2 7 6
0
7 6
6
7 6
6
5 4
4
1
G1
q n3
I1 l3 dSC
3
3 2
0
1
I2SC
1
G2
I2 l3 dSC
3
I1
7 6 7
7 6 7
7 6 7
6 7
0 7
7 6 I2 7 C I:
7 6 7
1 5 4 5
1
l3 dSC
3
6:85
203
From here
I1
l dSC
qn3 m1
I1SC 3 3
m1
;
I1
I2
qn3
1
I1G1 I2G2
I2
l dSC
qn3 m2
I2SC 3 3
m2
:
I1
I2
qn3
1
I1G1 I2G2
6:86
m11 yG1
1N
;
yG1
1N
1
m1 SC 1
y1N
y12N
m12 yG2
2N
:
yG2
2N
1
m2 SC 1
y2N
6:87
y0N 2:5;
yN 0:625;
V1 2;
V2 3:
For the rst load, characteristic currents (6.73) and (6.78) are
I1G1 3:125;
I1SC 1:785;
ySC
1N 6:25:
I2SC 5:357;
yG2
2N 6:25;
ySC
2N 4:166:
I21 1:
y12N 2:777:
m12 0:154:
The negative values mean that the point M1 and a unit point SC are on the
different sides from the innitely remote straight line G1 G2 .
Distances (6.34) and (6.33) are
204
l3 d13
l3 dSC
3
1
1
1 0:52;
3:125 6:25
1:785 5:357
1 0:428:
3:125
6:25
1
;
1:785
qn12
1
0:186;
5:357
qn13
0:52
:
0:428
1
1:785
0
0
6
6
1
6
C 6
5:357
6
4
1
1
3:125 0:428 6:25 0:428
0
0
1
0:428
7
7
7
7:
7
5
1
6
6
6
6
4
1:785
5:357
1:785
3:125
5:357
0:428
6:25
3
7
7
7:
7
5
5:357
0:82
0:461
0:153
0:428
1:785
3:125
6:25
5:357 0:153
1:
I21
0:82
I11
1
1:785 0:428
1
1
3:125 6:25
1
5:357 0:428
1
1
3:125 6:25
1
1
0:2397
0:461;
0:52
0:07989
0:154:
0:52
References
205
References
1. Alexander, C.K., Sadiku, M.N.O.: Fundamentals of Electric Circuits, 5th edn. McGrawHill,
New York (2009)
2. Bennett, M.K.: Afne and Projective Geometry. Wiley, Hoboken (2011)
3. Frank, J.A.: Schaums Outline of Theory and Problems of Projective Geometry. McGraw
Hill, New York (1967)
4. Glagolev, N.A.: Proektivnaia Geometria. (Projective geometry). Nauka, Moskva (1963)
5. Irwin, J.D., Nelms, R.M.: Basic Engineering Circuit Analysis, 10th edn. Wiley, Hoboken
(2011)
6. Penin, A.: Projective equivalency of operating regimes for two-ports. Electrichestvo 8, 4754
(1993)
7. Penin, A.: Fractionally linear relations in the problems of analysis of resistive circuits with
variable parameters. Electrichestvo 11, 3244 (1999)
8. Penin, A.: Normalized representation of the equations of active multi-port net works on the
basis of projective geometry. Moldavian J. Phys. Sci. 10(34), 350357 (2011). http://sfm.
asm.md/moldphys/2011/vol10/n3-4/index.html. Accessed 30 Nov 2014
9. Penin, A.: Recalculating the load currents of an active multiport with variable parameters on
the basis of projective geometry. Electrichestvo 10, 6673 (2012)
10. Penin, A.: Projective geometry method in the theory of electric circuits with variable
parameters of elements. Int. J. Electron. Commun. Electr. Eng. 3(2), 1834 (2013). https://
sites.google.com/site/ijeceejournal/volume-3-issue-2. Accessed 30 Nov 2014
11. Penin, A.: Recalculation of the loads current of active multiport networks on the basis of projective
geometry. J. Circuits Syst. Comput. 22(05), 1350031 (13 pages) (2013) doi: 10.1142/
S021812661350031X. http://www.worldscientic.com/doi/abs/10.1142/S021812661350031X.
Accessed 30 Nov 2014
12. Penin, A.: Comparison of regimes of active two-port networks with stabilization of load
voltages. Inter. J. Electron. Commun. Electr. Eng. 3(6), 118 (2013). https://sites.google.com/
site/ijeceejournal/
13. Penin, A., Sidorenko, A.: Balanced multi-port electric network and its projective coordinates.
Moldavian J. Phys. Sci. 14(12), 102112 (2015)
14. Penin, A., Sidorenko, A.: Sistemul distribuit de alimentare cu energie. MD Patent application
S20150047, 2015
15. Penin, A., Sidorenko, A., Donu, S.: Metod de transmitere a trei semnale prin linia de
comunicaie cu patru re. MD Patent application S20150082, 2015
Chapter 7
7.1
7.1.1
Let us continue with the matter we began in Sect. 6.1.3. Let a subsequent regime
2
2
; YL2
and currents I12 ; I22 of loads.
correspond to a point M 2 with conductivities YL1
The non-uniform coordinates are dened similarly to (6.28) and (6.29)
m21
2
YL1
2
YL1
;
G1
YL1
m22
2
YL2
2
YL2
:
G2
YL2
m21
2
2
YL2
1
YL2
G2
YL2
2
YL1
0
Y1 0
1 L1 G1 m21 m11 ;
G1
2
YL1 YL1 YL1 YL1
m22
7:1
m12 :
Using (6.31) and (6.34), we dene the homogeneous coordinates of the point M 2
q n21
I12
I22
d23
SC
; q n22 SC
; q n23 SC
; l3 d23
I1
I2
d3
I12
I22
1 :
I1G1 I2G1
207
208
I2
G2
YL2
21
m2
I2
YL2
1
I2
0
G1
2
I1
I1
YL1
YL1
21
m1
I1SC
I1G1
I1SC m21
;
I SC
m21 2G2 m22 l3 dSC
3
I2
I22
I1SC
I1G1
I2SC m22
:
I SC
m21 2G2 m22 l3 dSC
3
I2
Taking into account (7.1), we present the non-uniform coordinates m21 ; m22 as
m21 m21
1
n11
;
n13
m22 m21
2
n12
:
n13
I12
I SC
1
I1G1
m21
1
n11
n13
1
I1SC m21
1 n1
SC 3
I
I2G2 m21
2
n11
n13
2
I2SC m21
2 n1
SC 3
I
I2G2 m21
2
n12
n13
l3 dSC
3
n12
n13
l3 dSC
3
n1
I22 I SC
1
I1G1
m21
1
I1
209
C21 1
I1SC m21
1
6 0
4
I1SC
I1G1
m21
1
0
I2SC m21
2
I2SC
I2G2
0
0
l3 dSC
3
m21
2
3
7
5:
7:2
m21
1
6
J 21 4 0
1
m21
1 1
I G1
1
0
m21
2
1
m21
2 1
I G2
3
0
07
5:
1
q I12
q1
q I2
2
q1
I11 m21
1
I11
I1G1
m21
1 1
I21
I2G2
m21
2 1 1
I21 m21
2
I11
I1G1
m21
1 1
I21
I2G2
m21
2 1 1
;
7:3
:
m32
1
6 0
32
J 4 32
m1 1
I1G1
0
m32
2
m32
2 1
I2G2
3
0
07
5;
1
m31
1
6 0
31
J 4 31
m1 1
I1G1
0
m31
2
m31
2 1
I2G2
3
0
07
5:
1
210
21
Projective transformation (7.2) or (7.3) with parameters m21
1 ; m2 translates any
initial points of the plane I1 ; I2 into a new position shown in Fig. 7.1. The xed
points and xed straight lines are shown too.
m21
2 0:6486 0:3158 2:0538:
Transformation (7.2)
2
qI12
1:6
0
6 27 4
0
2:0538
4 qI2 5
0:04 0:0648
q1
3 2 I1 3
1
0
6 7
0 5 4 I21 5:
1
1
1:4335;
0:04
0:979
0:0648
0:825
1
1:0926
q1
qI 2 2:0538 0:825
1:5507:
I22 2
1:0926
q1
I12
Once again we notice that the offered formulas of recalculation are especially
21
convenient, when the xed values m21
1 ; m2 are used for any values of initial
currents.
7.1.2
Active Three-Port
211
2
YL1
;
2 Y G1
YL1
L1
m22
2
YL2
;
2 Y G2
YL2
L2
m23
2
YL3
:
2 Y G3
YL3
L3
2
1
m21
2 m2 m2 ;
2
1
m21
3 m3 m3
7:4
1 I12
I22
I32
1 :
l4 I1G1 I2G2 I3G2
Then
q n21
I12
;
I1SC
q n22
I22
;
I2SC
q n23
I32
;
I3SC
q n24
d24
:
dSC
4
Repeating a mental step for the plane, we obtain the required transformation
similarly to (7.2) and matrix
2
6
6
J21 6
4
m21
1
0
0
0
m21
2
0
m21
1 1
I1G1
m21
2 1
I2G2
0
0
m21
3
m21
3 1
I3G3
3
0
07
7
:
07
5
1
7:5
qI12
q1
I22
qI22
q1
I32
qI32
q1
I11 m21
1
I11
I1G1
m21
1 1
I21
I2G2
m21
1 1
I21
I2G2
m21
1 1
I21
I2G2
m21
2 1
I31
I3G3
m21
3 1 1
I31
I3G3
m21
3 1 1
I31
I3G3
m21
3 1 1
I21 m21
2
I11
I1G1
m21
2 1
I31 m21
3
I11
I1G1
m21
2 1
;
;
:
7:6
212
Special case
Only the load conductivity YL3 is changed. The matrix
2
1
60
6
J21 6 0
4
0
1
0
0
0
m21
3
m21
3 1
I3G3
3
0
07
7
:
07
5
1
Example 2 We continue Example 3 of Sect. 6.2 and rewrite the initial regime
parameters
1
1
1
YL1
0:5; YL2
0:5; YL3
1;
m22 0:6486;
m23 0:9266:
m21
2 0:6486 0:3158 2:0538;
m21
3 0:9266 0:387 2:3948:
Matrix (7.5)
2
1:6
6 0
21
J 6
4 0
0:04
0
0
2:0538
0
0
2:394
0:0648 0:0939
3
0
07
7:
05
1
1:254;
0:04 0:974 0:0648 0:82 0:09396 1:61 1 1:2433
2:0538 0:82
2:3394 1:61
I22
1:356; I32
3:1:
1:2433
1:2433
I12
7.2
213
7.2.1
y1N
1.25
y0N
V0
5
0.833
YL1
V2
y2
I2
0.25
y2N
V1
y1
0.25
yN
YL2
I2
G2
1
SC
m2(yN)
SC
yN
0
yN yN
m1(yN) 1
G1
yN
yN
m1(yN) 1
G1
yN
I1
yN
m1(yN)
214
(7.2) and (7.3), can recalculate the initial currents I1 ; I2 of the point M to the
subsequent currents I1 ; I2 of the point M.
The point M has the homogeneous coordinates n1 ; n2 ; n3 , which are set by
reference triangle G1 0 G2 and a unit point SC. On the other hand, the xed reference triangle G1 0 G2 and a new unit point SC form a subsequent system of
has the same homogeneous coorprojective coordinates. Therefore, the point M
dinates n1 ; n2 ; n3 ; that is,
qn n:
Using (6.36), we may write
I;
qn C
qn C I:
3
0
0 5;
J33
J11
J 4 0
J31
0
J22
J32
I1SC
;
I1SC
I2SC
;
I2SC
7:7
J22
J33
!
SC SC
dSC
I
d
SC3 G1 1SC 3SC 1 ;
I1 d3
d3 I1
!
dSC
I2SC dSC
3
3
SC G2 SC SC 1 :
I2 d3
d3 I2
dSC
3
;
dSC
3
Dividing all these elements by J33 , we obtain the resultant matrix in the form
2
I SC dSC
3
1
I1SC
dSC
3
6
6
6
J 6 0
6 I1SC dSC
4 I SC d3SC 1
1
I1G1
3
0
I SC dSC
3
2
I2SC dSC
3
I SC dSC
2 3 1
I SC dSC
2 3
I2G2
7
7
0 7:
7
7
5
1
215
I1SC dSC
3
;
I1SC dSC
3
m2 yN
I2SC dSC
3
:
I2SC dSC
3
7:8
Because the conductivity yN has an equal effect on the load currents, these
parameters are equal to each other; that is, m1 yN m2 yN mN .
Finally, we obtain the above desired transformation
2
3 2
mN
qI1
4 qI2 5 4 0
mN 1
q1
I1G1
0
mN
mN 1
I2G2
3 2 3
0
I1
0 5 4 I2 5 :
1
1
7:9
I1
I1G1
I1 mN
;
mN 1 IIG22 mN 1 1
2
I2 mN
I2
:
I1
I2
m
1
mN 1 1
N
I G1
I G2
1
7:10
The obtained relationships carry out the recalculation of currents at the respective change of conductivity yN in the form of parameter mN . Transformation (7.9) is
a projective transformation and possesses group properties.
Let us formulate the value mN by the initial and subsequent currents I11 ; I11 . To do
this, we use expression (7.10) for the current I1 as I2 0.
Then
I1
I1 m N
I1
I1G1
mN 1 1
7:11
I1
I1G1 :
216
I1
mN
SC
I1
I1
SC
I1
G1
I1
I1
The points I1 0; I1 I1G1 are xed points and do not depend on the value
yN . Therefore, expression (7.11) is the transformation I1 ! I1 with parameter mN as
it is shown in Fig. 7.4.
Therefore, the cross-ratio has the same value for the points I1SC ; I1SC
mN 0 I1 I1 I1G1 0 I1SC I1SC I1G1
I1 0
I SC 0
I1 0
I SC 0
SC1
SC1
:
G1
G1
G1
I1 I1
I1 I1
I1 I1
I1 I1G1
7:12
Now, let us formulate the value mN by yN ; yN and determine the sense of the
parameter mN .
Viewing expressions (6.28), we may consider expression (7.8) of mN as a
non-uniform coordinate
mN
I1SC dSC
I SC I SC
3
1SC 1SC n13 n13 :
SC SC
d
I 1 d3
d3
3
Hence, the value mN is the cross-ratio of the initial SC and subsequent points SC
shown in Fig. 7.3.
0
The base values of conductivity yN are values y0N ; yG1
N . The value yN 1
G1
G1
determines the current I1 0. The value yN presets the current I1 I1 as I2 0.
Analysis of the circuit gives the following relationship for yG1
N
yG1
N y0N
y1 y1N
y2 y2N
yiN :
y1 y1N
y2 y2N
7:13
yN
yiN
yN 1 yN 1 yN yiN
:
yN yiN yN yiN yN yiN
7:14
Now, we formulate the value mN by the yN ; yN for the general case of currents
I1 ; I2 .
217
y1N
:
yR
;y
y0N y1N y2N ;
I1SC yR ySER yN
SC
SC
I1
I2SC
dSC
I1
I2SC
3
G1 G2 1
1
I1
I2
I1G1 I2G2
dSC
3
Finally, we have
mN
I1SC dSC
yN yiN
3
:
SC
SC
yN yiN
I1 d3
yN
yN
yN
yN
mN
mN
218
y1N
(a)
y0N
1.25
y1
y2N
0.25
0.833
2.5
y2
yN
y1N
(b)
y0N
V0
5
0.25
1.25
y1
y2N
0.25
0.833
2.5
+
yN
y2
0.25
3 2
qI1
0:8494
6 7 4
0
4 qI2 5
0:01
q1
0
0:8494
0:009262
3 2I 3
1
0
6 7
5
0 4 I2 5 :
1
1
0:8465;
0:01 0:979 0:009262 0:8247 1 0:9825
I2 0:8494 0:8247 0:713:
0:9825
I1
YL1
G1
I1=I1
219
0:25 1:25
0:25 0:833
2:9 :
0:25 1:25 0:25 0:833
7.2.2
yN yiN
1:25 2:9
We consider again the circuit in Fig. 7.2. Let the conductivity y0N be changed,
and short circuit point
y0N ! y0N . Corresponding changes of an initial point M ! M
SC ! SC are shown in Fig. 7.7. Also, the coordinates of points G1 ; G2 by
(6.24) and (6.26) are proportional to the value y0N . Therefore, the subsequent straight
2 is parallel to the initial line G1 G2 ; the values Y G1 ; Y G2 do not change.
1 G
line G
L1
L2
Let us determine xed points and lines as y0N change. Naturally, the point 0 does
not depend on this element. If the current across conductivity y0N is equal to zero, a
straight line S1 S2 is the xed line and we have the following equation of this line:
y1N
y2N
I1
I2 yi0N V0 0:
y1 y1N
y2 y2N
I2
G2
G2
SC
SC
M
SC
d3
S1
M
0
S1
I1
G1
S2
I2
G1
S2
I1
220
y1N
(a)
1.25
y1
y2N
0.25
N
i
yN
y 0N
0.833
y2
0.25
0.625
y1N
(b)
i
y0N
V0
5
1.25
y1
y2N
0.25
0.833
+
yN
0.625
y2
YL1
S1
I1=I1
0.25
y1 y1N
y2 y2N
:
y1 y1N
y2 y2N
7:15
1 0;
I1S1 I2S2
where the points
y1
I1S1 yi0N 1
V0 ;
y1N
y2
I2S2 yi0N 1
V0
y2N
7:16
In turn, if y0N yi0N , the determinant DY 0. In this case, we have the load
1
current I1 I1S1 for the rst given load YL1
and the disconnected second load, as it is
shown in Fig. 7.8b. And vice versa, the current I2 I2S2 is preset for the second
1
given load YL2
and the disconnected rst load.
221
I1
;
I1SC
q n2
I2
;
I2SC
q n3
d3
:
dSC
3
1
;
l3 I1S1 I2S2
1 I1SC
I2SC
1
:
l3 I1S1
I2S2
d3
dSC
3
7:17
qn1
;
qn3
m2
qn2
:
qn3
I2
1
m1
0
m1
1 SC
1
1
m2
m2
0
0
S1
I1
S2
222
1
I2SC
1
I1S1 l3 dSC
3
1
I2S2 l3 dSC
3
l 1dSC
C1
1
I1SC
I1SC
6
0
6
4 SC
I1
I1S1
0
I2SC
0
0
I2SC
I2S2
l3 dSC
3
7
7
7;
5
3 3
3
7
7:
5
On the other hand, the same reference triangle S1 0S2 and a new unit point SC
form the other system of projective coordinates and distance dSC
3 .
Then, the required transformation is similar to (7.9)
3 2 3
3 2
2
I1
qI1
0
0
m 1 y0N
7 6 7
6 7 6
0
m
y
0
2
0N
q
I
5 4 I2 5;
4 25 4
m 1 y0N 1
m 2 y0N 1
1
S1
S2
I
I
q1
1
1
I1SC dSC
3
;
I1SC dSC
3
m2 y0N
I2SC dSC
3
;
I2SC dSC
3
7:18
Because the conductivity y0N has an equal effect on the load currents, these
parameters are equal to each other; that is, m1 y0N m2 y0N m0N :
Finally, we obtain
2
3 2
3 2I 3
qI1
0
0
m0N
1
6 7 4 0
7
m0N 0 5 6
7:19
4 qI2 5
4 I2 5 :
m0N 1 m0N 1
1
S1
S2
I
I
q1
1
1
I1
I1S1
I1
I1S1
m0N
I1 m0N
;
1 IIS22 m0N 1 1
m0N
I2 m0N
:
1 IIS22 m0N 1 1
7:20
Let us formulate the value m0N by the initial and subsequent currents I1 ; I1 .
223
I1
I1S1
I1 m0N
:
m0N 1 1
7:21
I1
I1S1 0 I1SC
I1S1 0 I1G1
I1SC
I1G1
I1S1 :
Now, we formulate the value m0N by y0N , y0N and determine the sense of this
parameter m0N . According to the above case, we get
m0N 0 y0N y0N yi0N
y0N 0
y0N 0
:
y0N yi0N y0N yi0N
7:22
The value m0N can be considered as a relative change of the conductivity y0N .
In turn, the value yi0N is a scale for y0N . The correspondence of values m0N and
y0N is shown in Fig. 7.11.
Example 4 We use the date of Example 1 for the initial regime
y0N 2:5; I1 0:979; I2 0:8247;
I1SC 3; I2SC 2; l3 dSC
3 0:677:
Let the subsequent regime parameters be given as
y0N 2; I1 0:896; I2 0:7545;
I1SC 2:6548; I2SC 1:77; l3 dSC
3 1:6969:
m0N
m0N
S1
I1
I1
I1
SC
m0N
SC
G1
G1
I1
I1
y0N
y0N
y0N
m0N
I1
I1
y0N
m0N
I1
224
3 2
qI1
0:9323
0
6 7 4
0
0:9323
4 qI2 5
0:011 0:01016
q1
3 2I 3
1
0
6 7
0 5 4 I2 5:
1
1
0:896;
0:011 0:979 0:01016 0:8247 1 1:01914
0:9323 0:8247
0:7688
I2
0:7545:
0:011 0:979 0:01016 0:8247 1 1:01914
I1
y1 y1N
y2 y2N
0:9323:
m0N
7.3
Let two active two-ports be given with different parameters. Bunches of load
straight lines and the conformity of points of characteristic and running regimes are
presented in Fig. 7.12.
The problem is to nd such values of conductivities and currents of loads of the
that its relative regimes would be equal to regimes of
second two-port (a point M)
the rst two-port (a point M with preset values of conductivities YL1 ; YL2 and
currents I1 ; I2 ).
For this purpose, we consider systems of projective coordinates of these
10 G
2
two-ports [1]. Such systems are set by the triangles of reference G1 0 G2 , G
and unit points SC, SC respectively.
225
I2
G2
G2
SC
SC
M
M
G1
G1
I1
YL1
YL1
;
G1
G1
YL1 YL1
YL1 YL1
m2
YL2
Y
L2 G2 :
G2
YL2 YL2
YL2 YL2
7:23
m2
:
m2 1
7:24
m1
;
m1 1
G2
YL2 YL2
7:25
226
1
C
I1SC
6 0
4
2
6
C 6
4
0
I2SC
I SC
1
I G1
1
I SC
2
I G2
2
1
I1SC
0
1
I1G1 l3 dSC
3
3
0
0 7
5;
l dSC
7:26
3 3
1
I2SC
1
I2G2 l3 dSC
3
0
l 1dSC
3
7
7:
5
7:27
3 3
J11
J 4 0
J31
0
J22
J33
3
0
0 5:
J33
7:28
J11
J31
J32
The form of these expressions shows that we may divide all the elements by J33
and introduce values
1
m
I1SC l3 dSC
3
;
I1SC l
3 dSC
3
2
m
I2SC l3 dSC
3
:
I2SC l
3 dSC
3
7:29
q1
1
m
G1 0
1 I1
m
1
G1
G1
1
I
I
1
0
m
G2 2
1 I2
m
1
G2
G2
2
I
I
2
3 2 3
I1
0
6 7
07
5 4 I2 5:
1
1
7:30
227
In turn,
I1
I2
I1
I1G1
I1
I1G1
G1
I1
I G1
1
G1
I1
I G1
1
1
I1 m
1 1
m
I2
I2G2
2
I2 m
1 1
m
I2
I2G2
G2
I2
I G2
2
G2
I2
I G2
2
;
2 1 1
m
2 1 1
m
7:31
I G2
N2 m
2 2G2 :
m
I2
q
6
4q
I1
I G1
1
I2
I G2
2
1
q
6
7 6
56
6
4
N1
m
0
N1 1
m
2I 3
1
7
I1G1
7
6
I2 7
N2
m
07
7 4 I2G2 5:
5
1
N
2 1 1
m
0
7:32
Let the relative regimes of these two-ports be equal to each other by (7.23)
m1 0:25;
m2 0:3158:
228
G1
YL1 YL1
YL2
SC currents
I1SC 2:434;
I2SC 1:549:
The distances
3
dSC
l
3 0:7618;
3 d13 0:9211;
l
3 0:08362:
l
Distinctions (7.29)
1 0:721;
m
2 0:6882:
m
Transformation (7.30)
3 2
I1
q
0:721
6 7 4
I2 5
0
4q
0:0216
1
q
2
3 2I 3
1
0
0
6 7
0:6882 0 5 4 I2 5:
0:0233 0
1
Currents (7.31)
0:721 0:979
0:7058
0:7355;
0:0216 0:979 0:0233 0:8247 1 0:9596
I2 0:6882 0:8247 0:5915:
0:9596
I1
7.4
We use the circuit in Fig. 6.20. For convenience, we redraw this circuit in Fig. 7.13.
Let us consider the two similar circuits with different values of element parameters and regime parameters. It is necessary to prove an approach to the comparison
or recalculation of running regimes of such circuits. The characteristics of these
comparable circuits are given in Fig. 7.14. The condition of regime comparison is the
conformity of characteristic regimes as a projective transformation [4].
229
y1N
y0N
V1
y2N
V2
V0
+
yN
YL1
YL2
y2N
I2
SC
y2N
I0M
I0M
G2
y2N
G2
G2
G2
y2N
1
y2N
SC
y2N
M
M
0
SC
I1
I0M
1
y1N
y1N
G1 G1
I0M
SC
y1N
y1N
7:33
7:34
230
M11
M 4 0
M31
0
M22
M32
3
0
0 5;
M33
7:35
I1SC
I SC
dSC
l
; M22 2SC ; M33 3 3SC ;
SC
I1
I2
l3 d3
SC
I1SC 1
I SC 1
d
dSC
l
l
G1 SC G13 3 SC ; M32 2G2 SC G23 3 SC :
I I
I I
I ld
I l d
M11
M31
3 3
3 3
The similar condition for the conformity of regulator conductivities follows from
(6.79) and (6.80). In particular, the conductivity y1N is dened by the equality
y11N
y1
ySC
y11N
ySC
SC 1N G1 1 1N G1 SC 1N G1 :
G1
y1N y1N y1N y1N y1N y1N y1N
7:36
y21N
ySC
1N
:
y21N yG1
ySC
yG1
1N
1N
1N
y21N
y11N
21 m
11 :
m
y21N yG1
y11N yG1
1N
1N
This change is also equal to the change of regime of the rst circuit
21
2
1
21
m
1 m1 m1 m1 :
7:37
231
Let us express these changes through load currents. Using (6.81) and (6.31), we
obtain
m21
1
I12 l3 d13
;
I11 l3 d23
m21
2
I22 l3 d13
:
I21 l3 d23
7:38
21
Let the changes m21
1 ; m2 and the initial equal regime, the points M1 ; M1 , be
given.
At rst, it is necessary to nd load currents of these regime changes, the points
2 . Similarly to (7.2), we have for the second circuit
M2 ; M
i
h i
2
1
qI J21 I ;
7:39
The matrix
2
m21
1
21
6
0
J 4
1
21
I G1 m1 1
0
m21
2
1
21
I G2 m2 1
3
0
07
5:
1
I 1
1
I G1
I11 m21
1
I 1
21
2
m21
1 1 I G2 m2 1 1
I 1
1
I G1
1
21
2
m21
1 1 I G2 m2 1 1
;
7:40
I21 m21
2
I 1
7:41
J11
21
J 4 0
J31
0
J22
J32
3
0
0 5;
J33
7:42
232
where
I1SC
SC
3 dSC
l
21 I2
3
J11 m21
;
J
m
;
J
;
22
33
1 SC
2 SC
I1
I2
l3 dSC
3
I1SC 1
3 dSC
l
3
J31 m21
;
1 G1 SC G1
I1 I1
I l dSC
1
3 3
3 3
I2SC 1
3 dSC
l
3
J32 m21
:
2 G2 SC
I2 I2
I G2 l dSC
From the obtained expressions, the procedure of regime comparison of two
circuits follows. For this purpose, we consider an example.
Example We use the date of Example 5 for Sect. 6.3. The element parameters have
the following values:
V0 5 ; y0N 2:5 ; yN 0:625:
We have for the rst circuit.
The characteristic values of currents and conductivities
SC
SC
I1G1 3:125; yG1
1N 3:125; y1N 6:25; I1 1:785;
SC
SC
I2G2 6:25; yG2
2N 6:25; y2N 4:166; I2 5:357:
y12N 2:777:
m12 0:154:
l3 dSC
3 0:428:
233
M11
M31
M32
M33
0:358
0:836;
0:428
Currents (7.34)
M11 I11
M32 I21 M33
1:293 1
1:293
1:4528;
1:194:
M31 I11
234
I22 2;
y22N 7:333:
1:179
0:916
0:3839 0:2749 1:396:
1:179 4:25 0:916 4:25
Similarly, we get
7:333
2:0475
m21
2 7:333 7 2:0475 7 0:5116 0:2263 2:261:
1:5;
1:45 0:0935 1:194 0:18 1 1:35
I22 1:194 2:261 2:
1:35
I12
Matrix (7.42)
2
1:8073
J 4 0
0:1574
21
0
2:4089
0:2103
3
0
0 5:
0:8368
2 by (7.41)
We check the currents of point M
1:8073 1
1:8073
1:5;
0:1574 1 0:2103 1 0:836 1:2045
I22 2:4089 2:
1:2045
I12
References
235
References
1. Penin, A.: Normalized representation of the equations of active multiport networks on the basis
of projective geometry. Moldavian J. Phys. Sci. 10(34), 350357 (2011). http://sfm.asm.md/
moldphys/2011/vol10/n3-4/index.html Accessed 30 Nov 2014
2. Penin, A.: Recalculating the load currents of an active multiport with variable parameters on the
basis of projective geometry. Electrichestvo 10, 6673 (2012)
3. Penin, A.: Recalculation of the loads current of active multiport networks on the basis of
projective geometry. J. Circ. Syst. Comput. 22(05), 1350031, 13 p. (2013). doi: 10.1142/
S021812661350031X http://www.worldscientic.com/doi/abs/10.1142/S021812661350031X
Accessed 30 Nov 2014
4. Penin, A.: Comparison of regimes of active two-port networks with stabilization of load
voltages. Int. J. Electron. Commun. Electr. Eng. 3(6), 118 (2013). https://sites.google.com/site/
ijeceejournal/
Chapter 8
8.1
We use the results of Sect. 6.1.1. Let us consider a four-port circuit with changeable
load voltage sources in Fig. 8.1 and give necessary relationships I3 V1 ; V2 ; I4 V1 ; V2
for this network.
Taking into account the specied directions of currents, this network is described
by the following system of Y parameters equations
2
I1
Y11
6I 7 6
6 2 7 6 Y12
6 74
Y13
4 I3 5
Y14
I
Y12
Y22
Y23
Y24
Y13
Y23
Y33
Y34
3 2 V1 3
Y14
6V 7
Y24 7
27
76
6 7 Y V;
Y34 5 4 V3 5
Y44
V
8:1
I3
I4
Y13
Y14
8:2
I3SC;SC
I4SC;SC
Y33
Y34
" #
V3
Y34
:
Y44
V4
8:3
237
238
V3
I1
V1
Y
+
I2
I4
+
_
V2
V4
+
V1
V2
1 Y
34 24
DY Y14
#
" SC;SC
I3
I3
Y23
SC;SC
;
Y13
I4
I4
8:4
Y24
D34
I3 I3SC;SC Y V1 ;
Y23
Y23
8:5
I4 I4SC;SC
Y14
D34
I3 I3SC;SC Y V2 :
Y13
Y13
8:6
Y14
I3 I3SC;SC :
Y13
8:7
This straight line passes through the point SC; SC or the point 0V . The value
Y14 =Y13 corresponds to a slope angle. The currents I3SC;SC ; I4SC;SC determine the point
SC; SC.
Similarly, the equation of the axis V2 is dened by the expression (8.5) as V1 0
239
I4
1,SC
I3
SC,SC
I4
(SC, SC)
0V
1,SC
I4
V1
OC,SC
I4
V1
V1
SC,1
I4
I4
OC
V2
1
1
I3
SC,OC
I4
OC
V2
OC,OC
I4
(OC, OC)
V2
OC,SC
SC,SC
I3
0I
I4 I4SC;SC
OC,OC
I3
Y24
I3 I3SC;SC :
Y23
I3
SC,OC
I3
I3
SC,1
I3
8:8
The open circuit regime of both loads corresponds to the point OC; OC or to
point 0I . Let us determine the voltages V1OC ; V2OC . From (8.1), we get
"
I1
I2
Y11
Y12
8:9
I1SC;SC
" #
V3
Y14
:
Y24
V4
8:10
#
" SC;SC
I1
I1
Y12
SC;SC
;
Y11
I2
I2
8:11
Y13
Y23
I2SC;SC
The inverse expression of (8.9)
"
V1
V2
1 Y22
12
DY Y12
2
where determinant D12
Y Y11 Y22 Y12 .
Using (8.11), we get, similarly to (6.7) and (6.8), that
V1OC
8:12
240
V2OC
8:13
8:14
8:15
We must map the point OC; OC onto the axes V1 ; V2 by parallel lines to these
axes. Then, the coordinates or components I3OC;SC ; I4OC;SC will correspond to the
point V1OC on the axis V1 . Using (8.12) and (8.2) as V2 0, we get
I3OC;SC Y13 V1OC I3SC;SC ;
8:16
8:17
In turn, the coordinates I3SC;OC ; I4SC;OC correspond to the point V2OC on the axis V2 .
Using (8.13) and (8.2) as V1 0, we obtain
I3SC;OC Y23 V2OC I3SC;SC ;
8:18
8:19
Let an initial regime be given by values V11 ; V21 or by a point M 1 . Then, the
coordinates I31;SC ; I41;SC dene the point V11 on the axis V1 . Using (8.2) as V2 0, we
have
I31;SC Y13 V11 I3SC;SC ;
8:20
8:21
In turn, the coordinates I3SC;1 , I4SC;1 dene the point V21 on the axis V2 . Using (8.2)
as V1 0, we have
I1SC;1 Y23 V21 I3SC;SC ;
8:22
8:23
Let us introduce normalized coordinates for the point M 1 . The point OC; OC is
a length scale for the system of coordinates V1 0V V2 .
Then, similarly to (6.17)(6.20), the normalized coordinates have the view
n11
241
V11 0V
V11
I31;SC I3SC;SC
;
V1OC 0V V1OC I3OC;SC I3SC;SC
8:24
V21
I4SC;1 I4SC;SC
:
V2OC I4SC;OC I4SC;SC
8:25
n12
N21
V11
I31;SC I3SC;SC
;
V11 I3OC;SC I31;SC
8:26
V21
I4SC;1 I4SC;SC
SC;OC
:
1
V2 I4
I4SC;1
8:27
V1OC
V2OC
The presented approach will be used for transmission of two measuring signals.
Example 1 We consider the following four-port circuit with given parameters in
Fig. 8.3.
System of Eq. (8.1)
2
I1
0:6813
6I 7 6
6 2 7 6 0:1393
6 74
0:147
4 I3 5
0:0464
I
0:1393
0:7727
0:087
0:159
0:147
0:087
0:3247
0:029
3 2 V1 3
0:0464
6V 7
0:159 7
27
76
6 7:
0:029 5 4 V3 5
0:2803
V
4
Currents (8.3)
"
I3SC;SC
0:3247
0:029
I4SC;SC
y3K
12
I3
V4
10
0.25
I4
I2
V2
0.333
V1
y10
y2L
0.666
0.2
y4L
yKL
yK0
I1
y1K
0.416
V3
#
" # "
12
3:6064
0:029
:
0:2803
10
2:455
yL0
y20
0.2
0.25
Fig. 8.3 Four-port represents two two-port networks connected by conductivity yKL
242
I3
I4
0:147
0:0464
#
" # "
V1
3:6064
0:087
:
0:159
2:455
V2
V1
V2
1 0:159
34
DY 0:0464
#
"
3:6064 I3
0:087
;
0:147
2:455 I4
D34
Y 0:147 0:159 0:087 0:0464 0:01934:
Equation (8.7) of the axis V1
I4 2:455 0:3156 I3 3:6064:
Equation (8.8) of the axis V2
I4 2:455 1:8275 I3 3:6064:
The obtained system of coordinates is shown in Fig. 8.4.
SC currents (8.10) of both loads
"
I1SC;SC
I2SC;SC
0:147
0:087
#
" # "
12
2:228
0:0464
:
0:159
10
2:634
0:507
0:507
0:6813 0:7727 0:13932 0:507:
V1OC
V2OC
D12
Y
I4
0V
2.5
OC
V1 =4.119
V1=2.202
V2=2.663
V1
1.5
M
(OC, OC)
1
OC
V2 =4.1514
V2
1
2
2.5
3.5
I3
243
V21 2:663:
Currents (8.2)
"
I31
I41
0:147
0:0464
# "
# "
#
"
2:202
3:6064
3:051
0:087
:
0:159
2:663
2:455
1:93
0:5346;
4:119
3:0 3:6064
2:663
2:032 2:455
n12
0:6415:
4:1514 1:795 2:455
n11
244
1:1446;
4:119 2:202
3:0 3:2827
2:663
2:032 2:455
N21
1:789:
4:1514 2:663 1:795 2:032
N11
8.2
Now, we consider a four-port circuit in Fig. 8.5. Let us give necessary relationships
between input currents and load conductivities [1]. This network is described by
system of Eqs. (8.1). Also, we have that
V1 I1 =YL1 ;
8.2.1
V2 I2 =YL2 :
Output of a Four-Port
I3
I1
V3
V1
I4
V4
I2
YL1
V2
YL2
245
YL2
I2
G2
G2
YL2
G2
I2
YL2=
SC
SC
SC
I2
SC
YL2
1
I2
1
SC
G1
G1
SC
I1
I1
I1
1
YL1
I1
G1
YL1=
YL1
YL1
m 1= 1
m 1=
m1
m1
1
YL1
;
1 Y G1
YL1
L1
m12
1
YL2
:
1 Y G2
YL2
L2
8:28
qn1
;
qn3
m2
qn2
;
qn3
8:29
qn1
1
7
6
7
6
I SC
7 61
6
6 qn2 7 6 0
7 4
6
7
6
1
5
4
I1G1 l3 dSC
3
qn3
where l3 dSC
3
SC
I1
I1G1
I2SC
I2G2
0
1
I2SC
1
I2G2 l3 dSC
3
1 .
I1
I1
6 7
6 7
6 7
6 7
6 7
7 6 7
0 7 6 I2 7 C 6 I2 7;
6 7
5 6 7
6 7
6 7
1
SC
5
4
4 5
l3 d3
1
1
0
8:30
246
I1
I1G1 l3 dSC
3
1
I
I1SC 1
I G2 lI2 dSC
3 3
2
l 1dSC
I1
I1G1 l3 dSC
3
1
I
I2SC 2
I G2 lI2 dSC
3 3
2
l 1dSC
;
8:31
3 3
3 3
C
6
7 4 SC
5 6 7
4 n2 5 ;
I1
I2SC
4 5
4
5
SC
l3 d3
n3
I1G1
I2G2
q1
n3
8:32
qI1
I1SC m1
I SC
;
I2SC
SC
q1
1
m
m
l
d
G1
G2
1
2
3
3
I
I
1
I2 I SC
1
I1G1
8.2.2
m1
8:33
I2SC m2
I2SC
I2G2
m2 l3 dSC
3
Input of a Four-Port
Let us consider the input currents I3 ; I4 of our four-port. We may superpose the
system of coordinates I3 0 I4 with the system of coordinates I1 0 I2 in Fig. 8.7.
Then, any point with coordinates I1 ; I2 corresponds to a point with coordinates
I3 ; I4 . In terms of geometry, a projective transformation takes place which
transfers points of the plane I1 ; I2 into points of the plane I3 ; I4 . Therefore, the
reference triangle G1 0 G2 , point SC, and running regime point M 1 correspond to
1
the triangle G1 0 G2 , point SC, and point M , as it is shown by arrows in Fig. 8.7.
Next, the axes of currents I1 ; I2 correspond to the axes I 1 ; I 2 . Also, two bunches
of the straight lines I1 ; I2 ; YL1 0, I1 ; I2 ; YL2 0 correspond to two bunches of
the lines I3 ; I4 ; YL1 0, I3 ; I4 ; YL2 0 with centers in the points G2 ; G1 . Thus,
1
the point M is set by other currents I31 ; I41 . Also, this point is dened by projective
non-uniform and homogeneous coordinates which are set by the reference triangle
G1 0 G2 and a unit point SC.
1
247
I2
I4
G2
I2
G2
YL2
YL2
I1
G1
YL1
0
0
I3
G1
YL1
I4
I1
I1
G1
I4
SC
SC
I4
SC
I4
G1
OC
I4
OC
I3
SC
1
I3 I 3
G1
I3
I3
1
1
loads YL1
; YL2
. Therefore, this property gives required invariant relations between
input and output currents.
1
For nding of the point M projective coordinates, it is necessary to obtain
equations of sides of reference triangle. The normalized equation of the side 0 G1
or axis I 1 in Fig. 8.8 looks like
I4OC
I4
k 1 I3
1 0;
k 1 I3OC I4OC k1 I3OC
k 1 tga1
I4G1 I4OC
;
I3G1 I3OC
l2 d2
I4OC
248
l 2 d2
I4SC
k1 I3SC
1:
I4OC k 1 I3OC I4OC k 1 I3OC
I4G2 I4OC
:
I3G2 I3OC
:
k2 I3OC I4OC
k2 I3OC I4OC
1
l1 d1
k 2 I3OC
SC
l 1 d1
I4SC
k2 I3SC
1:
k2 I3OC I4OC k2 I3OC I4OC
I4G1
I4
k 1 I3
G1
1 0;
G1
k 1 I3
I4 k1 I3G1
k1
I4G2 I4G1
:
I3G1 I3G2
l 3 d3
I4G1
249
SC
l3 d3
I4SC
k1 I3SC
1:
I4G1 k 1 I3G1 I4G1 k 1 I3G1
qn11
d1
SC
d1
k2
k2 I3OC
1
k 2 I3OC
I1
SC 4
OC
I4 l1 d1
qn12
d2
SC
d2
I4OC
k1
d3
SC
I1
SC 3
k1 I3OC l2 d2
SC
l1 d1
I4OC
I1
SC 4
OC
k1 I3 l2 d2
qn13
I1
SC 3
I4OC l1 d1
1
l2 d2
k1
I1
SC 3
I4G1 k1 I3G1 l3 d3
1
1
I1
:
SC 4
SC
G1
G1
I4 k 1 I3 l3 d3
l 3 d3
SC
d3
qn1
I3
C11
7
6 7
6
7
6 7 6
6
7
6 7 6
6
6 qn2 7 C 6 I4 7 6 C21
7
6 7 6
6
7
6 7 4
6
5
4 5
4
C 31
qn3
1
C 11
k2
C 21
k1
C 31
k1
1
SC
l1 d 1
1SC
l2 d 2
1SC
l3 d 3
I3
6 7
7 6 7
7 6 7
7 6 I4 7:
7 6 7
5 6 7
4 5
1
k2
k2 I3OC
SC
I4OC l1 d1
k1
SC
I4G1 k1 I3G1 l3 d3
;
:
C 21
k1
SC
I4OC k1 I3OC l2 d2
8:34
250
From here, the non-uniform coordinates have the form similar to (8.31)
m1
m2
C 11 I3
C 31 I3
C31
I
k1 4
C 21 I3
C 31 I3
C11
I
k2 4
C21
I
k1 4
C 31
I
k1 4
1
SC
l3 d 3
1
SC
l1 d 1
1
SC
l2 d 2
1
SC
l3 d3
;
8:35
:
The obtained expressions have the general appearance in comparison with (8.31)
because of the non-orthogonal coordinates I 1 0I 2 .
In practice, characteristic values of input current, output currents (as vertexes of
coordinate triangles), and loads are precalculated or preprogrammed by the calculation or testing of four-port network.
Further, using running values of input currents, we nd or, more precisely,
restore values of non-uniform coordinates (8.35) and values of given load conductivities according to the expressions YL1 m1 ; YL2 m2 . These expressions are
inverse to expression (8.28).
Such a formulated algorithm represents a practical interest for transfer of two
sensing signals via an unstable four-port network or a three-wire line; it is analogous to a signal transmission via a two-port network of Chap. 4.
The inverse transformation of (8.34)
2
qI3
n1
7
7
6
6
7
7
6
6
7
7
6
6
6 qI4 7 C1 6 n2 7
7
7
6
6
7
7
6
6
5
5
4
4
q1
n1
1
C11
6 1
6 C11
4
1
C11
I4G1
I3G1
1
I3G1
1
C 12
1
C 12
1
C 12
I4G2
I3G2
1
I3G2
1
C13
1
C13
1
C13
1
1
k 2 I3OC I4OC
SC
I G1 l1 d1 ;
OC
G1 3
k 1 k2 I3 I3
C 12
1
C 13
n1
6 7
6 7
7 6 7
7 6 n2 7 :
5 6 7
6 7
1
4 5
OC
I3
n3
I4OC
I3OC
k 1 I3OC I4OC
SC
I G2 l2 d2 ;
k 1 k2 I3OC I3G2 3
k 1 I3G1 I4G1
SC
I OC l3 d3 :
OC
G1 3
k 1 k 1 I3 I3
8:36
251
I3
I4
1
1
C11 m1 C12 m2 C 13
C 11
1 1
I3G1
m1 C12
1 1
I3G2
m2 C 13
1 I4G1
I3G1
1 1
C 11 I G1
3
m1 C12
1 I4G2
I3G2
1 1
m1 C12 I G2
3
m2 C 13
C 11
1 1
I3OC
1 I4OC
I3OC
1 1
m2 C 13 I OC
3
;
8:37
:
The obtained expressions have the general view in comparison with (8.33)
because of non-orthogonal coordinates.
Convenience of expressions (8.35) and (8.37) consists in their identical form; we
replace load conductivities by their non-uniform projective coordinates and currents
are already non-uniform coordinates too.
8.2.3
2
2
Let a subsequent regime be corresponded to a point M 2 with loads YL1
; YL2
,
2
2
2 2
2 2
non-uniform coordinates m1 ; m2 , output currents I1 ; I2 , and input currents I3 ; I4 .
At rst, using the result of Sect. 7.1.1, we rewrite subsequent currents I12 ; I22 (7.2)
qI12
I11
I11
6 27
6 7
6 7
1
4 qI2 5 J21 4 I21 5 C21 C 4 I21 5:
q1
The matrix
2
21
J11
J21 4 0
21
J31
0
21
J22
21
J32
3 2 m21
1
0
60
5
0
4 21
m1 1
1
I G1
1
0
m21
2
m21
2 1
I2G2
m21
1
21
4
M 0
0
0
m21
2
0
3
0
0 5:
1
3
0
07
5:
1
8:38
252
2
1
m21
2 m2 m2 :
8:39
These changes of the non-uniform coordinates are also true for the input currents. Therefore, it is possible at once to obtain similar relationships for the
recalculation of the input currents.
2
qI32
I31
I31
21 6
7
6 27
6 7
1
4 qI4 5 C M21 C 4 I41 5 J 4 I41 5:
q1
8:40
21
We calculate the matrix J . This matrix has the general view in comparison
with (8.38)
2
21
3
21
J 13
21 7
J 23 5:
21
J 33
21
J
21
6 11
J 4 J 21
21
21
J 31
J 12
21
J 22
21
J 32
8:41
1
qI12
m21
1 I1
m21 1
;
21 1
m
q1
1
1
1
2
G1 I
G2 I 1
1
I1
I22
I2
1
qI 2
m21
2 I2
2 m21 1
:
21 1
m
1
1
q1
1
2
G1 I
G2 I 1
1
I1
I2
8:42
21
21
21
21
21
21
I32
qI 2 J I31 J 12 I41 J 13
3 11
;
q1 J 21 I 1 J 21 I 1 J 21
31 3
32 4
33
I42
qI 2 J I31 J 22 I41 J 23
4 21
:
q1 J 21 I 1 J 21 I 1 J 21
31 3
32 4
33
8:43
We note that the denominators of expressions (8.42) and (8.43) are equal to each
other.
8.2.4
Let us consider cascaded four-port networks in Fig. 8.9. The rst four-port is given
by a matrix Y 36 of Y parameters and the second four-port corresponds to a matrix
Y 14 .
253
I3
I5
I1
V1
V5
I6
3-6
I4
1-4
I2
YL1
V2
V6
+
YL2
I2, I4, I6
G2
G2
G2
0
0
0
G1
I1, I3, I5
G1
G1
254
8.2.5
Examples of Calculation
0:147
0:0464 0:159
0:0464
0:087
0:3247
0:159 7
7
7:
0:029 5
0:029
0:2803
I2SC 2:634:
I1
I2
2
3
" #
I1SC;SC
V1
Y12
5
4
Y22
V2
I2SC;SC
#
" # "
V1
2:228
0:6813 0:1393
:
0:1393
0:7727
V2
2:634
Y11
Y12
I
I2SC;SS
0:1393
Y12 1
1 SC;SS
2:228
I G2
G2
16; YL2
I1
L2
0:9375:
G2
Y12
0:1393
VL2
I1G1
V1G1
I2G2
V2G2
y3K
0.416
V3
12
I3
V4
I4
0.666
y10
0.25
y2L
0.333
10
0.2
y4L
yKL
yK0
I1
y1K
V1
I2
V2
yL0
y20
0.2
0.25
YL2
YL1
255
I11 1:101;
1
0:333;
YL2
V11 2:202;
I21 0:8868;
V21 2:663:
1
YL1
1
YL1
0:5
0:3846;
G1
0:5 0:8
YL1
1
YL2
0:333
0:2622:
G2
1
0:333
0:9375
YL2 YL2
Matrix (8.30)
2
1
2:228
C 4 0
1
15:110:6769
1
2:634
1
150:6769
0
0
3
5;
1
0:6769
I1SC
I SC
2G2 1
G1
I1
I2
2:228 2:634
1 0:6769:
15:11
15
SC
2:228
2:634
I1
I2
1:101
0:8868
1
qn13
qn11
qn1 0:4942
0:3846;
qn3 1:2823
m12
qn2 0:3366
0:2622:
qn3 1:2823
2:228
C1 4 0
0:1474
0
2:634
0:1757
3
0
5:
0
0:6769
256
Currents (8.33)
2:228 0:3846
0:86
1:101;
0:1474 0:3846 0:1757 0:2622 0:6769 0:7797
2:634 0:2622
I21
0:8868:
0:7797
I11
I4SC;SC 2:455:
Using (8.2), we may nd the bunch centers G1 ; G2 . Then, we get the following
systems of equations:
2
3
" G1 #
" G1 #
I3
I3SC;SC
V1
Y13 Y23
5;
4
SC;SC
Y14 Y24
0
I4G1
I4
" G2 #
"
# 2 SC;SC 3
0
I3
I3
Y13 Y23
5:
4
Y14 Y24
V2G2
I4G2
I4SC;SC
Therefore,
I3G1 0:147 18:9 3:6064 6:4423;
I4G1 0:0464 18:9 2:455 3:333;
I3G2 0:087 16 3:6064 5;
I4G2 0:159 16 2:455 5:
The currents, corresponding to the open circuit,
I3OC 2:6397;
I4OC 1:6038:
I41 1:93:
I4OC
I4
k 1 I3
I4
I3
1 0:
OC
1
OC
OC
0:3835 0:8305
k 1 I3
I4 k 1 I3
0:3835
0:8305
1
l2 d2
l2 d2
2:455
3:606
1 1:0576:
0:3835 0:8305
2:1961
1:5258
1
l1 d1
l1 d1
2:455
3:606
1 0:2458:
2:1961 1:5258
1 0;
1
11 9:166
I4G1 k1 I3G1 I4G1 k1 I3G1
k1
257
258
1 0:4918;
11
9:166
s
2
2
1
1
0:142:
l3
11
9:166
1
l3 d3
l3 d3
2:455 3:606
1 0:3835:
11
9:166
qn11
d1
SC
d1
0:1216
0:4942;
0:2458
0:3566
0:3366;
1:057
0:4918
1:2823:
0:3835
qn12
d2
SC
d2
1
qn13
d3
SC
d3
Matrix (8.34)
2
1
0:24581:5258
6
1
4 1:0570:8305
1
0:38359:166
1
0:24582:1961
1
1:0570:3834
1
0:383511
2:666
1:852
6
4 1:1383 2:4653
0:2844 0:237
1
0:2458
7
1
1:057 5
1
0:3835
3
4:067
7
0:9454 5:
2:607
0:3846:
0:2844 3:051 0:237 1:93 2:607 1:282
1:1383 3:051 2:4653 1:93 0:9454 0:339
m12
0:2622:
0:2844 3:051 0:237 1:93 2:607
1:282
m11
259
1
C
0:9422
4 0:4915
0:1474
0:8789
0:8791
0:1757
3
1:7868
1:0847 5:
0:6768
3:051;
0:1474 0:3846 0:1757 0:2622 0:6768 0:7796
0:4915 0:3846 0:8791 0:2622 1:0847 1:5042
I41
1:93:
0:7796
0:7796
I31
I12 1:459;
V12 1:459;
2
1;
YL2
I22 1:602;
V22 1:602:
1
0:555;
1 0:8
m22 0:516:
Currents (8.2)
I32 3:253;
I42 2:132:
Matrix (8.38)
2
1:442
J21 4 0
0:0292
0
1:968
0:0645
3
0
0 5:
1
1:442 1:101
1:5876
1:459:
0:0292 1:101 0:0645 0:8868 1 1:0893
260
1:3225
21
J 22
0:2982
3
2:503
21
J 23 5:
0:5728
8.3
3:253:
0:01927 3:051 0:2982 1:93 0:5728 1:0893
8.3.1
:
G2
YL2 YL2
OC
SC G1
G1
YL1 YL1
YL1 0 YL1 1YL1
m1 YL1
Since these cross-ratios m1 ; m2 are equal to the transferable signals VS1 ; VS2 , then
the information conductivities YL1 ; YL2 are expressed by the formulas
YL1
G1
YL1
VS1
;
VS1 1
YL2
G2
YL2
VS2
:
VS2 1
8:44
261
5
G1
YL1
YL1
YL1 =
SC
G1
I1
V0
Line
V1
1
OC
YL1 =0
I2
F
VS1
Gen
I3
I4
5
F
VS1
-1
G2
YL2
YL2 =
YL2
SC
VS2
G2
V2
1
OC
YL2 =0
VS2
Table 8.1 Correspondence between output conductivities, input currents, and output currents
Number of set
Load conductivities
Output currents
Input
currents
YL1 0
2
3
YL1
YL1 1
YL2 0
I1 0
I2 0
YL2
YL2 1
I1
I2
I3OC
I3
I4OC
I4
YL1
G2
YL2
I1SC
I2SC
I3SC
I4SC
I2G2
I2G1
I3G2
I3G1
I4G2
G1
YL1
YL2
I1G2
I1G1
I4G1
VS1 m1
VS2 m2
C 11 I3
C 11
I
k2 4
1
SC
d1
C 31 I3
C 31
I
k1 4
1
SC
d3
C 21 I3
C 21
I
k1 4
1
SC
d2
C 31 I3
C 31
I
k1 4
1
SC
d3
;
8:45
:
262
So, it is possible to separate (or restore) two signals by only the input currents
(ve pairs of current values) of line. We note also that multiplicative and additive
errors of the measurement of input conductivities are mutually reduced in expression (8.45). Therefore, such a communication line is subject to action of electromagnetic noises to a lesser degree.
8.3.2
Afne ratios (8.24) and (8.25) or (8.26) and (8.27) are accepted as transferable
analog signals VS1 ; VS2 . The corresponding appliance is shown in Fig. 8.13 [2].
Similarly, for a short time, four pairs of voltage sources are connected to the output
terminals of the line by turns. This connection is realized by the contact pairs 1, 2,
3, 4 of two multichannel switches with a pulse generator Gen. The sets of four pairs
of input and output currents of the line correspond to these pairs of voltages, as it is
shown in Table 8.2. The voltages V1SC ; V2SC are minimum values; in particular case,
these values equal zero. The voltages V1OC ; V2OC are maximum values; these values
are greater than the information values V1 ; V2 .
So, the normalized coordinates of the point MV1 ; V2 are set by afne ratios
(8.24) and (8.25)
n1
V1 V1SC
;
V1OC V1SC
n2
V2 V2SC
:
V2OC V2SC
4
3
V1
I1
V0
Line
F
SC
I2
V1
OC
V1
VS1
Gen
I3
I4
4
F
VS1
-1
VS2
2
1
F
SC
V2
OC
V2
VS2
263
Number of set
Load voltages
Input currents
V1SC
V2SC
I3SC;SC
I4SC;SC
V1OC
V2SC
I3OC;SC
I4OC;SC
V1SC
V2OC
I3SC;OC
V1
V2
I3
I4SC;OC
I4
Since these afne ratios n1 ; n2 are equal to transferable signals VS1 ; VS2 , then the
information voltages V1 ; V2 are expressed by the formulas
V1 V1SC V1OC V1SC VS1 ;
8:46
I31;SC I3SC;SC
I3OC;SC
I3SC;SC
VS2
I41;SC I4SC;SC
I4OC;SC I4SC;SC
8:47
For other afne ratios (8.26) and (8.27), we get similarly to (8.46) that
V1
V2
8:48
The function unit F works according to these expressions. Also, the inverse
function unit F 1 calculates the transmitted signals VS1 ; VS2 by formulas
VS1
I31;SC I3SC;SC
;
1;SC
I3OC;SC I3
VS2
I4SC;1 I4SC;SC
I4SC;OC I4SC;1
8:49
These two variants of use of afne ratios allow to separate (or restore) two
signals by only the input currents (four pairs of current values) of line. We note also
that multiplicative and additive errors of measurement of input conductivity values
are mutually reduced in expressions (8.47) and (8.49).
8.4
Let us continue with the matter we began in Sect. 6.2.2. We consider the currents
I4 ; I5 ; I6 of the voltage sources V4 ; V5 ; V6 for the balanced network in Fig. 6.18.
Let us superpose the coordinate system I4 ; I5 ; I6 with the coordinate system
I1 ; I2 ; I3 in Fig. 8.14. Then, any point with coordinates I1 ; I2 ; I3 corresponds to a
264
I3
I6
G3
I3
G3
G2
I 2 , I5
G2
SC
I2
SC
G1
I1
G1
I 1 , I4
I6G1 I6SC :
8:50
8:51
Output
Points
I1
I2
I3
Input
Points
I4
I5
I6
I4OC
I5OC
I6OC
G1
I1G1
0
G1
I4G1
I5G1
I6G1
I4G2
I4G3
I4SC
I5G2
I5G3
I5SC
I6G2
G2
I2G2
G2
I3G3
I3SC
G3
G3
SC
I1SC
I2SC
SC
I6G3
I6SC
265
8:52
8:53
SC
SC
SC
SC
I6OC
I6G1
I6G2
I5OC
I5G1
I5G2
1
1 ;
1
1
1 ;
1
OC OC
I
I4
1
6
B12 I6G1 I4G1 1 ;
I G2 I G2 1
6
4
OC OC OC
I
I5
I6
4
G1
G1
D12 I4
I5
I6G1 :
I G2 I G2 I G2
4
6
5
y4a
y5b V5 ;
yab
I4G1 I4G2
y4a
y5b V4 :
yab
8:54
266
A23
C23
OC
I
5
I5G2
I G3
5
OC
I
4
I4G2
I G3
4
I6OC
I6G2
I6G3
I5OC
I5G2
I5G3
1
1 ;
1
1
1 ;
1
OC OC
I
I4
1
6
B23 I6G2 I4G2 1 ;
I G3 I G3 1
6
4
OC OC OC
I
I5
I6
4
G2
D23 I4
I5G2 I6G2 :
I G3 I G3 I G3
4
6
5
8:55
A13
C13
OC
I
5
I5G1
I G3
5
OC
I
4
I4G1
I G3
4
I6OC
I6G1
I6G3
I5OC
I5G1
I5G3
1
1 ;
1
1
1 ;
1
OC OC
I
I4
1
6
B13 I6G1 I4G1 1 ;
I G3 I G3 1
6
4
OC OC OC
I
I5
I6
4
G1
D13 I4
I5G1 I6G1 :
I G3 I G3 I G3
4
6
5
8:56
8:57
267
If V4 V5 V6 , then A1 B1 C1 .
At last, we give expressions for the distances.
1
SC
SC
SC
SC
qn11
l 1 d1
SC
l 1 d1
A23
I1
SC 4
l 1 d1
qn12
l 2 d2
SC
l 2 d2
A13
I1
SC 4
l 2 d2
qn13
qn14
l 3 d3
SC
l 3 d3
B23
I1
SC 5
l1 d1
C23
I1
SC 6
l 1 d1
D23
SC
l1 d1
B13
I1
SC 5
l2 d2
C13
I1
SC 6
l 2 d2
D13
SC
l2 d2
A12
B12 1
C12 1
D12
I1
I
I
;
SC 4
SC 5
SC 6
SC
l 3 d3
l3 d3
l 3 d3
l3 d3
1
l 4 d4
A1 I41 B1 I51 C1 I61 D1
SC
A1 I4SC B1 I5SC C1 I6SC D1
l 4 d4
A1
I1
SC 4
l 4 d4
B1
I1
SC 5
l4 d4
C1
I1
SC 6
l 4 d4
D1
SC
l4 d4
268
qn11
I41
6 17
6 17
6I 7
6 qn2 7
6
7 C 6 5 7;
6 17
6 17
4 I6 5
4 qn3 5
1
qn4
2 A
23
6 l dSC
6 1 1
6 A
6 13
6
6 l dSC
6
C 6 2 2
6 A12
6
6 l dSC
6 3 3
6
4 A1
SC
l 4 d4
B23
C23
SC
l1 d1
SC
l1 d1
B13
C13
SC
l2 d2
B12
SC
l3 d3
B1
SC
l4 d4
SC
l2 d2
C12
SC
l3 d3
C1
SC
l4 d4
D23 3
SC
l1 d1 7
7
D13 7
7
7
SC 7
l2 d2 7
7:
D12 7
7
SC
l3 d3 7
7
7
D1 5
8:58
SC
l4 d4
m11
;
A1 I41 B1 I51 C1 I61 D1
qn14 l1 dSC
1
m12
;
A1 I41 B1 I51 C1 I61 D1
qn14 l2 dSC
2
SC
8:59
SC
m13
So, we use only the input currents for (8.59). In practice, the characteristic values
are obtained by calculation or testing the multi-port only by manipulations at the
output terminals. It is technically more convenient than determination of all the Y
parameter.
In turn, we obtain or restore the load conductivities
1
YL1
G1 1
YL1
m1
;
1
m1 1
YL2
G2 1
YL2
m2
;
1
m2 1
YL3
G3 1
YL3
m3
:
1
m3 1
8:60
269
Input currents
I4
I5
I6
I4OC 0:1793
I5OC 0:2986
I6OC 0:1382
G1
I4G1 2:6
I5G1 2:0
I6G1 0:6666
G2
I4G2
I4G3
I4SC
3:6
I6G2 0:6666
2:0
I6G3 2:2666
1:501
I6SC 0:5786
G3
SC
1:0
I5G2
I5G3
I5SC
1:0
0:8272
A23
B23
0:2986
3:6
2:0
0:1382
0:6666
2:2666
0:1382
0:6666
2:2666
0:1793
1:0
1:0
1
1 6:1264;
1
1
1 1:3131;
1
270
C23
D23
0:1793
1:0
1:0
0:1793
1:0
1:0
0:2986 1
3:6
1 1:3131;
2:0
1
0:2986 0:1382
3:6
0:6666 0:5253:
2:0
2:2666
A13
B13
C13
D13
0:2986
2:0
2:0
0:1382
0:6666
2:2666
0:1793
2:6
1:0
0:1793
2:6
1:0
0:1382
0:6666
2:2666
0:1793
2:6
1:0
0:2986
2:0
2:0
0:2986
2:0
2:0
1
1 2:7222;
1
1
1 4:7186;
1
1
1 2:7222;
1
0:1382
0:6666 0:5382:
2:2666
A1
2:0
3:6
2:0
0:6666 1
0:6666 1 2:56;
2:2666 1
B1
0:6666
0:6666
2:2666
2:6 1
1 1 2:56;
1 1
C1
D1
271
2:6
1
1
2:6
1
1
2:0
3:6
2:0
2:0
3:6
2:0
1
1 2:56;
1
0:6666
0:6666 13:4825:
2:2666
1
YL2
0:25;
1
YL3
0:1666;
I11 0:4232;
I21 0:5673;
I31 0:335;
I41 0:4921;
I51 0:8654;
I61 0:354:
SC
The distances d1 ; d1
SC
The distances d2 ; d2
1
SC
The distances d3 ; d3
272
SC
The distances d4 ; d4
1
qn11
l 1 d1
SC
l1 d1
0:8883
0:4895
1:8147
qn12
l 2 d2
SC
l2 d2
1:242
0:457
2:7176
qn13
l 3 d3
SC
l3 d3
0:6792
0:5069
1:3399
qn14
l 4 d4
SC
l4 d4
9:101
1:5065
6:0411
2:56 1
2:56 1
2:56 1 13:4825
I
I
I
6:0411 4 6:0411 5 6:0411 6
6:0411
1
1
1
0:4238I4 0:4238I5 0:4238I6 2:2318:
0:7236
1:7363
0:6309
0:4238
1 0 I4 1
0:7236 0:2895
BI C
1:0017 0:198 C
5C
CB
B C:
4:9
0:379 A @ I6 A
0:4238 2:2318
1
273
0:3254;
1:5065
1:0017 0:4921 1:7363 0:8654 1:0017 0:354 0:198
m12
1:5065
0:457
0:3022;
1:5065
0:6309 0:4921 0:6309 0:8654 4:9 0:354 0:379
m13
1:5065
0:5069
0:3367:
1:5065
m11
0:3022 1
0:3284 0:3367
0:1666:
0:3367 1
1
YL1
1
YL2
1
YL3
References
1. Penin, A.: Invariant properties of cascaded six-pole networks. Int. J. Circuit Syst. Signal
Process. 6(5), 305314 (2012)
2. Penin, A., Sidorenko, A.: Method for transmitting signals through the three- wire direct current
line. MD Patent 692, 31 Oct (2013)
3. Penin, A., Sidorenko, A.: Transmission of measuring signals and power supply of remote
sensors. In: Bonca, J., Kruchinin, S. (eds.) Nanotechnology in the Security Systems. NATO
Science for Peace and Security Series C: Environmental Security, pp. 267281. Springer
Science + Business Media, Dordrecht (2014)
Chapter 9
9.1
9.1.1
Let us consider, for example, a familiar active two-port network in Fig. 9.1 with
changeable load conductivities YL1, YL2.
We rewrite the main relationships of Sect. 6.1.3. These active two-port is
described by the system of equations
I1
I2
SC
Y12
V1
I
1SC :
Y22
V2
I2
Y11
Y12
I1
y1N
1.25
y0N
V0
5
2.5
y2N
N
yN
9:1
0.833
0.625
V1
y1
0.25
y2
0.25
I2
YL1
V2
YL2
275
276
where Y parameters
Y11 y1 y1N
Y12 y2N
y1N
;
yR
y21N
;
yR
Y22 y2 y2N
y22N
:
yR
9:2
Expression (9.1) shows that the active two-port represents a passive part, which
SC
is set by conductivities of Y parameters, and two sources of currents ISC
1 , I2 , as it is
illustrated in Fig. 9.2 [1].
We may notice that SC currents are set by parameters Y10, Y20 which depend
practically on all elements of this circuit except y1, y2.
Therefore, at a possible change, for example, of conductivity yN, it is necessary
to make the recalculation of these SC currents what is inconvenient. The conductivity yN can be a part of a possible third load. In turn, the conductivity y1N belongs
more to the rst load, but also influences on the SC current of the second load. And
mutually, conductivity y2N influences on the SC current of the rst load.
All these features determine inconveniences for the evaluation of circuit characteristics, complicates regime calculations. In this sense, the parameters of the
generalized equivalent generator, proposed in Chap. 3, do not depend on a chosen
element of circuit.
9.1.2
We use results of Sect. 3.3. At rst, let us consider the above two-port network
concerning the load YL1. Therefore, this circuit will be an active two-pole with
changeable element YL2 in Fig. 9.3.
The family of load straight lines I1(V1) is presented by expression (3.79)
I2
I1
V1
Y
SC
I1
YL1
V2
SC
I2
YL2
277
I1
V1
YL1
YL2
9:3
Y12 2
:
YL2 Y22
9:4
The family of load straight lines shown in Fig. 9.4 represents a bunch of straight
lines with familiar center G1.
The bunch center coordinates correspond to expressions (3.77), (3.78), (3.80), or
(6.26), (6.27). Therefore, we get
V1G1
Y20
y0N
V0
V0 ;
Y12
y1N
y1
I1G1 y0N 1
V0 ;
y1N
9:5
I1
G1
YL1
YL1
G1
I1
G1
SC
I1
I1
OC
V1
V1
G1
V1
V1
YL2
YL2=
YL2= 0
YL2
Yi1
Fig. 9.4 Family of load straight lines with the parameter YL2
Yi1
278
VL
I1
V1
G1
IL
Yi
YL1
G1
YL1
I1G1
y1 y1N :
V1G1
9:6
9:7
G1
G1
Thus, the values IG1
are scales for corresponding values.
1 , YL1 , and V1
Expressions (9.3) and (9.7) determine a generalized equivalent generator in Fig. 9.5.
We remind that the value Y0i1 = YG1
L1 is the characteristic value for the conductivity Yi. In this case, the conductivity Yi voltage is equal and opposite to the
source VG1
1 voltage. Therefore, I1 = 0, V1 = 0 for all conductivity YL1 values.
Second, let us consider now the above two-port network concerning the load
YL2. This circuit will be an active two-pole with changeable element YL1 in Fig. 9.6
too.
Similar relationships are obtained for the second load YL2
y0N
V0 ;
y2N
Yi2 Y22
y2
I2G2 y0N 1
V0 ;
y2N
G2
YL2
y2 y2N :
I2
V2
A
YL1
YL2
Y12 2
:
YL1 Y11
9:8
9:9
G1
V1
279
G2
V2
G2
G1
I2
I1
Yi1
YL1
Yi2
YL2
Fig. 9.7 Formal equivalent generator as the association of the two active two-poles
9.1.3
It is possible also to obtain one more scheme of the equivalent generator. This
principal variant is demonstrated by Fig. 9.8 [2].
To prove that, we consider expression (6.43) for the rst input conductivity
Y11
I1G1 I1SC
J1
:
U1
V1G1
This expression denes the input conductivity for the passive part of two-port
network at the short circuit for the rst load and short circuit for the second pair of
terminals in Fig. 9.9a.
The similar relationship is for the second input conductivity
Y22
I2G2 I2SC
J2
:
U2
V2G2
G1
V1 /2
G2
U1
V2 /2
U2
V1
V2
Y
I1
YL1
J1
G1
I1 /2
I2
J2
G2
I2 /2
YL2
Fig. 9.8 Electrical scheme of the generalized equivalent generator of the active two-port
280
(a) V G1
U1
U2=0
Y
SC
I1
G1
J1
I1
(b)
U1=0
Y11
G2
V2
U2
Y
J2
Y22
G2
I2
SC
I2
Analogously, we have the input conductivity for the passive part of two-port
network at the short circuit for the second load and short circuit for the rst pair of
terminals in Fig. 9.9b.
The association of these two schemes leads to the above scheme in Fig. 9.8.
Taking into account the superposition theorem, the values of all constituent sources
of current and voltage are decreased twice. The calculation proves such a scheme
for equivalent generator.
Next, we may obtain the system of equations which describes the proposed
equivalent generator. Using the specied designations in Fig. 9.8, it is possible to
write down
V1G1
V1 ;
2
G2
V
U2 2 V2 ;
2
U1
I1G1
I1 ;
2
G2
I
J2 2 I 2 :
2
J1
9:10
The following relationships take place for the passive part of two-port network
J1 Y11 U1 Y12 U2
J2 Y12 U1 Y22 U2 :
9:11
:
2
2
2
9:12
281
Let us compare this form to expression (9.1). Then, the term in the brackets is
equal to the SC current ISC
1 . In addition, the structure of this short circuit current for
(9.12) demonstrates its components.
The component Y11VG1
1 /2 corresponds to the own current of the two-port network and depends on its parameters.
The component IG1
1 /2 is dened by the constituent current source.
The component Y12VG2
2 /2 corresponds to a mutual current on account of the
constituent voltage source for the second pair of terminals.
The expression for the current I2 is obtained similarly to (9.12)
V2G2
V1G1 I2G2
Y12
9:13
15
I1 6:521;
2
16:25
I2 7:3:
J2
2
J1
282
9.2
Let us use the above approach for an active three-port network with changeable
load conductivities in Fig. 9.10. We use now the results of Sect. 6.2. This network
is described by the following system of equations
2
3 2
Y11
I1
4 I2 5 4 Y12
I3
Y13
Y12
Y22
Y23
3 2 3 2 SC 3
V1
Y13
I1
Y23 5 4 V2 5 4 I2SC 5:
Y33
V3
I3SC
9:14
Taking into account the voltages V1 = I1/YL1, V2 = I2/YL2, and V3 = I2/YL3, we get
three bunches of planes. The intersection of planes of one bunch among themselves
denes a bunch axis.
The point of intersection of this bunch axis with the axis I1 is dened by (6.47)
y1
I1G1 V0 y0N 1
;
y1N
V1G1
y0N
V0 ;
y1N
G1
YL1
y1N y1 :
9:15
This point gives the center G1. Similarly, we get the center G2 by (6.45)
y2
I2G2 V0 y0N 1
;
y2N
V2G2
y0N
V0 ;
y2N
G2
YL2
y2N y2 :
9:16
I1
y1N
V1
y1
y0N
I2
y2N
YL1
V2
V0
y2
YL2
I3
y3N
V3
y3
YL3
283
V3G3
y0N
V0 ;
y3N
G3
YL3
y3N y3 :
9:17
Using these parameters of the centers, we get form (6.55) of system (9.14)
I1
V1
I1SC
I SC V2
I SC V3
I SC
1
1G1 G2 1G1 G3 1G1 ;
G1
G1
G1
I1
V1
I1
I1 V 2
I1 V3
I1
SC
SC
SC
I2
I V1
V2
I
I V3
I SC
2G2 G1 G2 1 2G2 2G2 G3 2G2 ;
G2
I2
I2 V 1
V2
I2
I2 V3
I2
SC
SC
SC
I3
I3 V1
I3 V2
V3
I3
I3SC
1
:
I3G3
I3G3 V1G1 I3G3 V2G2 V3G3
I3G3
I3G3
9:18
I1G1 I1SC
;
V1G1
I G2 I SC
2 G22 ;
V2
Y12
Y23
I1SC
I2SC
;
V2G2 V1G1
I SC
I SC
2 G3 3 G2 ;
V3
V2
Y13
Y33
I1SC
I3SC
;
V3G3 V1G1
I G3 I SC
3 G33 :
V3
9:19
Let us obtain the generalized equivalent generator similar to Fig. 9.8. For this
purpose, we consider also expression (9.19) for the rst conductivity
Y11
I1G1 I1SC
J1
:
G1
U1
V1
This expression denes the input conductivity for the passive part of three-port
network at the short circuit for the rst load and short circuit for the second and
third pairs of terminals in Fig. 9.11a.
The similar formulas for the second and third conductivities have the view
Y22
I2G2 I2SC
J2
;
G2
U2
V2
Y33
I3G3 I3SC
J3
:
G3
U3
V3
284
(a) V G1
U1
SC
I1
U2=0
Y11
U3=0
J1
G1
I1
(b)
U1=0
G2
U2
V2
Y
J2
U3=0
I2
I2
(c)
U1=0
SC
G2
Y22
G3
U3
V3
Y
J3
U2=0
I3
I3
G1
V 1 /3
U1
Y
SC
G3
Y33
J1
I1
G1
I 1 /3
V1
YL1
G2
V 2 /3
U2
J2
I2
G2
I2 /3
V2
YL2
G3
U3
V 3 /3
V3
J3
I3
G3
I 3 /3
YL3
285
U1
I1G1
I1 ;
3
I G2
J2 2 I 2 ;
3
I3G3
I3 :
J3
3
J1
9:20
The following relationships take place for the passive part of three-port network
J1 Y11 U1 Y12 U2 Y13 U3 ;
J2 Y12 U1 Y22 U2 Y23 U3 ;
9:21
9:22
Y12
;
3
3
3
3
9:23
9:24
The values of the constituent voltage and current sources of the generalized
equivalent generator do not depend on the conductivity yN in Fig. 9.10. It represents
a practical interest. This conductivity can be a part of the fourth changeable load.
Besides, the bunch center parameters of one load do not depend on the conductivities entering into the circuit of other loads. For example, the center G1 is
dened by elements y1, y1N and does not depend from y2, y2N, and y3, y3N.
286
Example 2 We use the date of Example 3, Sect. 6.2. Expressions (9.20) give the
values
10
15
1:948 5:28133; J1
0:974 4:026;
3
3
15
16:25
U2
1:64 6:64; J2
0:82 4:596;
3
3
9:377
14:844
U3
1:61 4:7356; J3
1:61 3:338:
3
3
U1
I1SC
I2SC
I3SC
10
15
9:377 15
0:176
0:282
1:236
2:641;
3
3
3
3
10
15
9:377 16:25
0:188
0:176 0:966
1:761;
3
3
3
3
10
15
9:377 14:844
0:188 1:283
0:282
2:817:
3
3
3
3
References
1. Bessonov, L.A.: Teoreticheskie Osnovy Elektrotekhniki. Elektricheskie tsepi, Izd.9. (Basic
Electrical Engineering Theory: Electric Circuits, 9th edn.). Vyshaia shkola, Moskva (1996)
2. Penin, A.: About the denition of parameters and regimes of active two-port networks with
variable loads on the basis of projective geometry. WSEAS Trans. Circuits Syst. 10(5), 157
172 (2011). http://www.worldses.org/journals/circuits/circuits-2011.htm. Accessed 30 Nov
2014
3. Penin, A.: Parameters and characteristics of the modied equivalent generator of an active
multiport network. Elektrichestvo 5, 3239 (2012)
Part III
Chapter 10
10.1
In power supply systems with limited voltage source power, the restriction of load
power, two-valued regulation characteristic, and interference of several loads is
observed [15]. Distributed power supply systems, autonomous or hybrid power
supply systems with solar cells, fuel elements, and accumulators may be examples
of such systems [8].
At the present time, a digital control of voltage converters is used. One way of
the digital control performance is a predictive technique. In one switching period, a
duty cycle for the next switching cycle is calculated, based on a sensed or observed
state and input/output information [2, 3].
Also, a feed-forward control method improves a load regulation dynamics of
converter [1]. This method calculates the required duty ratio variation by a predicted load current.
Therefore, it is necessary to take into account an internal resistance of power
supply, to carry out analysis of the load interference and obtain relationships for
denition of regime parameters at the possible coordinated predictive control for
preset load regimes. To simplify the solution of task and to reveal basic moments of
this influence, it is expedient to consider static regulation characteristics and idealized models of voltage converters.
Anyway, we consider a power supply system with two regulated voltage converters (or voltage regulators VR1, VR2) and given loads R1, R2 in Fig. 10.1 [911,
13]. Generally, voltage converters with switched tapped secondary windings of
transformers, multicell or multilevel voltage converters, pulse-width modulation
PWM converters, and so on may be examples of these regulators.
The regulators dene transformation ratios n1, n2
n1
V1
;
V
n2
V2
:
V
10:1
289
290
10
VR1
Ri
V0
V1
R
I0
VR2
n1
V2
R
n2
The interference of the regulators on the load voltages V1, V2 is observed
because of an internal resistance Ri.
Let us obtain equations describing behavior of this circuit at change of the values
n1, n 2.
From (10.1) it is evident
V1 n1 V;
V2 n2 V:
1
1
V1 2
V2 2
R1
R2
1
1
n1 2
n2 2 V2 :
R1
R2
V0 V
V:
Ri
R1
R2
2
4
10:2
This expression represents a sphere by the coordinates V1, V2, V in Fig. 10.2. For
simplication of drawing, the axes V1, V2 are superposed.
In turn, the variables n1, n2 are resulted by a stereographic projection of spheres
points from the pole 0, 0, 0 [14]. These variables dene the conformal plane n1n2
[4] the axes n1, n2 are superposed too.
10.1
291
V
V0
n1
n1M
n2
n2M
n1
n2
n1
n2
V0
2
V1 V1M
V1
V2 V2M
V2
In the plane V1, V2 we have an area of voltage changes. This area is dened by
the internal area of circle (ellipse) in Fig. 10.3a and corresponds to the spheres
equator. In this case V = V0 / 2. Taking into account (10.2), we obtain the equation
of this circle
Ri
Ri
V0 2
:
V1 2
V2 2
R1
R2
4
10:3
10:4
V0
2
r
R1
;
Ri
V2M
V0
2
r
R2
:
Ri
10:5
r
R1
;
Ri
n2M
r
R2
:
Ri
10:6
Let, for example, the rst load voltage regime V1 = const (there is a line L1 in
Fig. 10.3a) be supported by n1, n2 changes. Then, a spheres circular section L1 and
corresponding circle L1 in the plane n1n2 (there is a line L1 in Fig. 10.3b) turns out.
292
10
V2
(a)
V2M
2
k2
1
k2
K1
1
-V1
-V1M
L1
1
V1M V
1
V1 V1
n2
(b)
n2 M
2
k2
V1
k2
V1
n1 M
2
1
n1 n1
n1
L1
K1
V2 2 n2 2 V 2 :
Then, by (10.2)
V
V0
1
Ri
R1
Ri
R1
n1 2
Ri
R2
n2 2
Ri
R2
n2 2
10:7
10:8
Therefore, we have
V1
n1 V 0
n1 2
10.1
293
10:9
R1 V0 R1
0:
Ri V1
Ri
Roots n1 ; ~
n1 of this equation are characterized by the inverse property
n1 ~n1
R1
n1M 2 :
Ri
Therefore, in the plane n1n2, the two families of circles L1 correspond to the
parallel straight lines in the plane V1V2 for different values V1. These two families of
circles have the centers n1M. In turn, the values n1M conform to maximum
voltages V1M.
Such families of circles L1 describe the regime V1 = const as the rotation group
of sphere, as it is shown by arrows in Figs. 10.2 and 10.4.
This motion of points n1 n2 has the two xed points n1M and it is the elliptic
projectivity in the form
n2
an1 b
;
cn1 d
10:10
where coefcients a, b, c, d are parameters [5, 6]. The initial point n1 = n11 + jn12 and
subsequent point n2 = n21 + jn22 are complex values.
Fig. 10.4 Regime V1 = const
as the rotation group of sphere
n2
n1
L1
K1
294
10
(a)
V0
V1=0
1
k2
k2
K1
V2
(b)
n2
n1
K1
Let a set of initial points be situated on the spheres equator for different values
of V1. Projective transformation (10.10) shifts each initial point along its circle L1.
In turn, all subsequent points constitute an arc of circle K1 in Fig. 10.4. This circle
K1 is orthogonal to all circles of family L1.
Let us consider this shift in detail. The rotation of the equatorial plane of sphere
about its diameter, shown by arrows in Fig. 10.2, gives the circle family K1 with
parameters k12, k22, and so on in Fig. 10.3. This rotation by the coordinates V V2 is
shown in Fig. 10.5a.
In turn, in the plane V1V2, the projected circles k12, k22 give the family of ellipses
K1, as shown in Fig. 10.3a. Let us obtain the equation of this ellipse.
By Fig. 10.5a, we have
V0
V 2 k2 V
n2 V;
2
where k2 is an angular coefcient. Therefore
V
V0
2
V2
:
k2
10:11
10.1
295
V
2
R1
R2 k2 2
4
10:12
Next, we obtain the equation of corresponding circles in the plane n1n2. From
(10.11) it follows that
V0
k 2 n2
2
:
V
k2
Using (10.7), we get the required equation of circle K1
Ri
Ri
n2
n2 2
n1 2 2 1 0:
R2
R1
k2
10:13
10.2
At the increase of the values n1, n2 on some step of switching cycle, a running point
may pass over the equator and the voltage V2 is going down that is inadmissible.
Therefore, it is better to use such groups of transformations or movements of points
in the plane n1n2, when it is impossible to move out the running point over the circle
or equator of sphere. So, we must decrease the next values n1, n2 by some rule.
In this sense, we come to hyperbolic geometry [7]. There is Poincares model in
the plane n1, n2 shown in Fig. 10.6. The corresponding circle carries the name of
absolute and denes an innitely remote border. The equation of absolute conforms
to (10.4).
Fig. 10.6 Poincares model
of hyperbolic geometry
n2
L1
K1
n1
296
10
The arcs of circles L1 are straight lines of this model. In turn, the arcs of circles
K1 are the lines too. The distance between these lines is constant. The straight lines
L1 are orthogonal to these equidistant lines K1.
In turn, the Beltrami Klein model is the other hyperbolic geometry model in the
plane V1V2 in Fig. 10.3a. The lines L1, K1 of this model have the same sense.
Now, we may obtain required expressions or rules of regime parameters and
their changes.
n1 V 0
1
Ri
R1
n1 2
10:14
The value of this cross-ratio is constant for different initial points, as it is shown
in Fig. 10.7b.
It is also possible to constitute the cross-ratio m1n for the initial or running regime
1
n1 relatively to the origin of coordinate n1 = 0; that is,
m1n n1M n11 0 n1M
n11 n1M
:
n1M n11
10:15
10.2
297
(a) 0.5
V1
V0
0.0
n1M
-0.5
-2
-1
n1
21
21
(b)
-V1M
V1
21
21
21
n1
mn
n1
mn
mn
mn
1
n1M
21
21
mn
V1
n1
2
n1
V1M
21
n1
0
V1
2
V1
n1
-n1M
21
V1
V1
mn
The conformity of the values n11, m1n is shown in Fig. 10.7b too.
The value m1n determines the non-uniform projective coordinate of the value n11.
Hence, the point n1 = 0 is a unite point.
In the same way, the cross-ratio for the subsequent regime n21
m2n n1M n21 0 n1M
n21 n1M
:
n1M n21
10:16
n21
;
n1M
n11
n11
;
n1M
298
10
n21 1 n11 1
:
1
n21 1 n11
10:17
21
n11 m21
n 1=mn 1
:
21
1 n11 m21
n 1=mn 1
10:18
m21
n 1
:
m21
n 1
10:19
n11 n21
1
:
1 n11 n21
1
10:20
n21 n11
:
1 n21 n11
10:21
So, there is a strong reason to introduce the transformation ratio change as n21
1 .
Let us now consider the variable V1. Then, the cross-ratio mV for the initial
regime V11 and subsequent regime V21 has the form
m1V V1M V11 0 V1M
V11 V1M
;
V1M V11
m2V
V12 V1M
:
V1M V12
10:22
10:23
10:24
10.2
299
1
11 V1 ;
V
V1M
2n1
1 n1 2
10:25
10:26
12 V
11
V
2
11 :
1 V
1V
10:27
V121
Thus, there is a strong reason to introduce the value of voltage change as V21
1 .
21
The validity of such denition for the changes n21
1 and V1 is conrmed by
expression (10.25); that is,
V121
2n21
1
2
1 n21
1
10:28
300
10
n2
1n
2n
k2
1n
2n
2
1
k2
3n
2n
1n
2n
3n
2n
F3n
1n
K1
1n
2n
3n
0
2
V1
V1
n1
A
L1
3n
2n
A1n
V0
k2
2V1
2
#
Ri
Ri
V0
2
2 Ri
k2
1 n1
n1
R2 R1
V1
R2
2 Ri
k2
1 0:
R2
In turn,
n2 k 2
V0
1 n1
:
2V1
10:29
10:30
Further, we must determine the coordinates of points A1, F1 with the aid of
Eqs. (10.4) and (10.9)
Ri
R2
Ri
R2
n2 2
n2 2
Ri
R1
Ri
R1
n1 2 n1 VV01 1 0
n1 2 1 0:
10.2
301
Then
V1
n1 2 ;
V0
s
R2 R2
V1 2
:
n2
2
Ri R1
V0
10:31
10:32
The coordinates of the points A1n, C1n, D1n, F1n are dened by normalized
complex values
nA1 nA1
nA1
1 j
2 ;
nC1 nC1
nC1
1 j
2 ;
nD1 nD1
nD1
1 j
2 ;
nF1
nF1
nF1
1 j
2 :
jnA1
nF1
2 j
2 ;
2
2
1 nF1
where
nF1
nF1
1 V1 ;
1
2 1:
Then, the cross-ratio has the form
nD1
nA1 nC1 nA1
:
nD1
nF1 nC1 nF1
10:33
The result of calculation shows that this cross-ratio is a real value. The
cross-ratio m21
nL1 corresponds to the points C2n, D2n and so on.
Let us determine transformation (10.10) for the line L1 with parameter V11. Using
(10.33), we get the subsequent value
nD1
F1
C1
21
n1 jnF1
1 m21
n 1
nL1 =mnL1 1
2F1
;
21 =m21 1
nC1 n1 jnF1
1
m
2
nL1
nL1
10:34
where
nC1 is the initial value.
In particular, when the initial and subsequent value situated on the axis n2 , this
expression has the form
302
10
nD1
2
21
21
nC1
2 1 mnL1 =mnL1 1
:
21
21
1 nC1
2 1 mnL1 =mnL1 1
m21
nK1 1 C2n C1n 1
10:35
nC2 1 nC1 1
:
1
nC2 1 nC1
10:36
The result of calculation shows that this cross-ratio is a real value. The
cross-ratio m21
nK1 corresponds to points, D1n, D2n and so on.
Let us determine transformation (10.10) for line K1. Using (10.18) and (10.19),
we get the subsequent value
nC2
nC1 n21
K1
;
1 nC1 n21
K1
10:37
m21
nK1 1
m21
nK1 1
10:38
10.2
303
V2
k
F1V
F2V
2
2
k2
3V
k2
1V
1V
1V
K1
1V
2V
2V
E
D
3V
3V
2V
2V
3V
3V
V1
V
1
1
V1
A1V
A2V
A3V
10:39
The coordinates of the points A1V, D1V, C1V, F1V are dened by normalized
complex values
A1 V
11 jV
2A1 ;
V
11 jV
2C1 ;
C1 V
V
D1 V
11 jV
2D1 ;
V
F1 V
11 jV
2F1 :
V
10:40
m21
VL1
D1
V
C1
V
V
2
V
2
D1 V
C1 V
F1 V
F1 V2F1 1 V2F1 1
V
:
V2C1
2D1 2F1 2C1 2F1 V2D1
V2 V2
V2 V2
2
2
F1 1
F1 1
10:41
304
10
10:42
Let us determine transformation (10.10) for line L1 with parameter V11. Using
(10.41), (10.36) and (10.37), we get the subsequent value
C1
V
21
2
2D1
F1 VL1
V
V
2
;
C1
V
F1
21
V
2
1 V2F1 VL1
10:43
D1
V
2
21
VL1
F1 V F1
m21 1
V
VL1
2 V D1 V2C1
21
mVL1 1 1 2F1 2F1
10:44
V
V
2
2
2 .
determinates the voltage change for V
Moving of points along the line K1
Similarly to (10.36) let us constitute the cross-ratio
2 1
m21
VK1 1 V1 V1 1:
10:45
The result of calculation shows that equality (10.23) and (10.42) takes place
21 2
m21
VK1 mnK1 :
1 .
Let us determine the transformation which similar to (10.26) for voltage V
Then
21
1
12 V1 VK1 ;
V
21
11 VK1
1V
10:46
12 V
11
V
m21
VK1 1
11
12 V
m21
1V
VK1 1
1 .
determinates the voltage change for V
10:47
10.2
305
2 . To do this, we use
Now, let us determine the transformation for voltage V
ellipse Eq. (10.12) for normalized values
!
R2 1
V1 1
V2 2 1:
Ri k2 2
2
10:48
11 ; V
21 , V
12 ; V
22 of the initial C1V and subsequent
Substituting the coordinates V
C2V points into (10.48), we get the system of equation
8
!
>
R2 1
>
1 2
>
21 2 1
>
V 1
V
>
< 1
Ri k2 2
!
>
>
R
1
2
2
>
2
2 2
>
>
: V1 1 Ri k 2 V2 1:
2
22
Using (10.46), we obtain the equation for V
21
11 VK1
V
21
11 VK1
1V
2
1 2
22 2 1 V1 1:
V
21 2
V
q
21
V
21 2 :
1 VK1
21
11 VK1
1V
10:49
10.3
Example
Consider the circuit in Fig. 10.1. Let the parameters be given as follows:
V0 5;
Ri 1;
R1 1:25;
R2 2:
306
10
p
V2M 2:5 2 3:5353:
p
n2M 2 1:4142:
5
1 0:
V1
V2 2 6:25;
k2 2
n2
0:5n2 2 0:8n1 2 2 1 0:
k2
0:8V1 0:5
V12 2:5:
n21 0:6909:
10.3
Example
307
m21
n
0:2068 1:118
1:4539:
1:118 0:2068
0:6909 1:118
4:2353 2:913 1:4539:
1:118 0:6909
0:6909
0:618;
1:118
n11
0:2068
0:185:
1:118
2:913 1
0:4888:
2:913 1
0:185 0:4888
0:618:
1 0:185 0:4888
n21 0:4888
0:618 0:4888
0:85:
and so on.
We see that actual values of the transformation ratio changes are decreasing at
each time.
Cross-ratio (10.22) for initial and subsequent regimes by voltages V11, V21
m1V
1 2:795
2:1142;
2:795 1
m2V
2:5 2:795
17:949:
2:795 2:5
308
10
Equality (10.23)
m1V 2:1142 1:45392 ;
11
V
1
0:3577:
2:795
0:8944 0:3577
0:7892:
1 0:8944 0:3577
2 0:4888
0:7892:
1 0:48882
10.3
Example
309
A1n
C1n
D1n
F1n
n1
n2
n1
n2
0.4
0.2269
0.26
0.4
1.3206
0.4326
0.7
1.3206
0.3578
0.203
0.2323
0.3578
0.9338
0.3059
0.4954
0.9338
0:4;
nF1
2
r
2
0:42 1:3206:
2
1:25
We obtain the same cross-ratio for the points C2n, D2n with the voltage V21.
The coordinates of all required points are presented in Table 10.2.
Using (10.34), we dene
1 m21
1 1:6282
nL1
4:1837:
1:6282
1
m21
1
nL1
Then, the subsequent value
nD1
0:2323 j0:4953:
A2n
C2n
D2n
F2n
n1
n2
n1
n2
1.0
0.7325
0.7948
1.0
0.6324
0.2674
0.4104
0.6324
0.8944
0.6552
0.71088
0.8944
0.4472
0.1891
0.2902
0.4472
310
10
2:9135 1
0:4889:
2:9135 1
1 V
2
Plane V
Moving of points along the line L1
11 0:3577 are given in
The coordinates of required points for the voltage V
Table 10.3.
Cross-ratio (10.41)
m21
VL1
Equality (10.42)
2:6519 1:62822 :
Voltage change (10.44)
V 21
L1
A1V
C1V
D1V
F1V
2:6519 1
0:4523:
2:6519 1
n1
n2
V2
2
V
0.4
0.2269
0.26
0.4
1.3206
0.4326
0.7
1.3206
3.3015
1.906
2.6956
3.3015
0.9338
0.5391
0.7624
0.9338
10.3
Example
311
F1
V2
1 0:5391
0:9338 0:4523
0:8944 1 0:3577 1
8:4867:
0:8944 1 0:3577 1
21 2
Let us check the equality m21
VK1 = (mnK1) . Then
8:4867 2:91352 :
Voltage change (10.47)
V 21
K1
8:4867 1
0:7892:
1 8:4867
1 0:78922 0:2582:
1 0:3577 0:7892
2C2
V
A2V
C2V
D2V
F2V
n1
n2
V2
2
V
1.0
0.27325
0.7948
1.0
0.6324
0.2674
0.4104
0.6324
1.5811
0.9128
1.2909
1.5811
0.4472
0.2582
0.3651
0.4472
312
10
1 0:3577 0:7892
2D2
V
We see that the results of this calculation coincide with the data of Table 10.4.
References
1. Chae, S., Hyun, B., Agarwal, P., Kim, W., Cho, B.: Digital predictive feed-forward controller
for a DCDC converter in plasma display panel. IEEE Trans. Power Electron. 23(2), 627634
(2008)
2. Chattopadhyay, S., Das, S.: A digital current-mode control technique for DCDC converters.
IEEE Trans. Power Electron. 21(6), 17181726 (2006)
3. Chen, J., Prodic, A., Erickson, R.W., Maksimovic, D.: Predictive digital current programmed
control. IEEE Trans. Power Electron. 18(1), 411419 (2003)
4. Conformal geometry. Encyclopedia Wikipedia. http://en.wikipedia.org/wiki/Conformal_
geometry (2014). Accessed 30 Nov 2014
5. Frank, J.A.: Schaums Outline of Theory and Problems of Projective Geometry. McGraw
Hill, New York (1967)
6. Glagolev, N.A.: Proektivnaia geometria. (Projective geometry). Nauka, Moskva (1963)
7. Hyperbolic geometry. Encyclopedia Wikipedia. http://en.wikipedia.org/wiki/Hyperbolic_
geometry (2014). Accessed 30 Nov 2014
8. Lui, X., Wang, P., Loh, P.: A Hybrid AC/ DC micro grid and its coordination control. IEEE
Trans. Smart Grid 2(2), 278286 (2011)
9. Penin, A.: Geometrical properties of regulated voltage converters at limited capacity voltage
sources. Tehnicheskaia elektrodinamika 3, 6164 (1991)
10. Penin, A.: Fractionally linear relations in the problems of analysis of resistive circuits with
variable parameters. Electrichestvo 11, 3244 (1999)
11. Penin, A.: Analysis of regimes of voltage regulators with limited capacity voltage sources.
Geometrical approach. WSEAS Trans. Circuits Syst. 12(1), 12 (2013). http://www.wseas.org/
wseas/cms.action?id=6933. Accessed 30 Nov 2014
12. Penin, A.: Projective geometry method in the theory of electric circuits with variable
parameters of elements. Int. J. Electron. Commun. Electr. Eng. 3(2), 1834 (2013) https://sites.
google.com/site/ijeceejournal/volume-3-issue-2. Accessed 30 Nov 2014
13. Penin, A.: Non-Euclidean geometry and regulated characteristics of limited capacity power
supply. J. Electr. Eng. 2(4), 175186 (2014)
14. Stereographic
projection.
Encyclopedia
Wikipedia.
http://en.wikipedia.org/wiki/
Stereographic_projection (2014). Accessed 30 Nov 2014
15. Yehia, D.M., Yokomizu, Y., Iioka, D., Matsumura, T.: Delivaraible-power dependence on
distribution-line resistance and number of loads in low-voltage DC distribution system. IEEJ
Trans. Electr. Electron. Eng. 7(1), 2330 (2012)
Chapter 11
11.1
We use the same power supply system with two idealized voltage regulators
VR1; VR2, and load resistances R1, R2 in Fig. 11.1 [5, 6]. The regulators dene the
transformation ratios n1 = V1/V2, n2 = V2/V; an internal resistance Ri determines the
interference of the voltage regulators on load regimes.
For convenience, we rewrite Eq. (10.2)
Ri
Ri
V0 2 V0 2
2
2
:
V1
V2 V
R1
R2
2
4
11:1
R1
2
4
11:2
For different load values (R11 ; R21 and so on), this expression represents a bunch of
circles (ellipses) by the coordinates V1, V in Fig. 11.2.
Let the stabilized load voltage V1 = V1= be given [7]. Then, the vertical line with
the coordinate V1= intersects the bunch of circles in two points.
At the same time, the transformation ratio n1 is resulted by a stereographic
projection of circles points from the pole 0,0. Therefore, the load resistance R11
corresponds to the variable n11 ; R21 corresponds to n21 and so on.
313
314
11
VR1
Ri
V1
V0
n1
VR2
V2
R
n2
Fig. 11.2 Stereographic
projection of bunch of ellipses
on a tangent line n1
V
n1
-R1
3
V0
n1
R1=
n1max
n1
1
n1
n1
R1min
2
R1
V0
R1
V1max
0
-R1
V1
V1=
For a minimum value R1min of the load resistance, the circle is tangent to the
vertical line V1=. In this case, V = V0/2. Using (11.2), we get the condition
Ri
R1min
V1 2
V0 2
:
4
11.1
315
V1 2
V0 2
11:3
V1
:
V0
11:4
The operating area of all the circles must be above the diameters of these circles;
that is, V V0 =2. Therefore, we use the upper point of the intersection.
In turn, on some step of switching period at increase of the parameter n1, a
running point may pass over the diameter that is inadmissible. Therefore, it is better
to use such groups of transformations or movements of points along the line n1
when it is impossible to move out the running point over the diameter. So, we must
decrease the next values n1 by some rule. In this sense, we come to hyperbolic
geometry.
V1
:
V0
11:5
However, the question arises about the range 0\n1 \n11 of this transformation
ratio. According to Fig. 11.2, this range corresponds to negative load values R1
and expression (11.2) determines a hyperbola. In this case, the load gives energy
back and the voltage source V0 consumes this energy as it is shown in Fig. 11.3.
In this regard, we consider a physical realization of such a power source, as a
negative resistance. For this purpose, we remind something of the electric circuit
theory by examples of two circuits in Fig. 11.4.
Let a voltage source V1 be connected to a resistance R1 for the left-hand circuit.
Then, the negative resistance R1 and the drain current I1 correspond to this
voltage source.
We consider Fig. 11.3 again. Then, it is possible to connect up the voltage source
V1 instead of the resistance R1 . At the same time, the input resistance of this
316
11
Ri
I1
VR1
V0
I0
V1
-R1
-
n1
Fig. 11.4 Negative resistance
R1 corresponds to a voltage
source V1 and its practical
realization by PWM boost
converter
I1
I0
I1
R1
V1
Ri
V1
V0
-R1
R1
circuit must be equal to the constant value R1 at the voltage V1 change. This
condition is satised due to the variable value n1. We obtain so-called a loss-free
resistance [8]. The example of another such a circuit is the right-hand circuit in
Fig. 11.4 as a high-power-factor boost rectier [4].
Taking into account the value n1 = V1=/V and Eq. (11.2), we obtain
V1
n1 V0
1
Ri
R1
n1 2
11:6
V0 R1 R1
0;
V1 Ri
Ri
R1
n1 2
:
Ri n1 VV0 1
1
11:7
n11 0
n11 0
;
n1max n11 n1max
n11
where the points 0; n1max are base ones and n11 is a unit point.
11:8
11.1
317
R1
R1min
n1
n1max
n1
0 n11
V1
V0
V1
V0
n11
n11
:
2 VV10 n11 n1max n11
11:9
The conformity of the points n11 ; m1n is shown in Fig. 11.6. In this case, the value
m1n is a non-homogeneous coordinate of n11 .
1
2
Further, the cross ratio m21
n , which corresponds to the regime change n1 ! n1 ,
has the form
2 1
m21
n 0 n1 n1 n1max
11:10
n21
n1max
n11
n11
n1max
21
n1
0
n1 n1
1
n1max
n1
mn
R1
R1min R1
I1max
n1
21
mn
0
mn
mn
21
R1
0
R1
21
I1
I1
I1
I1
318
11
we get the regime change (segment n21 n11 ) or cross ratio (11.10) in the view
21 n11 1
m21
n 0 n
n21 n11 1
:
n11 n21 1
11:11
Similarly to (10.19), we may obtain the analogous expression for the change n21
1
of the transformation ratio so that the following relationships are performed
n21
1
m21
n 1
;
m21
n 1
m21
n
n21
1 1
:
1 n21
1
11:12
For this purpose, we make the substitution of variables so that to use ready
expressions (10.19). Therefore, we introduce the value
~n1 2n1 1
11:13
~n21 ~n11
:
1 ~n21 ~n11
n21
n21 n11
n21
1 :
n11 2n21 n11
11:14
In this expression, the changes of the variables are equal to among themselves;
21
that is, ~
n21
1 n1 .
There is a following foundation for this equality. Linear expression (11.13)
preserves a cross ratio and, consequently, the regime change m21
n . On the other
21
hand, the change of the transformation ratio n1 is expressed by regime change
(11.12).
0.5
n1
-1
n1
n1
n1
11.1
319
n11 1 n21
1
:
21
1
1 n1 2n1 1
11:15
There is a group transformation. Moreover, if the initial value n11 1, then the
subsequent value n21 1 regardless of the value n21
1 . Therefore, such movement of
point corresponds to hyperbolic geometry.
Similarly to the above, let us consider the cross ratio for the load resistance R1 .
Using the dependence R1 n1 in Fig. 11.5, we demonstrate the conformity of the
variables R1 ; n1 by Fig. 11.6. The cross ratio for the initial point R11 relatively to the
base points 0; R1min , and a unit point R1 1 has the form
m1R 0 R11 1 R1min
R11 0
1
:
R1min 1 4 R1i V1 22
R V
R11
11:16
Expression (11.16) equals the corresponding cross ratio for the conductance Y1
1=R1 and the load current I1 U1 =R1 . Also, the following equality takes place
mR mn 2 :
11:17
11:18
11:19
1 R21
1
:
1 R21
1
1
11 R1 :
R
R1min
11:20
320
11
11:21
Then, there is a strong reason to introduce the value of load resistance change
21
as R21
1 and the value of transformation ratio change as n1 .
The validity of such denitions for changes is conrmed by the following
expression similar to initial expression (11.6); that is,
R21
1
2n21
1
2
1 n21
1
11:22
Using (2.21), we obtain the subsequent value R21 of the load resistance
21
R
11 1 R21
R
1
:
1 1
1 R21
2
R
1
1
11:23
21 1
11 1, then the subsequent value R
As well as (11.15), if the initial value R
regardless of the value R21
1 .
Thus, a concrete kind of a circuit and character of regime imposes the
requirements to denition of already system parameters.
Therefore, arbitrary and formal expressions for regime parameters are excluded.
Example Let the circuit parameters be given as follows
V0 5;
Ri 1;
V1 2:5:
R21 1:25:
V1 2
V0 2
1:
V1
1:
V0
11.1
321
V1
0:5:
V0
n21 0:691:
0:585
1:41:
1 0:585
1:41
n11 n21 n1max
n21 n11
0:106
0:226:
2
1
1
2
n1 n1 2n1 n1 0:467
1
1
2
4 RRi1VV12
1 0
1
2:0:
1 4 6:25
225
The equality
21 2
2
m21
R mn 1:581 2:5:
322
11
2n21
1
2
1 n21
1
2 0:226
0:429:
1 0:051
R1
R2
2
4
11:24
By denition,
n1
V1
;
V
n2
V2
:
V
11:25
2
R2 2
n2 V0
R
R
1 R i n1 2 R i n2 2
R1
11:26
It follows that
(
Ri
R1
Ri
R1
n1 2
n1 2
Ri
R2
Ri
R2
V0
n2 2 nV11
1 0
2
n2 V0
n2 V2 1 0:
11:27
11.1
323
By denition (11.25)
n2 n1
V2
:
V1
11:28
#
1
1 V2 2
n1 V 0
n1 Ri
1 0:
R1 R2 V1
V1
2
11:29
YT :
RT R1 R2 V1
11:30
If, for example, the load voltages are equal to among themselves, V2 V1 ,
then these loads are connected in parallel and
1
1
1
YT :
RT R1 R2
Finally, we get the expression
n1 2
Ri n1 V0
1 0:
RT
V1
11:31
n2max 2 VV20 :
11:32
Further, it is possible to use the above idea of hyperbolic geometry in the case of
one load.
324
11.2
11
We consider again the circuit shown in Fig. 11.1. Let the rst load voltage V1 be
stabilized. But for all that, the rst load resistance may be both positive R1 [ 0 and
negative R1 \0. Also, the second constant load resistance is positive R2 [ 0.
For example, the circuit in Fig. 11.8 corresponds to the positive load R1 [ 0 and
PWM regulators in Fig. 11.9a conform to the negative load R1 \0.
We rewrite Eq. (11.1)
Ri
Ri
V0 2 V0 2
2
2
;
V1
V2 V
R1
R2
2
4
which correspond to a surface with a parameter R1 in the coordinates V1, V2, V3.
If R1 [ 0, this expression represents a sphere (ellipsoid) similarly to the circle in
Fig. 11.2. The both loads consume energy; the voltage source V0 gives energy.
If R1 \0, we get a one-sheeted hyperboloid in Fig. 11.9b [3]. The rst load, as a
constant voltage source V1 , gives energy. In addition, the voltage source V0, as
energy storage, may consume and give back energy. The corresponding direction of
the current I0 determines these regimes.
For different values R1, our expression represents a bunch of spheres or
hyperboloids. If R1 1, as the open circuit regime, the corresponding surface
degenerates into a cylinder.
VR1
Ri
V0
I1
V1=
I0
VR2
n1
I2
V2
R
n2
Fig. 11.8 Power supply system with the invariable voltage V1 and given load R2
11.2
325
I1
(a)
I2
Ri
V1=
V2
I0
V0
R2
R1
(b)
V
V0
V2
0
V1
Fig. 11.9 a Power supply system with resistance R1 . b Its geometric model
R2 const:
R2
2
4
R1
11:33
The second member of this equation is a radius of circle for the given value R1.
It is possible to consider the voltage V2 change as the radius-vector rotation. This
rotation determines a point movement along the straight line V1 in coordinates V1,
V in Fig. 11.11. This gure at V2 = 0 is analogous to Fig. 11.2.
326
11
V
V0
R1=
-R1
R1
V0
2
V2
V
-R1
V0
n1
n1max
R1 =
n1
V0
2
R1>0
V1
0
-R1
V1=
In the general case of the variable V2, we get the surfaces, which rotate around
the diameter V = V0/2, as it is shown by closed arrows. Also, the transformation
ratios n1, n2 are resulted by the familiar stereographic projection of spheres points
on the tangent plane or conformal plane. The axes n1, n2 are superposed in
Fig. 11.11.
Further, we use the rst equation of system (11.27)
Ri
Ri
n1 V0
n1 2
n2 2
1 0:
R1
R2
V1
11:34
11.2
327
11:35
11:36
V1 2
V0 2
V1
;
V0
n2
R1<0
R1=
2
R1
1
R1
0
n1max
n1
328
11
Next, we display the voltage V2 value for the trajectories in Fig. 11.12. Using
denition (11.25), we obtain
V2 V1
n2
:
n1
V0
2
rr
R2
R1min
1
;
Ri
R1
n2max
rr
R2
R1min
1
:
Ri
R1
11:37
It is interestingly to note that all these tangent lines to the curves correspond to
the value n1max . Therefore, the operating area of the transformation ratio is limited
by the value n1 n1max . So, we must decrease the next value n1, for the next
regulator switching period, by some rule. In this sense, we come to hyperbolic
geometry.
n1
n1max
n2
n2
:
n2ref
As the scale value n2ref , we may use a circle with some characteristic value of the
parameter R1. The resistance value R1 1 may be such a characteristic value.
Using (11.35) and the value n1max , we get
11.2
329
(b)
(a)
g2
n2
g2
R1= 1
R1<0
R1
0
0.5
1
R1
n1
g1
R1= 1
g1
-1
R1<0
n2ref
r
R2
:
Ri
11:38
11:39
1
;
1 g1
n2
g2
:
1 g1
11:40
Let us check the offered expressions. In this case, Eq. (11.39) transforms into the
equation of circle; that is,
g1 2 g2 2 1
R1min
r1 2 :
R1
11:41
The second member of this equation is a radius squared for the given R1. We
may term the variables g1 ; g2 as hyperbolic transformation ratios.
This geometric model allows to use a cross ratio for the determination of regimes
and their change.
330
11
11:42
D1
gD1 gD1
2 jg1 ;
gF1 gF1
2 j0:
F1
In particular, radius of half-rounds (11.41) denes the coordinates gA1
2 ; g2 as the
follows
gA1
2 r1 ;
gF1
2 r1 :
g2
D2g
1
F1g
D1g
C2g
C1g
jg1
1 B2g
B1g
1
R1
A1g
-1
A2g
11.2
331
For Poincares model of hyperbolic geometry by Fig. 11.14, cross ratio (11.42)
looks like [2]
mDC
g
tghC r1 gC1
r1 gD1
2
2
:
D1
tghD
g
gC1
1
1
11:43
mDC
g
2
r1 gC1
r1 gD1
2
2
:
C1
r1 gD1
r1 g2
2
11:44
2
g2
gD1
2
r1 gC1
;
r1
1 r21 gDC
2
DC
gDC
2
2
mDC
1
g
11:45
11:46
This change corresponds to the points C2g ; D2g of the half-rounds with parameter
R21 and so on.
We may obtain an expression for the subsequent value of the transformation
ratios n1 ; n2 . To do this, it is necessary to apply to (11.45) the inverse change of
variables relatively to (11.40); that is,
g1
1 n1
;
n1
g2
n2
:
n1
11:47
11:48
332
11
V
B2V
V0
B1V
C1V
R1
A2V
C2V
D1V
V0
2
A1V
D2V
F1V
V2
2
V2DC
V2D1
V2max V C1
;
V2max 1 2 V DC
V2max 2
11:49
mDC
V 1
gDC
2 :
mDC
V 1
11:50
11:51
11.2
333
11.2.3 Example
Let the circuit parameters be given as follows
V0 5;
Ri 1;
R2 2;
V1 2:5;
R1min 1;
n1max 1:
3:5
:
1:5
Further, we use the voltage value V C1 3:5 because this value is greater than
V0 =2 2:5.
Maximum values (11.37)
V2max
n2max
r
pp
5 p
1
2 1
p
Radius (11.41), r1 0:2 0:447.
Transformation ratios (11.25)
nC1
1
2:5
0:714;
3:5
nC1
2
0:707
0:202:
3:5
nC1
2
0:202
0:143:
1:414
1 0:714
0:4;
0:714
gC1
2
0:143
0:2:
0:714
334
11
Therefore,
2:5
1:414
0:833; nD1
0:333;
2
3
3 1:414
1 0:833
0:333
0:2; gD1
0:4:
2
0:833
0:833
nD1
1
gD1
1
0:4
0:2
p
2
0:2 0:2
0:2 0:4
mDC
p
6:8528 2:6182 :
p
g
0:2 0:2
0:2 0:4
mDC
g
6:8528 1
0:7453:
6:8528 1
Equality (11.51)
1:414
0:4
0:8944:
1:581 0:4472
References
1. Glagolev, N.A.: Proektivnaia geometria. (projective geometry). Nauka, Moskva (1963)
2. Kagan, V.F.: Osnovania geometrii, Chasti II. (Geometry basics. Part II). Gostekhizdat, Moskva
(1956)
3. Korn, G.A., Korn, T.M.: Mathematical handbook for scientists and engineers. MacGrawHill,
New York (1968)
4. Maksimovic, D., Jang, Y., Erickson, R.W.: Nonlinear-carrier control for highpowerfactor
boost rectiers. IEEE Trans. Power Electron. 11(4), 578584 (1996)
References
335
5. Penin, A.: Analysis of regime of voltage regulators with limited capacity voltage sources.
Geometrical approach. WSEAS Trans. Circuits Syst. 12(1), 12 (2013). http://www.wseas.org/
wseas/cms.action?id=6933 Accessed 30 Nov 2014
6. Penin, A.: Non-Euclidean geometry and regulated characteristics of limited capacity power
supply. J. Electr. Eng. 2(4), 175186 (2014)
7. Penin, A.: NonEuclidean geometrical transformation groups in the electric circuit theory with
stabilization and regulation of load voltages. Int. J. Circuits Syst. Sign. Process. 8, 182194
(2014). http://www.naun.org/cms.action?id=7621 Accessed 30 Nov 2014
8. Singer, S., Erickson, R.W.: Powersource element and its properties. IEE ProceedingsCircuits,
Devices and Systems. 141(3), 220226 (1994). http://ecee.colorado.edu/*rwe/papers/IEE94.
pdf
Chapter 12
12.1
Introduction
As it was noted in Chap. 10, into power supply systems with limited capacity
voltage sources, the limitation of load voltage is appeared. Renewable power supply
with solar array, fuel cell, and rechargeable battery may be the example of ones [4].
The output voltage of these sources changes over a wide range. In such systems it is
convenient to apply the pulse-width modulation PWM boost and buckboost regulators or converters which can step up or step down the input voltage [2, 10], but
these regulators have a nonlinear regulation curve or characteristic. Therefore, the
problem of linearization is appeared [9].
The linearization methods are proposed for idealized, without losses buckboost
regulators [3, 5].
Also, the real boost and buckboost converters have the two-valued regulation
characteristic. In this case, the up-slope direction or the forward branch of its
regulation characteristic is used and the down movement of the operating point on
the back branch is restrained.
Therefore, it is necessary to correctly determine the regime parameters of the
converter relatively to the maximum permissible load voltage and control pulse
width. It will allow estimating, for example, reserves of the control voltage and load
voltage, to use this data for a direct digital or predictive control, and to carry out
some kind of the linearization of the regulation characteristic in a wide range of
load voltage changes.
12.2
Let us consider a PWM boost voltage converter with a given load resistance RL in
Fig. 12.1 [8].
Springer International Publishing Switzerland 2016
A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4_12
337
338
12
VL
Ri
CL
RL
V0
0
PWM
Vreg
We know the expression of the static regulation characteristic for the continuous
current mode of the choke L with the loss resistance Ri [1]
VL V0
1D
1 D2
Ri
RL
V0
1
1D
r2
1 1D
2
12:1
where D 1 is a relative pulse width and r is a relative loss. For convenience, let
the converter be given by the following parameters:
V0 25;
r 0:08:
160
VL
120
80
40
0
0,0
0,2
0,4
0,6
0,8
1,0
12.2
339
1
1:
1D
12:2
This value is the inverse relative pause width or the transformation ratio of the
idealized boost converter
n
VL
:
V
12:3
n
1 r n2
12:4
12:5
160
VL
80
-80
nM
-160
-15
-10
nM
1
-5
10
15
340
12
V
n
V0
nM
V0
2
0
VL VLM
VL
V0
156:25;
2r
1
nM 12:5:
r
12:6
V0
1 r2
n 1;
D 0:
12:7
For the initial values n1 ; VL1 , the corresponding cross ratios are
m1n
1
n1 1 r
1
1
;
n1 1
:
r1
1 1r
r
r
rn
m1v
V0 1 V0 V0
V
2r L 1 r2 2r
V0
2r
V0
2r
VL1 1 r2
:
m1n 2 :
VL1 1 r2
12:8
12:9
12.2
341
V0
V0 1
21
VL
1+
VL
VL
V0 1
VL
21
n
1
1+
D 1
n
D
21
mV
SCL
mV
mV
mV
mV
SCL
SCL
0
0
By denition (12.2), the values D and n are connected among themselves by the
fractionally linear expression
D
n1
:
n
1 r D1 1 r
:
:
1 r D1 1 r
12:10
Expressions (12.8) and (12.9) lead to identical values if we take the logarithm
S1 Ln m1v 2Ln m1n :
Then, the hyperbolic distance
V0
1
VL1
1r
1r
2r VL
Ln
:
2Ln
2Ln
V
1
1
0
1r
1r
2r VL
2r VL
V0
S1 Ln 2r
V0
342
12
1r 2
\1;
1r
Ln mSCL
2Ln
V
1r
\0:
1r
The corresponding hyperbolic distance will be positive for inverse value of the
cross ratio:
SSCL 2Ln
1r
[ 0:
1r
Further, it is natural to introduce the normalized hyperbolic distance for a running regime (the index 1 is lowered), using the obtained scale
r
VL
1
1:
r
2Ln 11r
2r VL
V0
S
SSCL
Ln 2r
V0
12:11
Thus, the normalized distance considers all the characteristic points. In turn, the
inverse expression is
r2r 1
1
V0 11r
:
VL
1 r2r 1
2r
1
12:12
1r
:
1
1
n2 n1
2
1
r
r
rn
rn
r 1r2 n2 n1
2
m21
n
n2 n1
1 r2 n2 n1
12:13
Then,
n21
:
21
rn
r
m21
n 1
12:14
12.2
343
n1 n21
1 r2 n1 n21
12:15
Similarly, for the voltage change VL1 ! VL2 , it is possible to write down at once
VL21
;
21
2r VL
V0
2r
m21
V V0
VL21
VL2 VL1
1 4r2
VL2 VL1
V0 2
VL2
VL21 VL1
1 4r2
VL21 VL1
V0 2
12:16
1 r D2 1 r D1 1 rD21
;
1 r D2 1 r D1 1 rD21
D1 D2
;
1 r2 D1 D2 D1 D2
n21 D21 :
12:17
Thus, the mutually coordinated system of all the regime parameters turns out.
We will consider the linearization of the regulation characteristic of a converter
in Fig. 12.6 [6]
VL
Ri
RL
CL
inverse
nonlinear
function
calculator
V0
0
PWM
Vreg
Vfb
nonlinear
function
calculator
Vref
0
344
12
1 r r 1
1
1r
;
1 r r 1
1
1r
r r 1
1
1 11r
n
:
r
1
r 1r
1
12:18
1r
It turns out that the value r is equally expressed by D and VL. It is possible to
interpret this equality as the linearization of the dependence VL(D).
Further, it is possible to accept that the value D is equal to the regulating voltage
Vreg for PWM. We believe that the input voltage DV of a nonlinear function
calculator (introduced before PWM) is the hyperbolic distance r.
Therefore, we obtain, by (12.18), that
1 rDV 1
1
1
12:19
V0
Vfb Ln 2r
V0
12:20
So, we obtain that Vfb is equal to DV. If to use a feedback closed loop (shown by
the dashed line in Fig. 12.6), the output voltage of the error amplier will be the
voltage DV. ORCAD model of this converter is proposed by [8].
Fig. 12.7 Nonlinear
characteristic of the function
calculator
1.0
Vreg
0.8
0.6
0.4
0.2
0.0
0
10
15
12.2
345
16
Vfb
12
0
0
40
80
120
VL
160
VLM
Example 1 Let VL1 48:49 be the initial regime. We must dene the normalized
distance for this regime.
For n 1; we have a unit value
VL 25
1
24:841:
1 0:082
156:25 48:49
0:8522 1:378:
156:25 48:49
Then,
m1n
q p
m1V 1:378 1:174:
346
12
24.841 48.49
156.25 VL
69.85
156.25
2.949
12.5
1.986
12.5
0.92
0.661
1.08
0.496
1.9
0.7256
1.378
mV
0.32
0.32
The corresponding value n1 may be calculated from the inverse formula to (12.8)
n1
r 1
m 1
1 11r
1 r n1
1:986:
r 1r mn 1
In turn,
D1 0:496;
The cross ratio and distance for the scale point are
mSCL
V
1r 2
0:725;
1r
SSCL 2Ln
1r
0:320:
1r
S1
1:
SSCL
12.2
347
1:17422 1 1
2:618 1
69:85:
156:25
22
1
2:618 1
1
1:174
69:85 48:49
24:841:
1 4 0:0064 69:8548:49
252
On the other hand, as the change VL21 is equal to a unit value, the change is
n 1 by (12.7).
Then, by (12.15), we get the subsequent values
21
n2
1:986 1
2:949;
1 0:0064 1:986 1
D2 0:661:
The values VL ; D; n for the further steps are shown in Fig. 12.9.
It is visible that each time the actual voltage change is reduced and cannot reach
the maximum value.
Example 3 Let the regime with the initial value VL1 48:49 be changed to the
value VLN 1 69:85 by small steps consistently. In turn, the consecutive reduction
of the change step reduces undesirable transients.
Let the number of steps be N = 5. It is necessary to nd the values VLi ; ni on the
each step i.
We nd the hyperbolic distance r 61 corresponding to the initial (the zero step)
regime and nal (the fth step) regime. According to Example 2, this distance is as
follows: r 61 1.
For these ve steps, the regime change (as a hyperbolic distance) equals
Dr 0:2.
The value (or length) of the rst step is
r 2 r 1 r 21 1 0:2 1:2:
The rst step is VL2 according to (12.12) and n2 by (12.18) are given as
1:17422:2 1
52:96;
1:17422:2 1
1 1:174 2:2 1
2:183:
n2
0:08 1:1742:2 1
VL2 156:25
348
12
52:963 48:49
5:
1 4 0:0064 52:96348:49
252
The changes on the following steps are also equal to this value.
The transformer ratio change on the rst step according to (12.13) is
n21
2:183 1:986
0:202:
1 0:0064 2:183 1:986
For subsequent steps, the values n; VL are obtained via the recurrent relationships
(12.15), (12.16), since the changes on the following steps keep their values
2:183 0:2029
2:379;
1 0:0064 2:183 0:2029
52:963 5
VL3
57:34;
1 4 0:0064 52:9635
252
n3
n4 2:574;
VL4 61:617;
n5 2:768;
VL5 65:78;
n6 2:96;
VL6 69:84:
12.3
VL
V
Ri
L
RL
CL
V0
PWM
Vreg
12.3
349
The static regulation characteristic of the idealized converter for the continuous
current of the choke L has the well-known view:
VL V
D
:
1D
D1 D
2
1 D
Ri
RL
V0
D
1D
D 2
1 r2 1D
12:21
1
1:
1D
Then
VL V0
n1
1 n 12 r2
1
D
VL
1
:
1D
1D
V
12:22
160
VL
120
80
40
0
0,0
0,2
0,4
0,6
0,8
1,0
350
12
This value, at that ncon 0, is the transformation ratio of the idealized converter
itself.
Finally, we obtain
VL V0
ncon
1 r ncon 2
12:23
This equation corresponds to (12.4). Thus, we may use the above results.
Regulation characteristic (12.23) is given in Fig. 12.12. The working area involves
the zero points ncon ,VL .
Next, we may use ellipse Eq. (12.5) at once
r2 VL 2 V2 VV0 0:
The plot of this ellipse is given in Fig. 12.13.
Fig. 12.12 Example of
regulation characteristic via
the converter transformation
ratio
160
VL
80
-80
-160
-20
-15
-10
-5
10
15
ncon
ncon
V0
1 +1
V0
2
0
VL V0
VL
20
ncon
n
12.3
351
The values ncon , n turn out at the expense of a stereographic projection too. The
maximum values are
VL M
V0
;
2r
1
ncon M :
r
12:24
ncon 0;
n 1;
D 0:
As it is visible, this case of a unit point differs from above case (12.7).
V0
V0
2
1+
V0
1+
21
VL
VL
VL
V0
VL
21
n
1
1
1
0.5
1+ 1
1
1+
n
1
D
21
mV
0
SCL
1/mV
S
SCL
SCL
mV
mV
mV
SCL
mV
352
12
For the initial values n1con , n1 , VL1 , the corresponding cross ratios are similar to
(12.8), (12.9)
1
1
1
1
1
m1n n1con 0
n1 1 1
r
r
r
r
1
1
1
1
n
1
n
con
r ;
r1
12:25
1
1
1
n
1
con
r
r n
m1V
V0
VL1
V0 1
V0
VL 0
m1n 2 :
2r
V0
1
2r
2r
V
L
2r
12:26
1
1
D1 0
r1
r1
1
1
r1 D
1
1
r1 D
r1
:
r1
12:27
Expressions (12.25) and (12.26) lead also to identical values if we take the
logarithm
VL
:
VL
2r
V0
S1 Ln m1V Ln m1n ;
S1 Ln 2r
V0
VL
V0
1 r2
n 2;
D 0:5:
This scale point should be considered too. The cross ratio and hyperbolic distance for the scale point are as follows:
mSCL
V
SSCL
!
V0
V0
V0
1 r2
0
;
2
2r 1 r
2r
1 r2
1r
:
Ln mSCL
2Ln
V
1r
12:28
VL
V0
1 r2
n 0;
D 1:
12.3
353
This point is also specied in Fig. 12.14. In this case, the cross ratio will be the
inverse value to the above received value, but the hyperbolic distance will be
identical in the absolute value.
Further, it is natural to introduce the normalized hyperbolic distance for the
running regime (the index 1 is lowered), using the obtained scale
r
S
SSCL
VL
1r
:
2Ln
1r
2r VL
V0
12:29
Ln 2r
V0
12:30
1r
1 2
1
1
ncon ncon
r
r
1
n2con r1 n1con r
1
n2con r1 n1con
1
r
1
r
n2con n1con
1r2 n2con n1con
n2con n1con
1r2 n2con n1con
n2con n1con
1 r2 n2con n1con
n2 1 n1 1
1 r2 n2 1n1 1
Then
n21
:
21
rn
r
m21
n 1
12:31
n1con n21
1 r2 n1con n21
12:32
354
12
Similarly, for the voltage change, VL1 ! VL2 , it is possible to write down at once
VL21
;
21
2r VL
V0
2r
m21
V V0
VL21
VL2 VL1
1 4r2
VL2 VL1
V0 2
VL2
VL21 VL1
1 4r2
VL21 VL1
V0 2
12:33
1
2
r1 D
1
2
r1 D
1
1
r1 D
1
1
r1 D
12:34
D2
1D2
D
1D
1
D
D
1 r2 1D
2 1D1
2
D2 D1
1 D2 D1 D2 D1 1 r2
12:35
:
Thus, the mutually coordinated system of all the regime parameters turns out.
Similar to Fig. 12.6 and relationships (12.18)(12.20), we can use the linearization of the regulation characteristic.
For that, we give, similar to (12.30), the expression for the nonlinear function
calculator as
Vreg D
1 r r
1
1r
:
1r r
1 r 1r 1 r
12:36
V0
Vfb Ln 2r
V0
12:37
12.3
355
VL
Ri
RL
CL
V0
PWM
Vreg
D1 D
2
1 D r
V0
n1
1 rn2
12:38
The plots of these dependences are close to Figs. 12.11 and 12.12.
Similar to (12.22), let us introduce the variable V so that
VL
n 1:
V
12:39
12:40
The plot of this ellipse is shown in Fig. 12.16. The value n turns out at the
expense of a stereographic projection.
Let us obtain the characteristic values, using the data of the previous example.
So, we give the ready formulas for the maximum values
VLM
V0
V0
p 2
2
2rr 1 r 2r nM
144:2 VLM
25
;
2 0:080:08 1:0032
169:24 VLM
p (
13:54 nM
r 1 r2
nM
:
r
11:54 n
M
We accept the points VL 0; n 1 as a unit points, and VL 25; n 0 as the
scale points.
356
12
nM
V
n=0
V0
n=1
+
nM
169.24
VLM
V0 25
144.2
VL
mV
0.726
0
VLM
0.852
mn
+
DM
DM
1.086
0.926
VL1 VLM
VLM
;
1
VL VLM VLM
1
n n
M 1 nM
n1 1 nM 1
:
n nM 1 nM
m1V VLM
VL1 0 VLM
m1n n
M
V0 VLM
VLM
0:726;
V0 VLM VLM
n 1 nM
M
0:852:
nM 1 n
M
mSCL
V
mSCL
n
Vreg
p2r 1
1 r= 1 r2
1
p 1r=p
1 r2
D 1 r2 r 1 r2
p2r 1
1 r=p
1 r2
1
2
1r= 1 r
12:41
12.3
357
Vfb r
V0
VL r 1 r2
p
Ln 2r
V0
2
2r VL r 1 r
p
1 2rr 1 r2
p :
Ln
1 2rr 1 r2
12:42
References
1. Erickson, R.W., Maksimovic, D.: Fundamentals of Power Electronics. Springer, Berlin (2001)
2. Iskender, I., Genc, N.: Design and analysis of a novel zerovoltage-transition interleaved boost
converter for renewable power applications. Int. J. Electron. 97(9), 10511070 (2010)
3. Lo, Y.K., Chen, J.T., Lin, C.Y., Ou, S.Y.: Improved control-to-output characteristics of a
PWM buck-boost converter. Int. J. Circuit Theory Appl. 39(2), 203209 (2011)
4. Lui, X., Wang, P., Loh, P.: A Hybrid AC/ DC micro grid and its coordination control. IEEE
Trans. Smart Grid 2(2), 278286 (2011)
5. Michal, V., Premont, C.H., Pillonet, G.: Switched DC/DC boost power stage with linear
control-to-output conversion ratio, based on the ramp-modulated PWM generator. European
patent EP 2482433 A2, 1 Aug 2012
6. Penin, A.: Method of regulating the voltage with step-up and inverting pulse converters. MD
Patent 4067, 26 Aug 2008
7. Penin, A.: Regimes analysis of the voltage pulse regulators on the basic of the invariance
property of the control characteristics. Probl. Reg. Energ. 1 (2009)
8. Penin, A.: Analysis of regimes of voltage regulators with limited capacity voltage sources.
Geometrical approach. WSEAS Trans. Circuits Systems. 12(1), 12 (2013) http://www.wseas.
org/wseas/cms.action?id=6933. Accessed 30 Nov 2014
9. Sira Ramirez, H., Rios-Bolivar, M.: Sliding mode control of dctodc power converters via
extended linearization. IEEE Trans. Circuits Syst.1. Fundam. Theory Appl. 41(10),
652660 (1994)
10. Sundareswaran, K., Devi, V., Nadeem, S.K., Sreedevi, V.T., Palani, S.: Buckboost converter
feedback controller design via evolutionary search. Int. J. Electron. 97(11), 13171327 (2010)
Part IV
Chapter 13
13.1
Introduction
In the electric circuit theory, the concepts of power-source and power-load elements
are developed on the basis of generalization of results in power processing systems
[14, 8]. Analysis of a power supply system, which contains a power-load element,
shows the two-valued voltage of a limited capacity voltage source. But for all that,
the volt-ampere characteristic of this power-load element has one-valued representation. On the other hand, taking into account the losses of real power-source
and power-load elements, we get the two-valued volt-ampere characteristic of these
elements [6]. Also, the influence of supply line losses onto the power-load element
has practical importance [5, 7].
Therefore, it is desirable reasonably to determine the regime parameters of the
power-source and power-load element using the single-valued area of their characteristics; that is, to present the regime parameters in the relative form.
13.2
Let us consider the simplest power supply system with a voltage regulator (stabilizer) VR and given load resistance R1 in Fig. 13.1. The load power
P
V1 2
IV;
R1
or
I P=V
361
362
13
I1
VR
V0
Ri
V
V1=
I
R1
So, we have a power-load element P for different values of voltage V and current I.
In turn, the voltage source characteristic
I
V0 V
;
Ri Ri
13:1
~
as a straight line, intersects the hyperbola into two points M, M.
Using load power characteristic I P=V, we get the quadratic equation
V0 V V2
P:
Ri
Ri
13:2
s!
4PRi
1 1
;
V0 2
~ V0
V
2
s!
4PRi
1 1
:
V0 2
13:3
13.2
363
Note
The offered graphic representation of both power-load P and power-source
P elements corresponds by analogy to a voltage source and source of current.
Further, we will consider power-load elements using the results [5, 6].
13.3
364
13
For a different voltage source value V01 , a new point I 1 ; V01 will be the energy
equivalent point to the initial point I; V0 because this circuit does not possess the
own or internal scales, which can specify the qualitative characteristics of an
operating regime. It is true for the innite large values I 1; V 1 of our ideal
circuit. Therefore, regimes of different comparable circuits are equivalent always.
4PRi
V0 2
0:
13:4
I
Ri
V0
M2
P
M1
M
-VM
M1
0
VM
M2
V
V0M V 1
-V0M
V
2
V0
13.3
365
13:5
Ri P
V:
V
13:6
13:7
where the points VM V20M ; VM V20M are base ones and V 1 is a unit point.
The conformity of the points V 1 , m1V is shown in Fig. 13.8.
Fig. 13.6 Characteristics of
the power-load element in the
projective coordinates
M1
M1
VM
V0M
-V0M
S
-VM
366
13
V
0
VM
21
-VM
-V0M
V0
mV
21
V0
V0M
V0
V0
21
mV
0
mV
mV
1
2
Further, the cross-ratio m21
V , which corresponds to a regime change V ! V ,
has the form
m21
V
V0M 2 1
V0M
V V
2
2
V 2 V20M V 1 V20M
:
V 2 V20M V 1 V20M
13:8
1
1 2 V [ 1;
V
V0M
13:9
13.3
367
m21
:
V
2
1
V 1 V 1
2 V 1 1
V
13:10
13:11
V 21 1
;
V 21 1
V 21
m21
V 1
:
m21
V 1
13:12
13:13
1 1, then the
There is a group transformation. Moreover, if the initial value V
2
21
subsequent value V 1 regardless of the value V .
Similarly to the above, let us consider the cross-ratio m10 for the voltage V0 using
the conformity of the variables by Fig. 13.8 for the single-valued area. This
cross-ratio has the form
V 1 V0M
m10 V0M V01 1 V0M 01
:
V0 V0M
13:14
13:15
13:16
The base points V0 ; V0 and points VM ; VM , as the limit points, correspond to
the innitely large distance. Therefore, this limit regime has a clear physical sense.
Similarly to (13.8), the cross-ratio m21
0 , which corresponds to a voltage source
1
2
regime change V0 ! V0 , has the view
V02 V0M V01 V0M
2
1
m21
:
0 V0M V0 V0 V0M 2
V0 V0M V01 V0M
13:17
368
13
1
01 V0 [ 1;
V
V0M
13:18
m21
0
2 1 V
1 1
V
02
01
V0 1 V0 1
2V
1
V
0 0 1
1
2 V
V
0
0
2V
1 1
V
0 0
2 V 1
V
0
0
1
1
13:19
13:20
V021 1
;
V021 1
V021
m21
0 1
:
m21
0 1
13:21
13:22
Then, there is a strong reason to introduce a voltage source change as V021 and
a voltage change of power-load element as V 21 .
The validity of such denitions for the changes is conrmed by the following
expression similar to initial expression (13.6); that is,
2V021
1
V 21 :
V 21
13.3
:
V
2
~
2
5
2 5
2V
p
p
p
Voltages (13.5), VM PRi 3 2 6.
Cross-ratio (13.7) for the initial regime V 1 3
m1V
p
3 6 0:5505
p
0:10102:
3 6 5:4494
V 2 V20M V 1 V20M
0:4903 0:10102 4:8536:
V 2 V20M V 1 V20M
1 p3 1:2247:
V
6
2:924 1:2247 1
1:519:
2:924 1:2247
1:519 1
4:8536:
1:519 1
p
5 2 6 0:101
p
0:0102:
9:9
52 6
369
370
13
Distance (13.16)
H 1 Ln 0:0102 2Ln 0:10102 4:5848:
Voltage source regime change (13.17)
m21
0
p
82 6
p 0:0102 0:2404 0:0102 23:5578:
82 6
1
5
01 V0 p
1:021:
V
V0M 2 6
1:0887:
1:633 1:021
0:6124
V021 1 1:0887 1
23:5578:
V021 1 1:0887 1
V0 2
;
4P
VM
V0
:
2
13:23
13.3
371
I
Ri
V0
M1
Ri
RiM
M1
M
0
S
1
V V0
VM
V0 V V2
:
P
P
13:24
S
VM
V0
372
13
Ri
RiM
VM
RiM
Ri
mV
V0
Ri
mV
Ri
VM V 1 V0
Fig. 13.11 Dependence Ri V for a given load power and conformity of different regime
parameters
Therefore, the single-valued mapping of the parabola points onto the axis V
takes place.
Using expression (13.7), we may constitute the cross-ratio m1V for the initial
point V 1 ; that is,
V 1 VM V 1 V0 =2
:
m1V VM V 1 V0 1
V0 =2
V0 VM
13:25
The points VM V20M ; V 1 are base ones and V0 is a unit point. The
conformity of the points V 1 , m1V is shown in Fig. 13.11.
Similarly to the above, let us consider the cross-ratio m1i for the resistance Ri
using the conformity of the variables by Fig. 13.11 for the single-valued area. This
cross-ratio has the form
RtiM R1i
m1i RiM R1i 0 1
:
RiM
13:26
13:27
13.3
373
I
V0
Ri
M1
V
P
PM
+
M1
0
V
VM
V0
~ 1 . Also, the
For example, this line intersects the hyperbola P1 into points M1 , M
V0 2
;
4Ri
VM
V0
:
2
13:28
V0 V V2
:
Ri
Ri
13:29
13:30
P
PM
VM
PM
mV
V0
mV
VM V 1 V0
Fig. 13.13 Dependence PV for a given voltage source and conformity of different regime
parameters
374
13
Similarly to the above, let us consider the cross-ratio m1P for the power P1 using
the conformity of the variables by Fig. 13.13 for the single-valued area. This
cross-ratio has the form
PM P1
:
m1P PM P1 0 1
PM
13:31
13.4
13:32
V0 V
V0
P
:
R0 R0 R0 IR0
From here,
V0 I
P
IR0 :
I
13:33
V0 V
R0
V0
V
+
V0(I)
V0
1
V0
V(I)
V0M
VM
IM
13.4
375
Evidently, this equation contains the sum of the hyperbola and straight line.
For different voltage source values (V01 , V02 and so on), we get the voltage source
characteristic as parallel horizontal lines. The pairs of the currents I 1 , ~I 1 , and
~ 1 correspond to the given source voltage V01 . Therefore, the
voltages V 1 , V
volt-ampere characteristic V0 I of this voltage source is the two-valued one.
From (13.33), we get
V0 V
PR0
V:
V
Obviously, this expression coincides with (13.6) and corresponds to Fig. 13.7.
Therefore, all expressions (13.7)(13.22) remain in force.
I0
I1
"
Y00 Y10
# "
Y10 Y11
V0
V1
where Y parameters
Y00 y10 y0 ;
Y11 y10 y1 ;
I1
V1
y10
V0
y0
y1
YL
I0
Y10 y10 :
YIN
I0
YIN
YL
I0
V0(I0)
V0
0
+
V0 V1
V1(I1)
V0
V1
Fig. 13.15 Variable voltage source, power-load element with loss two-port and their
characteristics
376
13
3
V0
4 I0 5 chc
shy
q
ch2 c
Y00 Y11
Y10
3
V1
shc 4 5
I1 ;
chc
q
sh2 c
DY
Y10 2
13:34
p
DY is the characteristic admittance.
Y00 YL DY
;
Y11 YL
YL
YIN
q thc
;
q
1 YqL thc
13:35
3
V1
4 I1 5 chc
shy
q
3
V0
shc 4 5
I0 :
chc
q
I1 P1
I0
I0
Finally, we obtain
V0 2 q
I0 2
chc2 shc2
P1
V0 I0
0:
q
shc chc
shc chc
13:36
13.4
377
V0
The pairs of input conductivities YIN , Y~IN , and the pairs of currents I0 , ~I0 correspond to a given source voltage V0 that determines the above two-valued
volt-ampere characteristic V0 I0 . In that sense, the value YIN denes the
single-valued regime.
The maximum efciency value KP and the minimum input power value P0 MIN
corresponds to the admittance matching YL YIN q; that is,
KP chc shc2 ;
P0 MIN P1 =KP :
13:37
s
P1
;
q
V0
s
P0 MIN
:
q
13:38
For different voltage source values, we get the voltage source characteristic as
parallel vertical lines in Fig. 13.16. These lines intersect the hyperbola into pairs of
~ 1 by the voltage V01 . The arrow shows the
points, for example, the points M1 , M
conformity of these points. Also, the characteristic points M , M correspond to
the tangent lines V0M . The closed arrows illustrate these xed points.
Fig. 13.16 Two-valued
regime of a variable voltage
source
I0
M1
M1
V0M 0
V0
V0 M V0
V0
378
13
To nding the xed points, we obtain the following expression from (13.34)
V0 V1 chc
I1
P1
shc V1 chc
shc:
q
qV1
This expression is similar to (13.6) and Fig. 13.7. We suppose the derivation
equals zero; that is,
dV0
P1
chc
shc 0:
dV1
qV1 2
From here, we obtain
V1M
s
P1
thc:
q
13:39
s
P1
thc 2V1M chc:
2 chc
q
13:40
I0
S
M
V0M V 0
-V0M
V0
13.4
379
+
V0M
V0
V1M
V1
m0
m0
+
V0M
V1
V0
V0
V1M V1
m0
Further, we may use all expressions (13.7), (13.8), (13.14)(13.17). The conformity of all the points is shown in Fig. 13.18.
But, we must take into account the characteristic points for KP ; that is, the
voltages V1 , V0 .
Let us consider the cross-ratio mV for the voltage V1 similarly to (13.7)
V VM
mV VM V1 1 VM 1
:
V1 VM
13:41
13:42
13:43
Further, we may introduce the normalized distance or relative deviation from the
matched regime
D1
H1
Ln m10
Ln m1V
H
Ln m0
Ln mV
13:44
p
5 2:2361;
p
5:25 2:2913;
P1 8:
380
13
s r
P1
8
4;
0:5
q
V0
s r
P0 MIN
12:466
4:9932:
0:5
q
s
r
P1
8
0:2182 1:8685;
thc
0:5
q
I11 P1 =V11 1:
I01 1:9191:
8 1:8685
V11 V1M
m10
8:6448 3:8293
V01 V0M
13.4
381
4:9932 3:8293
V0 V0M
H Ln m0 Ln 0:1319 2:0257:
Relative deviation from the matched regime (13.44)
D1
13.5
H 1
0:9519
0:4699:
H 2:0257
V1M
; 1; V1M
. We must determine all the characteristic values for voltage V1 and
resistance r0 .
r0
V0
r1
r10
V1
P
+
382
13
I1
r0
M1
r0
r0
Ri =0
M1
V1M
r0
V1 M V 1
V1
VL
r0
R i=
IL
I1
r0
V
r0
M
r0
V1M
r0
V1
V1M V
1
G
As it was shown above, the load straight lines intersect at the point G, which has
coordinates ILG ; VLG . Physically, it means that regime parameters do not depend on
the value r0 ; that is, the current across this element is equal to zero at the expense of
the load voltage.
Therefore, it is convenience to use the generalized Thvenin/Helmholtz equivalent generator of our network shown in Fig. 13.22. Then, the parameters of this
generator are the following values
V0
;
r10
13:45
r1 r10
V0 :
r10
13:46
ILG
VLG
13.5
383
I1
Ri
G
IL
VL
V1
+
Fig. 13.22 Equivalent circuit for the power-load element with the supply line
r0 r10
:
r0 r10
13:47
In turn,
r0
r10 Ri r1
:
r10 Ri r1
13:48
r0V
RIi 1;
r10 r1
;
r10 r1
r0I r10 :
13:49
13:50
Thus, an equation of straight line, passing through the point G, has the form
I1
VLG V1
ILG :
Ri
13:51
VLG V1
V1 ILG :
Ri
13:52
VLG V1
:
P V1 ILG
13:53
Then, we get
Ri V1
384
13
Ri
Ri
Ri
1
Ri
V1M
V1M
V1
VL
V1
; V1M
. We suppose that
V1M
dRi VLG 2V1 VLG V1 V1 ILG
0:
dV1
P V1 ILG
P V1 ILG 2
The solution of this equation gives the two roots
V1M
1
p
1 VLG ILG =P
;
ILG
V1M
1
p
1 VLG ILG =P
:
ILG
13:54
Substituting (13.54) in (13.53), we get Ri ; R
i . Using (13.48), we obtain r0 ; r0 .
The conformity of the points V1 , Ri , r0 is shown in Fig. 13.24.
V1M
+
Ri
V1M V1
Ri
Ri
Ri
Ri
r0
r0
r0
r0
mV
r0
mV
VL
r0
V1
H
D
mV
1
H
1
Ri
13.5
385
Let us constitute the cross-ratio m1V for the initial point V11 ; that is,
1
V11 V1M
m1V V1M
V1 1 V1M
;
1
V1 V1M
13:55
where the points V1M
; V1M
are base ones and V1 1 is a unit point. The mV
values are shown in Fig. 13.24.
Similarly to the above, let us consider the cross-ratio m1i for the resistance R1i
using the conformity of the variables by Fig. 13.24 for the single-valued area. This
cross-ratio has the form
R1i Ri
1
m1i Ri R1i 1 R
:
i
Ri R
i
13:56
Next, let us consider the cross-ratio m10 for the resistance r01 using the conformity
of the variables by Fig. 13.24. This cross-ratio has the form
r 1 r0 r0I r0
m10 r0 r01 r0I r0 01
I
:
r0 r0
r0 r0
13:57
13:58
13:59
Further, we must support the characteristic points VLG , RVi , r0V . Then
G
VLG V1M
mVV V1M
G
VL 1 V1M
;
VL V1M
RVi Ri Ri
V
mVi Ri RVi 1 R
;
i
Ri
Ri R
i
V
I
r r0
r0 r0
:
mV0 r0 r0V r0I r0 0V
r0 r0 r0I r0
13:60
13:61
386
13
H1
:
HV
13:62
r10 5;
r1 1; P1 3:
10
2;
5
VLG
15
10 12:
5
r0V
51
0:8333;
51
V1M
3
1
q
1 122
3
2
3;
V1M
3
1 3
6:
2
Ri V1M
VLG V1M
12 3
3;
G 3
33 2
P V1M IL
R
i 6
12 6
12:
362
r10 Ri r1 5 3 1
r10 R
512 1
i r1
9:1666:
r0
r10 R
5
12 1
r
1
i
r0
V11 4;
r01 3:0882:
13.5
387
r01 r10
2:909:
r01 r10
Cross-ratios (13.55)(13.57)
m1V
V11 V1M
43
0:1;
46
V11 V1M
R1i Ri
2:909 3
0:01 0:12 ;
1
2:909 12
Ri Ri
r 1 r0 r0I r0
m10 01
I
r0 r0
r0 r0
3:0882 3:3333 5 3:3333
0:02 2 0:01:
VLG V1M
12 3
0:5;
G
12 6
VL V1M
R
3
mVi i
0:25 0:52 ;
12
Ri
r V r0 r0I r0
I
mV0 0V
r0 r0
r0 r0
0:8333 3:3333 5 3:3333
0:25:
mVV
4:6052
3:3219:
1:3863
388
13
References
1. Cid-Pastor, A., Martinez-Salamero, L., El Aroudi, A., Giral, R., Calvente, J., Leyva, R.:
Synthesis of loss-free resistors based on sliding-mode control and its applications in power
processing. Contr. Eng. Pract. 21(5), 689699 (2013)
2. Emadi, A., Khaligh, A., Rivetta, C.H., Williamson, G.: Constant power loads and negative
impedance instability in automotive systems: denition, modeling, stability, and control of
power electronic converters and motor drives. IEEE Trans. Veh. Technol. 55(4), 11121125
(2006)
3. Griffo, A., Wang, J., Howe, D.: Large signal stability analysis of DC power systems with
constant power loads. In: Vehicle Power and Propulsion Conference 2008. VPPC08. IEEE,
16 (2008)
4. Maksimovic, D., Jang, Y., Erickson, R.W.: Nonlinear-carrier control for high power factor
boost rectiers. IEEE Trans. Power Electron. 11(4), 578584 (1996)
5. Penin, A.: Analysis of regime changes of constant powerload. Elektrichestvo 12, 4349 (2008)
6. Penin, A.: Constant power sources and its properties. Elektrichestvo 4, 6065 (2010)
7. Sergeev, B., Romash, A., Nagovitsyn, V., Kurchenkova, N.: Analysis of electric circuit with
constant powerload. Elektrichestvo 6, 1622 (2002)
8. Singer, S., Erickson, R.W.: Powersource element and its properties. IEEE Proc. Circ. Devices
Syst. 141(3), 220226 (1994)
Chapter 14
14.1
In the above Part I, we have considered active two-poles with the load straight lines.
In turn, there are active two-poles with self-limitation of the output current, which
are characterized by nonlinear load curves [13]. So-called the zero-current
switching load resonant converters have these load curves [9]. The typical form
of this load curve is represented by a convex line 1 in Fig. 14.1. In the rst
quadrant, the active two-pole gives energy; the corresponding regime is changed
from the short circuit SC to the open circuit OC. Thus, depending on degree of
convexity, curve 1 can alter from line 2 to line 3.
In the second and fourth quadrants, the active two-pole consumes energy, but
there is limitation of the current even for the high load voltage.
Examples of load quasi-resonant converters
Example 1 It is known the load-resonant converters, for example, with the
zero-current switching [3, 9]. These converters regulate their output by changing of
a switching period TS. Let us consider DC resonant converter with given elements
in Fig. 14.2. It has self-limitation of the output current if the switching period
p
TS 2Tr; the value Tr 2p Lr Cr1 Cr2 is the own oscillation or resonance
period.
Let us suppose ORCAD simulation for TS = 32 S, Tr = 16 S [15]. The load
curve for the rst quadrant is almost rectangular and represents two obvious areas.
Area 1 corresponds to a voltage source; the load voltage depends on its current a
little. Area 2 corresponds to the start of the current limitation and determines the
maximum load power point PM. Areas 3, 4 correspond to a current source. In
particular, point 4 denes SC regime. So, the limitation of the load current takes
place.
Springer International Publishing Switzerland 2016
A. Penin, Analysis of Electrical Circuits with Variable Load Regime Parameters,
Power Systems, DOI 10.1007/978-3-319-28451-4_14
389
390
14
I
I
SC
3
2
3
V
OC
V
1
3
Fig. 14.1 Typical load lines of an active two-poles. 1 Convex. 2 Linear. 3 Rectangular
i T1
(a)
(b)
Cr1
T1
0.4 F
2V0
Cr2
250V
8 H
T2
Lr
30
2
PM
20
0.4 F
I, A
1
10
I
V
V0
0
0
40
RL
(c)
80
80
120
V, Volt
60
40
iT1 , A
1
20
0
t
-20
-40
Tr =16
Fig. 14.2 Known quasi-resonant converter. a Electric circuit. b Load curve with characteristic
areas by numbers 14. c Superposed current pulses of one transistor for these characteristic areas
14.1
391
The current pulses for these characteristic areas explain the line convex and its
horizontal part. The forward pulse amplitude 1 is determined by a light load, which
changes from OC to the maximum load power. The maximum amplitude corresponds to a point PM and equals about 80 A. Area 3 determines the overload regime
for a heavy load, the forward amplitude is decreased, and the reverse pulse appears.
For SC regime, the reverse amplitude is equal to the forward amplitude 4 if we do
not consider losses. We note that the total amplitude of forward and reverse pulses
does not change for areas 24 and equals about 80 A.
Example 2 The proposed DC resonant converter in Fig. 14.3 with given elements
has the natural self-limitation of the output current [14]. The switching period
p
TS Tr; the value Tr 2p Lr1 Cr 43 lS is the own oscillation period.
Fig. 14.3 Proposed
quasi-resonant converter.
a Electric circuit. b Load
curve with characteristic areas
by numbers 16. c Current
pulses of both transistors for
characteristic areas 3, 4
(a)
iT1
Lr1
Lr2
36 H
T1
36 H
T2
Cr
RL
2V0
1.3 F
150 V
voltage load
iT 2
resistive load
(b)
30
5
2
PM
44
20
I, A
TSTr
TS , S
1
2
50
PM
10
4
5
0
20
(c) 40
20
88
40
V, Volt
PM
60
V0 80
6 6
iT1
iT2
iT1
iT1,iT2, A 0
iT2
t
-20
-40
Tr /2=21.5 S
392
14
Let us consider ORCAD simulation, for the rst quadrant, of the load curves (a
resistive load) with the different switching period TS [15]. The period TS = 44 S is
the minimum working value and Ts = 88 S is the doubled period.
Area 1 for all the curves corresponds to a voltage source too. Area 2 corresponds
to the start of the current limitation and determines the maximum load power point
PM. Area 3 in the form of an inclined line corresponds to the further current
limitation up to SC point 4.
The current pulses for the characteristic areas 3, 4 explain the load line convex
and its inclined part. Area 3 determines the overload regime for a heavy load; the
forward amplitude does not decrease but the short reverse pulse appears.
For SC regime, the reverse pulse extends and its amplitude equals to the forward
current amplitude.
In the general case, for construction of the load line in all the quadrants, it is
necessary to use an alternating voltage source of variable amplitude and corresponding frequency as a load. Therefore, in the second and fourth quadrants, the
converter transfers power from this voltage load to the own voltage source 2V0. In
this case, there is a limitation of the current even for the high load voltage too. It is
clear for areas 5, 6. In particular, for area 5, the reverse pulse amplitude is being
increased more than the forward current amplitude. In turn, the forward current is
absent for area 6 and the converter is simply a rectier.
Further, we note that the load curve for the rst quadrant is almost rectangular
for the doubled period TS = 88 S too. On the contrary, if the period TS Tr, SC
current rises and the load line aspire to a straight line.
The represented load characteristics allow carrying out the analysis and justifying an equivalent generator or circuit of such an active two-pole with
self-limitation of output current.
In addition, similarity of characteristics of such a quasi-resonant converter and
characteristics of different nonlinear elements as transistors, solar cells, and so on
attracts attention.
14.2
Such points of the characteristic regimes as SC and OC are suggesting the possibility of using an equivalent generator with a nonlinear internal resistance Ri1 in
Fig. 14.4 [13]. The voltage source VOC and SC current ISC
1 determines the corresponding value of the internal resistance
RSC
i1
V0OC
:
I1SC
14.2
393
14:1
Accepting the linear dependence of line 1, it is possible to obtain an approximation expression of a load characteristic and analytically to check condition
(14.1). It will be the weighty logical justication of the above equivalent generator.
For obviousness, lines 2, 3 show different dependences of Ri1(V) in Fig. 14.5 too.
If the resistance Ri1(V) is not dependent on the voltage (the straight line 2), we get a
voltage source with this internal resistance and a usually linear load characteristic in
Fig. 14.1; that is, line 2. If the resistance Ri1(V) goes through the point VOC
Fig. 14.4 Equivalent
generator and its load
characteristic
XS
Ri2
I2
SC
SC
AI1 +I2
I
SC
S
SC
Ri1
V
OC
I1
I1
RL
SC
I2
I1
I2
AV
OC
I
0
OC
V
YS
394
14
SC
Ri1
3
OC
Ri1
OC
AV
OC
Fig. 14.5 Internal resistance dependence of the typical active two-pole. 1 With self-limitation of
the load current. 2 Real voltage source. 3 Ideal current source
(the straight line 3), we have a current source with the rectangular characteristic in
Fig. 14.1; that is, line 3.
Then, the given case represents the intermediate version, when the straight line 1
passes through the point (AVOC, 0). Line 1 equation has the following appearance
for the work area, V VOC,
Ri1
V
OC
V
:
A
I1SC
1
14:2
V OC V
:
I1
14:3
1V
:
AV
14:4
Here and further, the value V is a normalized voltage by the value VOC.
Expression (14.4) denes a hyperbole. The parameter A determines degree of
convexity of curve 1 in Fig. 14.1 for the working area V VOC. If A , this
curve degenerates into a straight line in Fig. 14.1; that is, line 2. If A = 1, we get a
rectangular characteristic; the hyperbola merges with its asymptotes (Fig. 14.1;
lines 3).
The load power has the view
P
1V
;
AV
A
V
PSC
G
14:5
14.2
395
where PSC
G is the maximum power of the equivalent generator for SC regime. This
expression determines a hyperbola and is similar to efciency expression (4.49).
Therefore, we may use the known results.
The maximum load power
p p2
PM
A A A1 :
SC
PG
14:6
p
AA 1 VM :
14:7
1V
I2SC 1 V;
AV
14:8
V OC
:
Ri2
14:9
Expression (14.8) denes a hyperbola with a center S and asymptotes SXS, SYS in
Fig. 14.4 for ISC
1 = 1. In practice, for a given resultant curve I(V), we must nd the
SC
following settings as ISC
1 , A, and I2 . To do this, the system of three Eq. (14.8) for
1
1
2
three pairs of points (I , V ), (I , V2), and (I3, V3) is being solved.
396
14.3
14
The maximum load power corresponds to different load voltage values depending
on the parameter A or proportions between the hyperbolic and linear component of
the currents. For clarity, this voltage coordinate can accept values from (14.7) for
the hyperbolic characteristic up to the value for the straight line one; that is,
1 V+M 0.5. Therefore, likewise to Sect. 4.4, the similar problem arises for scales
and relative expressions of the above deviation.
To do this, we consider regime symmetries of load characteristic (14.8) for a
given value of A in Fig. 14.6. Let the load be changed from SC to OC regime. We
pass to the projective coordinates, which is determined by a center F similarly to
Figs. 2.24 and 4.22. The point F is formed due to the intersection of the tangential
lines or asymptotes FXF, FYF at the xed or base points ISC, VOC. In this case, a
point on the hyperbola is assigned as the rotation of radius vector RFF from the
initial point P+M to a running point or load R1.
Therefore, the pole F and polar TQ (which passes through the points ISC, VOC)
determine the mapping or symmetry of the region of power consumption by the
load on the region of power return; that is, P P.
In turn, the pole T and polar FQ, as a complementary system, determine the
mapping or symmetry of the running points R1, R2 relatively to the maximum power
point P+M. The point T is formed due to the intersection of the tangential line TP+M and
the known line TQ. In this case, the point on the hyperbola is assigned as the rotation
of radius vector RTT from the base point P+M to the running point or load R1.
Further, we will consider the concrete example of the load characteristic in
Fig. 14.3b. For TS = 44 S, the following settings, as ISC
1 = 24.33, A = 1.016, and
ISC
=
9.805
are
obtained.
2
Equation (14.8), for all the normalized currents by ISC
1 = 24.33, has the view
IV 1:016
1V
0:4031 V:
1:016 V
14:10
T
SC
A+I2
XS
XF
I
SC
R2
Q
+
PM
RT
R1
RF
SC
I2
I2
0
VQ
1 A
V
OC
YF
V
YS
14.3
397
The maximum load power coordinates equal V+M = 0.86, I+M = 0.968. The point
F coordinates have the form
A
0:9845;
2A 1
VF
IF
A
A1
I2SC
0:99:
2A 1
2A 1
14:11
dI
A1
A
I2SC 1:071
dV
A V2
determines the slope angle for the tangential line TP+M. The solution of these
equations gives VT = 1.464, IT = 3.457.
Similarly, we nd the point Q coordinates using the equations of the straight
lines TQ and FP+M accordingly
IF IM
0:176
VF VM
determines the slope angle for the line FP+M. The solution of these equations gives
VQ = 0.371.
Let us map the harmonic conjugate points T, ISC, Q, VOC of the polar TQ onto
the axis V in Fig. 14.7. These points correspond to the points VT, 0, VQ, 1
accordingly.
The mutual mapping of the points VT, VQ relatively to 0, 1 is shown by arrows.
For these harmonic conjugate points, the cross-ratio equals 1; that is,
0 VT VQ 1 0 1:464 0:371 1
1:464 0 0:371 0
1:
1:464 1 0:371 1
Similarly, let us map the harmonic conjugate points Q, P+M, F, PM of the polar
FQ onto the axis V. These points correspond to the points VQ, V+M, VF, VM
accordingly.
398
14
VT
hyper.
0.5
VM
VQ
1.464
0.371
0.8
VF
0.86 0.984
V
VM linearly
1.016 1.19
(A1)/A
m(Q) 0
hyper.
0.651
0.182
linearly-hyper.
10.34
hyper.
m(F)
1 m(F)
m(Q)
0.096
-hyper.
D
0
linearly-hyper.
1
1
:
mQ 0:096
Also, we check the cross ratios for the rest characteristic points VT, V = A
mT 0 VT VM 1 0 1:464 0:86 1 0:096 mQ;
mA 0 A VM 1 0 1:016 0:86 1 10:33 mF
1
:
mQ
Therefore, we may accept the value m(Q) as a scale value. So, all the characteristic points are evaluated via this scale, as it is shown by the axis m in Fig. 14.7.
So, a running regime point V (for example, V = 0.8) is expressed by the following cross-ratio:
mV 0 V VM 1
V
V
M
V 1 VM 1
0:8
0:86
4 6:142 0:651:
0:8 1 0:86 1
14:12
14.3
399
The corresponding hyperbolic distance from the maximum load power point
HV LnmV 0:428:
Similarly, the scale hyperbolic distance
HQ LnmQ Ln 0:096 2:343 LnmF Ln 10:33:
Then, the normalized distance or relative deviation from the maximum load
power point
D
HV
0:182:
HQ
14:13
m0 V
:
1V
A
14:14
400
14.4
14
Let us return to Fig. 14.5. Two lines 1 for the corresponding areas V VOC,
V VOC we are replacing by one hyperbola 2 shown in Fig. 14.8a. In this case,
initial Eq. (14.2) obtain the following view
Ri
ROC
i
2
OC 2
1
Vi 2
r
2
iOC r 2 ;
OC
A V
Ri
14:15
2 :
AI SC
A2 r 2 V =rAV OC
14:16
The plot of this expression has the typical view in Fig. 14.8b.
If to introduce the coordinate system ID 0GS VGS, this curve is close to the typical
transfer characteristic ID(VGS) of MOSFET transistors and its approximation by a
hyperbolic tangent [6]. The values ICD, VCGS correspond to the cusp of this curve.
Fig. 14.8 Symmetrical load
characteristic. a Internal
resistance via its voltage.
b Current of the internal
resistance via its voltage and
transfer characteristic.
c Current via internal
conductivity
Ri
(a)
2
Ri
0
(b)
IDM
OC
ID
-V
OC
2V
OC
Yi
OC
yi
V
I
IM
SC
0
V
0DS
Vi
ID
SC
OC
OC
AI
1
OC
ri
(c)
SC
Ri
VGS
OC
VGS
Vi
14.4
401
Also, it is possible to note that expression (14.16) is the particular case of Rapps
model of a solid-state microwave power amplier [8, 19].
If to express the current I through the internal conductivity Yi = 1/Ri, yOC
= 1/rOC
i
i ,
the circle equation turns out
2
2
Yi
I
1:
14:17
AI SC
yOC
i
The plot of this circle is shown in Fig. 14.8c.
14.5
(a)
14:18
I
(c)
Ri
IM
Ri
Yi
Vi
Vi
YiM
IM
(b)
IDM
C
ID
ID
I
+
IM
I
Vi
0
0DS
IM
VGS
VGS
402
14
On the other hand, for the coordinates of point M, the following orthogonal
transformation is known [10]
" #
x
y
"
cos a
sin k
sin a
cos a
# "
Vi
Ri
14:19
14:20
a12 Vi2
a11 Vi2
I
0:
2
a22 Vi a33
a22 Vi2 a33
14:21
The plot of this expression has the typical view in Fig. 14.9b with different
values of the maximum currents I+M, IM.
If to express the current I through the internal conductivity Yi, the general circle
equation turns out
a11 2a12 I a22 I 2 a33 Yi2 0:
From here, we get the explicit circle equation
2
2
k
Y 2 I IC 1 IC ;
cos a sin a i
14:22
where
cos2 a sin2 a 1 1
tga :
I
2 cos a sin a
2 tga
C
IM
tga:
In turn, the current IC conforms to the center of the segment I+MIM and determines
the maximum value YiM.
14.5
403
If asymmetrical curve (14.21) is used for approximation of the transfer characteristic ID(VGS) of MOSFET transistor, it is necessary, in the initial coordinate
system ID 0GS VGS, to restore the coordinate system I 0 Vi. To do this, we may use
the following property of the tangent line into the cusp ICD, VCGS of the asymmetrical
characteristic; this line passes through the origin of the coordinate system I 0 Vi.
This property follows from circle (14.22).
We note, if = 450, the above symmetrical curve is realized. In our case,
expression (14.21) contains two parameters as k, , which realize comprehensive
facilities for approximation.
We get still more facilities, if to use a nonrectangular coordinate system. Then,
similar expression (14.21) will contain three parameters.
Case 2
We may consider the known hyperbola Ri(Vi) in Fig. 14.10a and rewrite Eq. (14.18)
for the coordinate system y0x
y k=x:
14:23
On the other hand, for the coordinates of point M, the following transformation
takes place:
" #
x
y
" #
Vi
0
:
1
Ri
1= cos a
tga
k cos a
:
Vi
14:24
Vi
;
Vi2 tga k cos a
Vi2
:
Vi2 tga k cos a
1
p ;
2 k sin a
ViM
s
k cos a
1
tga
2YiM tga
1
;
2YiM Vi
k cos a
ViM
:
2YiM
14:25
404
14
(b)
(a)
I
1
Ri
0.5
Ri
y
Vi
Yi
x
0
0.5
Vi
Vi
Vi
(c)
I
1
0.5
Vi
0
0.5
Vi
Vi
Yi
;
YiM Vi =ViM 2 1
I
Vi =ViM 2
;
IM Vi =ViM 2 1
where IM = 2I(YiM) = 2YiM ViM. The plots of these curves are shown in Fig. 14.10b
for the normalized values Yi, Vi, I.
If to express the current I through the internal conductivity Yi, the explicit circle
equation turns out
Yi
2YiM
2
I
1
IM 2
2
1
:
4
The plot of this circle is shown in Fig. 14.10c for the normalized values. In turn,
~i are resulted by the known stereographic projection.
the voltages Vi ; V
The considered dependences are similar to the regulation characteristic of the
above voltage regulators in Chap.10 that gives a certain physical meaning to the
parameters of the approximation curves.
Also, these dependences coincide with Salehs models of a modulation characteristic for traveling-wave tube [11, 21].
14.5
405
14.6
406
14
1
150
V, Volt 100
50
0.5
0
0.75 1 SUN
8
10
12
14
I, A
determines the maximum load power point PM. Then, area 3 corresponds to a
current source with nearly horizontal straight line.
Hence, the inverse similarity follows with the curves in Fig. 14.3b; that is, the
inclined area 1 in Fig. 14.11 corresponds to area 3 in Fig. 14.3b. Then, using
Fig. 14.4, it is possible to draw up a dual equivalent generator in Fig. 14.12.
The current generator ISC and the internal nonlinear conductivity Yi1 dene a
hyperbolic component, and the linear conductance Yi2 denes a linear component of
the load voltage. We may write the equation immediately, using expression (14.8)
and changing formally currents I by voltages V and resistances by conductivities
VI V1 I V2 I V1OC A
1 I=I SC
V2OC 1 I=I SC ;
A I=I SC
14:26
SC OC
where OC voltage for the hyperbolic component is VOC
1 = I /Yi1 , andfor the linear
OC
SC
component is V2 = I /Yi2.
Similarly to Fig. 14.5, we may propose the nonlinear conductivity Yi1 dependence on the load current I in Fig. 14.13 for the dual active two-pole.
XS
V
Yi1
I
V1
SC
V
RL
Yi2
OC
V2
OC
AV 1 +V2
OC
S
OC
V1
OC
V2
V1
V2
AI
I
SC
SC
YS
14.6
407
Yi1
OC
Yi1
3
SC
Yi1
SC
AI
SC
Fig. 14.13 Internal conductivity dependence on the load current of the typical active two-port. 1
With self-limitation of the load voltage. 2 Real current source. 3 Ideal voltage source
OC
For the given curves in Fig. 14.11, it is need to nd the settings A, VOC
1 , and V2
using SC, OC points and two points a, b points and two points a, b, which are on
the each side near to the maximum load power point (a working area) with Ia, Va
and Ib, Vb coordinates.
Let us give the necessary design parameters
aI a
;
ab
Vb
Ib
;
Ia Ib
V OC
Va
Vb
; b12
;
b12
1 Ia 1 Ib
1
b12
; V2OC V OC V1OC :
1
1
A Aia Ai
b
b 1I
V1OC
AI
RL V2OC =I SC
RL V2OC =I SC
14:27
The maximum power point parameters are found from the cubic equation
2I 3 1 3A AuI 2 2A1 AuI A2 u 0;
OC
where u = 1 + VOC
1 /V2 .
408
14
mkT Iph I I0
Ln
;
q
I0
14:28
where RS is the series resistance of the cell, the values m, k, T, q are the diode factor,
Boltzmann constant, temperature, electron charge correspondingly; the values Iph,
I0 are the photocurrent and reverse saturation current.
For the given values, this expression takes the view
V 0:9I
1
13:615 I 0:0081
Ln
:
0:042
0:0081
14:29
14:30
The curves V(I) under expressions (14.29), (14.30) are shown in Fig. 14.14.
Examination of these curves shows that proposed model simulates a solar cell
characteristic well for practical tasks. The found errors of the working area (for
example, from Ib to Ia) are less than 2 %.
In turn, the curves of the power P = IV are shown in Fig. 14.15.
Then, if R = 10.92, we get from (14.27)
14.6
409
(a)
200
150
b
I P
M
V, Volt 100
50
10
12
14
I, A
(b)
140
130
V, Volt
PM
120
110
100
10
11
I, A
12
13
1500
1250
1000
P, W 750
500
250
0
10
12
14
I, A
I 2 26:97I 177:2 0:
The solution gives IM = 11.33. Using (14.30), we are nding VM = 123.36.
Therefore, this power is equal to 1398 and coincides with the maximum load
power.
410
14
References
1. Akbaba, M., Alattavi, M.: A new model for I-V characteristic of solar cell generators and its
applications. Solar Energy Mater. Solar Cells 37, 123132 (1995)
2. Alghuwainem, S.: A close-form solution for the maximum-power operating point of a solar
cell array. Solar Energy Mater. Solar Cells 46(3), 249257 (1997)
3. Ang, S., Oliva, A.: Power-switching converters, 2nd ed., Taylor & Francis Group (2005)
4. Cubas, J., Pindado, S., Victoria, M.: On the analytical approach for modeling photovoltaic
systems behavior. J. Power Sources 247, 467474 (2014)
5. Das, A.: Analytical derivation of explicit J-V model of a solar cell from physics based implicit
model. Sol. Energy 86(1), 2630 (2012)
6. Diakov, V.P., Maximchuk, A.A., Remnev, A.M., Smerdov, V.Y.: Entsiklopedia ustroistv na
polevyh transistorah (Encyclopedia of equipments on eld transistors). Solon-R, Moskva
(2002)
7. El Tayyan, A.: An approach to extract the parameters of solar cells from their illuminated I-V
curves using the Lambert W function. Turkish J. Phys. 39(1), 115 (2015)
8. Eltholth, A., Mekhail, A., Elshirbini, A., Dessouki, M., Abdelfattah, A.: A new- trend modelbased to solve the peak power problems in OFDM systems. Model. Simul. Eng. 2008, 5
(2008)
9. Hui, S.R., Chung, H.: Resonant and soft-switching converters. In: Power Electronics Hand
book, pp. 405449. Elsevier Inc. (2006)
10. Korn, G.A., Korn, T.M.: Mathematical Handbook for Scientists and Engineers. MacGraw
Hill, New York (1968)
11. Lee, B.M., de Figueiredo, R.J.: Adaptive predistorters for linearization of high-power am
pliers in OFDM wireless communications. Circ. Syst. Signal Proc. 25(1), 5980 (2006)
12. Ma, J., Man, K., Zhang, N., Guan, S., Wong, P., Lim, E., Lei, C.: Improving
power-conversion efciency via a hybrid MPPT approach for photovoltaic systems.
Elektronika ir Elektrotechnika 19(7), 5760 (2013)
13. Penin, A.: The voltage-current characteristics of an active two-pole with self-limitation of
current. Elektrichestvo 7, 5460 (2008)
14. Penin, A., Semionov, A.: Method of regulated resonance DC-DC voltage conversion.
Europ. patent EP 1 504 517 B1. 05 Nov 2008
15. Penin, A.: A quasi-resonance voltage converter with improved parameters. Elektrichestvo 2,
5864 (2009)
References
411
16. Penin, A., Sidorenko, A.: A convenient model for IV characteristic of a solar cell generator as
an active two-pole with self-limitation of current. World Acad. Sci. Eng. Technol. 3(4), 905
909 (2009)
17. Penin, A., Sidorenko, A.: Determination of deviation from the maximum power regime of a
photovoltaic module. Moldavian J. Phys. Sci. 9(2), 191198 (2010)
18. Penin, A.: An empirical validation of the linear-hyperbolic approximation of the IV
characteristic of a solar cell generator. World Acad. Sci. Eng. Technol. 5(3), 13631368
(2011)
19. Rapp, C.: Effects of HPA-Nonlinearity on a 4-DPSK/OFDM-signal for a digital sound
broadcasting system. In: In ESA, Second European Conference on Satellite Communications
(ECSC-2) pp. 179184 (SEE N92-15210 06-32), vol. 1, pp. 179-184 (1991)
20. Rekioua, D., Matagne, E.: Optimization of Photovoltaic Power System. Modelization,
Simulation and Control. Spriger, London (2012)
21. Saleh, A.: Frequency-independent and frequency-dependent nonlinear models of TWT
ampliers. IEEE Trans. Comm. 29(11), 17151720 (1981)
Conclusions
In this book, the features of an electric circuit with changeable operating regimes
are considered. The reader will agree that for such a circuit it is important to
determine the parameters of the running regime in the normalized form using the
characteristic parameter values as scales, to receive the equivalent of this circuit.
But there are two problems.
On the one hand, the actual regime parameters of the load resistance, current,
voltage result in various values of the corresponding normalized quantities. The
hands-on experience as if agrees with that.
On the other hand, interference of loads or any resistance on the load regime
leads to change of scales.
The offered approach to interpretation of regime changes as projective transformation allowed connecting regime parameters in one system and to consider
these changes through an invariant value in the form of the cross ratio of four
points. The reader will agree that the adequate mathematical model of regime
behavior turned out and we obtain the basis for research of such circuits and
introduction of necessary concepts. Naturally, the presented results are only the
beginning of researches. In particular, it is possible to apply such approach to
alternating current circuits.
If we look more widely, it is possible to speak about representation of flowed
processes of the different physical nature, using known electromechanical analogy.
Also, the presented approach is being applied for a long time in other scientic
domains, as mechanics, biology, and so on; for example, it is possible to see the
following publications:
1. Vaseashta, A.K., Penin, A., Sidorenko, A.: On the analogy of non-euclidean
geometry of human body with electrical networks. Int. J. Electr. Comput. Eng.
4(3). 378388 (2014)
2. Stakhov, A.P.: The mathematics of harmony: From Euclid to contemporary
mathematics and computer science. Series on knots and everything, vol. 22,
World Scientic Publishing Co. Pte. Ltd. (2009)
413
Index
A
Active, 276, 279, 284
multi-port, 193, 282
three-port, 210, 282
two-pole, 3, 9, 1113, 17, 29, 30, 34, 35,
42, 56, 60, 70, 82, 83, 86, 88, 98, 184,
190, 276278, 389, 390, 392, 394, 406
two-port, 9, 14, 87, 88, 167, 168, 175, 207,
224, 244, 275, 276, 279, 281, 407
Additive property, 41, 262, 263
Afne transformation, 29, 30, 32, 9799, 102,
241
Approximation, 20, 393, 395, 400, 403405,
408, 410
Asymmetrical load characteristic, 401, 404
Attenuation coefcient, 18, 103, 376
Auxiliary load, 55, 70, 87
B
Balanced network, 190, 196, 263
Base values, 911, 14, 39, 61, 62, 75, 178, 186,
191, 194, 195, 216
Boost converter, 24, 337, 339, 343, 348,
354356
Buck converter, 23
Bunch center, 38, 55, 57, 66, 71, 79, 8486, 89,
174, 176, 182, 188, 197, 199, 200, 254,
256, 277, 281, 285, 370, 371
C
Cascaded connection, 9, 10, 99, 100, 109, 110
Changeable resistance, 9, 55, 58
Characteristic
admittance, 18, 103, 376
regimes, 1, 4, 9, 12, 20, 21, 32, 39, 53, 168,
177, 181, 188, 199, 238, 292, 340, 342,
351, 352, 392
415
416
G
Generalized equivalent circuit, 57, 71, 84, 279
Group property, 40, 63, 69, 76, 109, 297
H
Half
plane, 328, 329, 331, 332
rounds, 328, 330, 331
Harmonic conjugacy, 48, 52, 128
Homogeneous coordinates, 38, 174, 179, 182,
186, 189, 197, 201, 204, 214, 221, 245,
246, 253, 265, 267, 272
Hyperbolic geometry model, 295, 296, 331
I
Increments, 3, 1517, 22, 63, 76
Innitely remote, 38, 48, 50, 133, 136, 174,
203, 248, 257, 295, 328, 365, 371, 378
Influence of loads, 4, 279
Initial regime, 2, 9, 11, 3234, 36, 40, 41, 42,
45, 61, 63, 6769, 75, 76, 80, 81, 90,
98, 101, 107, 114, 169, 172, 196, 197,
200, 232, 240, 243, 255, 271, 296298,
307, 321, 333, 345, 356, 369, 380, 386
InputOutput Conformity, 97, 102, 237, 244,
263
Intermediate change, 8, 16, 207
Internal resistance, 3, 4, 6, 10, 11, 21, 29, 35,
42, 44, 55, 56, 58, 59, 67, 84, 141,
173175, 289, 290, 313, 349, 370, 383,
392394, 400, 401, 404, 405
Invariant, 3032, 34, 35, 37, 39, 55, 6163, 76,
99, 105, 106, 108, 109, 112, 116, 142,
150, 159, 241, 247, 253, 296, 315, 340
L
Limited capacity voltage
source, 21, 22, 141, 337, 361
Linearization, 337, 343, 344, 354, 356
Linearly hyperbolic, 398, 405, 410
Linear stabilization, 11, 228
Load
conductivities, 14, 113, 173175, 182, 190,
209, 244, 250, 251, 268, 273, 275, 282
power, 1, 3, 5, 6, 21, 39, 47, 48, 51, 102,
121, 131135, 289, 290, 361, 362, 372,
389, 392397, 399, 405410
resistances, 4, 1517, 313, 323
straight lines, 9, 11, 14, 4143, 45, 5760,
66, 71, 73, 74, 79, 8385, 89, 98100,
102, 103, 109, 110, 168, 173175, 224,
238, 245, 276, 277, 381, 382, 389
Loss resistance, 23, 91, 338, 355, 374, 393
Index
M
Maximum efciency, 1, 108, 120, 122, 124,
128, 377, 380
Maximum load power, 1, 3, 6, 21, 39, 48, 51,
102, 389, 392, 393, 395, 396, 397, 399,
405, 407, 409, 410
Measured input current, 118, 260
Minimum load resistance, 315, 320, 327
Modular connection, 126, 127, 128, 130
Moving of point, 294, 295, 299, 301, 302, 304,
309311
N
Negative resistance, 91, 92, 315, 316
Nonlinear
load characteristics, 389, 392
regulation curve, 337
Non-uniform coordinates, 174, 178, 181, 183,
186, 188, 207, 210, 212, 221, 245, 246,
250252, 260, 268, 273
Normalized values, 5, 1114, 32, 55, 131,
149151, 157, 158, 297, 299, 305, 317,
319, 321, 328, 366, 368370, 404
Norton equivalent circuit, 44, 70, 71, 78, 79,
84, 88, 89
O
One-sheeted hyperboloid, 324
One-valued, 21, 361, 363
Open circuit, 2, 3, 29, 31, 32, 39, 55, 56, 70,
83, 97, 141, 147, 168, 169, 172, 177,
186, 213, 238, 239, 256, 324, 389
P
Parallel connection, 127, 128
Paralleling, 4, 141143, 153
Parametric stabilization, 91
Passive multi-port, 237
Polar, 50, 52, 53, 121, 123, 124, 133, 136, 365,
371, 378, 381, 396398
Pole, 50, 52, 53, 121123, 133, 136
Power-load elements, 17, 21, 361, 363
Power-source elements, 17, 21, 361363
Power supply, 1, 2, 21, 22, 55, 91, 92, 94, 116,
131, 141, 183, 184, 195, 198, 289, 290,
296, 313, 314, 324, 325, 337, 339, 361,
362, 364, 381, 405
Power transfer ratio, 5, 17, 19, 120
Projection center, 32, 61
Projective
plane, 173, 174, 221
transformation, 29, 37, 38, 39, 41, 61, 102,
104, 105, 109, 142, 145, 149, 154, 155,
Index
417
157, 196, 244, 246, 247, 253, 294, 296,
307, 316
Q
Quasi-resonant, 17, 20, 389392
R
Recalculation formula, 5, 6, 40, 44, 46, 99,
101, 196, 259, 260
Reference triangle, 174, 177, 179, 200, 214,
245247, 253, 264
Regime change, 2, 3, 711, 23, 24, 35, 37,
3943, 45, 46, 62, 63, 65, 68, 69, 76,
78, 81, 82, 90, 91, 99, 100, 102, 107,
115, 252, 259, 296, 297, 299302, 307,
317319, 321, 330, 331, 333, 334,
340342, 346, 347, 351, 353, 366369
Regime symmetry, 47, 121, 123
Regulated converter, 361
Regulation, 22, 290
characteristic, 21, 2325, 289, 296, 297,
337340, 343, 349, 350, 354, 356, 404
of load voltages, 289
Relative form, 1, 3, 4, 7, 11, 19, 23, 25, 39, 47,
146, 155, 175, 180, 181, 188, 201, 202,
278, 361, 365, 371, 404
Running regime, 4, 5, 29, 31, 35, 39, 52, 53,
62, 98, 122, 124, 135, 138, 146, 149,
157, 168, 170, 173, 174, 178, 182, 186,
189, 196, 201, 238, 244, 246, 253, 264,
281, 296, 342, 353, 398
S
Scales, 2, 46, 9, 14, 20, 29, 32, 44, 47, 55, 63,
76, 173, 182, 188, 278, 364, 396
Self-limitation, 389, 391, 392, 394, 407
Short circuit, 2, 3, 29, 31, 32, 39, 55, 56, 70,
83, 88, 97, 142, 167, 168, 177, 185,
186, 213, 238, 254, 256, 276, 279, 281,
283, 389
Six-port, 193, 263, 264
Solar Cell, 17, 20, 289, 392, 405, 406, 408, 410
Space, 41
Stabilization, 11, 13, 9193, 198, 199, 313, 322
Static characteristic, 1
Stereographic projection, 290, 291, 313, 314,
326, 340, 351, 355, 404