Synchronized Phasor Measurements Based Power System Dynamic State Estimation

International Electrical Engineering Journal (IEEJ)
Vol. 6 (2015) No.8, pp. 2018-2023

ISSN 2078-2365
http://www.ieejournal.com/
Synchronized Phasor Measurements

Based Power System Dynamic State
Estimation
B.PhaniRanga Raja, B.Ramesh
V R Siddhartha Engineering College (Autonomous)
phaniboyina@gmail.com
Abstract The dynamic state estimation (DSE) applied to power

systems with synchrophasor measurements would estimate the
systems true state based on measurements and predictions. In
this application, as phasor measurement units (PMUs) are not
deployed at all power system buses; state predictions would
enhance the redundancy of DSE input data. The importance of
predicted and measured data in DSE is affected by their
confidence levels, which are inversely proportional to the
corresponding variances. In practical cases, power system states
may undergo drastic changes during hourly load fluctuations,
component outages, or network switchings. In such conditions,
the inclusion of predicted values could degrade the power
system state estimation. The proposed model accommodates
system wide synchronized measurements of PMUs, which could
be of interest to smart grid applications in energy management
systems. In this paper phasor measurement unit (PMUs) is
modelled in Simulink in MATLAB. In this paper, a new state
estimation method based on the extended weighted least squares
(WLS) method and Newton Raphson state estimation employing
Jacobin matrix for considering both measurement errors and
model inaccuracy. Simulation studies are conducted on an
illustrative IEEE three-bus system based on measurement
values obtained by PMUS.
Keywords:
State estimation (SE), weighted least square method, phasor
measurement unit, dynamic state estimation
Introduction
State estimation (SE) computes the best estimate of the true
operating state of a power system as a necessity for an
efficient and reliable operation [1]. Since SE serves several
crucial functions such as the detection of abnormal
conditions, generator correction actions, and contingency
analysis [2], its responsibility is increasing with the dimension
of power systems and emerging uncertain conditions in the
operation of restructured systems.
In general, SE is categorized into the following three
types: 1) static SE (SSE); 2) tracking SE (TSE); and 3)
Raja and Ramesh
dynamic SE (DSE) [3]. SSE estimates the power system state

associated with a given time and based on the measurement
set corresponding to that moment of time. The SSE algorithm
is iterative that is initialized as a flat start. Hence, it uses heavy
computations and cannot be executed in short intervals. To
overcome this problem, TSE was introduced in which the
estimation starts from the last calculated state variables
instead of a flat point.
In both SSE and TSE, the power system state is estimated
based on a single set of measurements. A most advanced state
estimation is DSE i.e., possesses the ability to predict the
power system state progressively in short time steps.
Furthermore, at each time step, the estimation uses both
measurement and prediction data sets. DSE techniques and
models along with the results of DSE implementation in
practical power systems have been presented in [4] and [5]
and other reviews of DSE techniques are available in [6] and
[7].
A new generation of power system monitoring
schemes embedded in the wide-area measurement system
(WAMS) is enabled by increasing the deployment of phasor
measurement units (PMUs) [8]. PMUs directly measure the
state variables, i.e., magnitude and phase angle of bus
voltages, with a very high accuracy. They can consequently
increase the robustness and precision of the estimation
process. The advantages of WAMS over the conventional
metering arrangement are as follows [9].
1) PMUs are installed with specific guidelines, tested during
the assigning process, and calibrated periodically.
2) PMU devices are furnished with advanced computation
algorithms and self-check/self-diagnostic capabilities. In
addition, PMUs are equipped with 16 bit or upper A/D
converters providing an extra high sampling rate and very
accurate measurements.
The integration of PMU measurements in the conventional
DSE was discussed in [10][15]. The Electric Power
Research Institute had a research project relevant to this
PMUs [16]. To avoid deficits associated with the
2018
Synchronized Phasor Measurements Based Power System Dynamic State Estimation

Vol. 6 (2015) No.8, pp. 2018-2023
ISSN 2078-2365
conventional metering system such as asynchronous
measurement and time skew errors, a plenary set of WAMS
are intended in future power systems.
The most common DSE algorithms are variants of weighted
least squares (WLSs) estimation. These models are solved
using iterative matrix-based calculations along with
mathematical heuristics to reduce the computation.
This paper presents the Weighted Least Square method
(WLS) and Newton Raphson State estimation methods.
Hence, the power system state variables would be calculated
based on PMU measurements at adjacent buses. The
calculated confidence level would be affected by
uncertainties associated with PMU measurements. An
illustrative example and the IEEE 3-bus test system are
examined in this paper for investigating performance of the
proposed DSE model.
II Modelling of PMUs
Synchronized phasor measurement units (PMUs) were first
introduced in early 1980s, and since then have become a
mature technology with many applications which are
currently under development around the world. The
occurrence of major blackouts in many major power systems
around the world has given a new impetus for large-scale
implementation of wide-area measurement systems (WAMS)
using PMUs. Data provided by the PMUs are very accurate
and enable system analysts to determine the exact sequence of
events which have led to the blackouts, and help analyze the
sequence of events which helps to identify the exact causes
and malfunctions that may have contributed to the terrible
failure of the power system.
The fundamental feature of PMU is that it measures positive
sequence (and negative and zero sequence quantities, if
needed) voltages and currents of a power system in real time
with precise time synchronization.
Figure 1 shows synchronized phasor measurement system
configuration. The GPS transmission is received by the
receiver section, which delivers a phase-locked sampling
clock pulse to the Analog-to-Digital converter system. The
sampled data are converted to a complex number which
represents the phasor of the sampled waveform. Phasors of
the three phases are combined to produce the positive
sequence measurement.
The basic Phasor measurement unit is that of estimating a

positive-sequence (also negative and zero are available),
fundamental frequency phasor representation from voltage or
current waveforms. As indicated by Fig. 1, the analog power
signal is converted into digital data by the analog to digital
converter. For example, if the voltage is needed to be
measured, the samples are taken for each cycle of the
waveform and then the fundamental frequency component is
calculated using (DFT). The positive sequence phasor can be
calculated as follows;
V=1/3 (Va + Vb + 2Vc)
(1)
Figure 2 shows a simple block diagram explaining the
procedure of measured voltage or current analog signal. The
external time source is an absolute time reference from a
global positioning system (GPS) receiver, which delivers a
phase-locked sampling
clock pulse to the Analog-to-Digital converter system. The
sampled data are converted to a complex number which
represents the phasor of the sampled waveform. Phasors of
the three phases are combined to produce the positive
sequence measurement.
The figure includes a hardware low-pass filter (LPF)
for anti-aliasing and an analog-to-digital (A/D) converter for
analog-to-digital conversion. The system permits capturing
records of the same event at different points in the power
system with a unique time reference, the phasor measurement
units at present are located strategically, with the purpose of
capturing information on the impact of contingencies at the
local or system level.
2019
Raja and Ramesh

Vol. 6 (2015) No.8, pp. 2018-2023
ISSN 2078-2365
function, and it needs only one iteration to converge.
Alternatively, the solution to (3) can be obtained as follows:
= (T R-1 )-1 T -1
(5)
The algorithm for WLS state estimation problem can be

outlined as follows:
a) Start iterations, Set the iteration index A=0.
b) Initialize the state vector X, typically as a flat start.
c) Calculate the gain matrix, x.
d) Calculate the right hand side = (T-1)-1T-1
e) Decompose x and solve for x
f) Test for convergence
g) If no, update k=k+1, and go to step c. Else, stop.
Fig 2 simulation diagram of PMU
IV. Newton Raphson state estimation:
III. Weighted least square method (WLS)
The least square estimate is referred as the ordinary least
square estimate and is obtained by minimizing the index
function that puts equal weightage to the errors of estimation.
In WLS, put different weightages on the different components
since some of the measurements may be more reliable and
accuracy than the least square. For this we define
measurement equation as follows:
Z=h(x) +e
(2)
Where:
Z is the (m x 1) measurement vector;
x is the an (n x 1) state vector to be estimated;
h is a vector of nonlinear functions that relate the states to the
measurements; and
e is an (m x 1) measurement error vector.
It is necessary that m n and the Jacobian matrix of
h(x) has rank n.
The optimal state estimate vector x may be determined by
minimizing the sum of weighted squares of residuals
Min () = [ h()]T-1[ h()]
(3)
Where,
R is the measurement covariance matrix.
(3) is linearized using the Taylor series expansion, retaining
the first two terms and ignoring higher-order terms. This leads
to a linear WLS problem having the solution.
= (T-1)-1T-1
(4)
Where, H is the Jacobian matrix of h(x).
Newton-Raphson State Estimator (NRSE) method [19][20]

that was subsequently introduced became more popular
because of exact problem formulation and very good
convergence characteristic. In NRSE method, elements of
Jacobian matrix are computed from the standard expressions
which are functions of bus voltages, bus powers and the
elements of bus admittance matrix. Nowadays, with the
introduction of fast computers, even huge amount of complex
calculations can be carried out very efficiently in much lesser
time. Therefore, there is no need to go for approximate
models. In this paper, an attempt is made to introduce more
physical meaning for the elements of the SE Jacobian matrix
H. Bus admittance matrix of transmission network does not
find place in computing the elements of the H matrix. The
power flows in the network elements are taken as the basic
components in constructing the H matrix. Network elements
are added one-by-one and the H matrix is updated in a simple
manner. Resulting final H matrix is exactly same as that
obtained in NRSE method.
V.Power flow in transmission network elements:
In NRSE method the transmission network is represented by
the bus admittance matrix and the elements of the H matrix
are computed using the elements of bus admittance matrix.
Alternatively, in this paper, the elements of the H matrix are
obtained considering the power flows in the transmission
network elements.
Consider the general transmission network between buses i
and j are as follows in fig 3.
The iterative approach is applied to obtain the state update

until the absolute value of the difference of the states between
successive iterations is less than the tolerance value ,
typically, is set to 1e-4. The function h(x) can be linear or
nonlinear. H is a constant matrix (m n) in the case of linear
2020
Raja and Ramesh

Vol. 6 (2015) No.8, pp. 2018-2023
ISSN 2078-2365
Fig 3 Transmission network element between buses i and j

The transmission line is represented by the series impedance
Rij+jXij or by admittance of Gij+jBij . For such a general
transmission network element, the real and reactive power
flows are given by the following expressions are
(6)
(7)
(8)
(9)
Where
(10)
(6) to (9) are the real and reactive power flows for the
transmission network of fig 3. The jacobian matrix H can be
calculated from the partial derivatives of available
measurements with respect to and V. The elements of
jacobian matrix H are formulated by partial derivatives pij, qij,
pji, qji with respect to
.
The algorithm for Newton Raphson state estimation problem
can be outlined as follows:
a) Read measurements from input Zi=hi(x)
b) Perform Ybus matrix of the system network.
c) Create the jacobian matrix H.
d) Calculate the power injection and power flows.
e) Read the weighting data for bus and line (R-1 or W)
f) Create the weight matrix
g) Calculate the Gain Gm=H*W*H.
h) Calculate the state mismatch vector Smm=inv
(gain)*H*W*mmf.
i) Calculate the difference between next and previous
state
j) Update X
Fig 4: flow chart for state estimation solution.
V. Simulation Studies
The three-bus power system is shown in Fig.5 and the
corresponding data are given in the Tables I and II. The PMU
located at bus 1 measures the voltage phasor of bus 1 and
current phasors of transmission lines 1 and 2. Accordingly,
the voltage phasors at buses 2 and 3 would be measured
indirectly. According to the power flow solution, unit1 would
generate 0.8 + j 0.25 (p.u. on the 100 MVA base) and the
remaining load plus transmission system losses are generated
by unit 2, which is located at the slack bus.
Fig 5 Three bus system

2021
Raja and Ramesh

Vol. 6 (2015) No.8, pp. 2018-2023
ISSN 2078-2365
The reference bus angle is not defined because of direct phase
measurement of PMUs, and state variables are considered as
{V1, 1, V2, 2, V3, 3}.The challenging situation for
measuring the performance of DSE is a sudden change in
system conditions for 24 time periods by changing the loads.
state variables results obtained through the proposed

algorithm for IEEE 3-bus system.
Table III: output obtained from the three bus system at
different loads from PMU (TRUE VALUES)
t
V3(p.u)
3(deg)
V3(p.u)
3(deg)
0.9806
-3.30
13
0.9753
-4.65
0.9801
-3.43
14
0.9756
-4.57
0.9796
-3.56
15
0.9762
-4.42
0.9743
-4.89
16
0.9812
-2.92
0.9749
-4.74
17
0.9816
-3.04
0.9746
-4.81
18
0.9813
-3.10
0.9740
-4.98
19
0.9816
-3.04
System data for three bus system:
0.9740
-4.98
20
0.9811
-3.16
Table I: Transmission line data:
0.9733
-5.14
21
0.9806
-3.29
Simulink block diagram of three bus system with PMU:
(a)
Line no
1
2
3
From
bus
1
1
2
To bus
R(p.u)
X(p.u)
G(p.u)
B(p.u)
10
0.9740
-4.97
22
0.9802
-3.41
2
3
3
0.05
0.02
0.02
0.2
0.15
0.15
0
0
0
0.025
0.02
0.02
11
0.9747
-4.81
23
0.9797
-3.54
12
0.9749
-4.73
24
0.9791
-3.67
Table IV: output obtained from WLS method (ESTIMATED

VALUES)
Table II: load data for different time periods:

t
P(p.u)
Q(p.u)
P(p.u)
Q(p.u)
P(p.u)
Q(p.u)
1.020
0.110
1.311
0.142
17
0.977
0.106
1.040
0.112
10
1.285
0.139
18
0.987
0.107
1.061
0.115
11
1.259
0.136
19
0.977
0.106
1.273
0.138
12
1.247
0.135
20
0.997
0.108
1.248
0.135
13
1.234
0.133
21
1.017
0.110
1.260
0.136
14
1.222
0.132
22
1.037
0.112
1.286
0.139
15
1.198
0.129
23
1.058
0.114
1.286
0.139
16
0.958
0.103
24
1.079
0.117
Simulation is carried out on three bus system in MATLAB

and the results coming from the three bus system taken as
measurement data from 3 bus and we have to estimate the best
suitable values from the measurement data obtain from PMU.
The MATLAB code based on WLS and NRSE
algorithms has been run in MATLAB environment. The
estimated voltage angles are the same for the two methods,
while the estimated power flows by the WLS method are
much closer to the NRSE method. Table III.IV, V shows the
V3(p.u)
3(deg)
V3(p.u)
3(deg)
0.9707
-3.28
13
0.9653
-4.60
0.9702
-3.40
14
0.9666
-4.53
0.9706
-3.50
15
0.9672
-4.40
0.9653
-4.83
16
0.9731
-2.90
0.9657
-4.70
17
0.9726
-3.01
0.9656
-4.79
18
0.9723
-3.09
0.9650
-4.95
19
0.9726
-3.01
0.9650
-4.95
20
0.9721
-3.12
0.9648
-5.09
21
0.9716
-3.24
2022
Raja and Ramesh

Vol. 6 (2015) No.8, pp. 2018-2023
ISSN 2078-2365
10
0.9650
-4.94
22
0.9712
-3.40
References:
11
0.9657
-4.79
23
0.9707
-3.50
12
0.9659
-4.70
24
0.9701
-3.62
[1] M. Shahidehpour, W. F. Tinney, and Y. Fu, Impact of security

on power systems operation, Proc. IEEE, vol. 93, no. 11, pp.
20132025, Nov. 2005.
[2] A. J. Wood and B. F. Wollenberg, Power Generation, Operation,
and Control, 2nd ed. New York, NY, USA: Wiley, 1996.
Table V: Output obtained

(ESTIMATED VALUES)
from
NRSE
Method
V3(p.u)
3(deg)
V3(p.u)
3(deg)
0.9806
-3.30
13
0.9753
-4.65
0.9801
-3.43
14
0.9756
-4.57
0.9796
-3.56
15
0.9762
-4.42
0.9743
-4.89
16
0.9812
-2.92
0.9749
-4.74
17
0.9816
-3.04
0.9746
-4.81
18
0.9813
-3.10
0.9740
-4.98
19
0.9816
-3.04
0.9740
-4.98
20
0.9811
-3.16
0.9733
-5.14
21
0.9806
-3.29
10
0.9740
-4.97
22
0.9802
-3.41
11
0.9747
-4.81
23
0.9797
-3.54
12
0.9749
-4.73
24
0.9791
-3.67
[3] A. M. Leite da Silva, M. B. Do Coutto Filho, and J. M. C.

Cantera, An efficient dynamic state estimation algorithm including
bad data processing, IEEE Trans. Power Syst., vol. 2, no. 4, pp.
10501058, Nov. 1987.
[4] M. B. Do Coutto Filho and J. C. Stacchini de Souza,
Forecasting-aided state estimationPart I: Panorama, IEEE
Trans. Power Syst., vol. 24, no. 4, pp. 16671677, Nov. 2009.
[5] M. B. Do Coutto Filho, J. C. S. de Souza, and R. S. Freund,
Forecasting-aided state estimationPart II: Implementation,
IEEE Trans. Power Syst., vol. 24, no. 4, pp. 16781685, Nov. 2009.
[6] N. R. Shivakumar and A. Jain, A review of power system
dynamic state estimation techniques, in Proc. IEEE Int. Conf.
Power Syst. Technol., Oct. 2008, pp. 16.
[7] A. Jain and N. R. Shivakumar, Power system tracking and
dynamic state estimation, in Proc. Power Syst. Conf. Expos., Mar.
2009, pp. 18.
[8] A. G. Phadke and R. M. de Moraes, The wide world of
wide-area measurement, IEEE Power Energy, vol. 6, no. 5, pp.
5265, Oct. 2008.
[9] G. Phadke and J. S. Thorp, Synchronized Phasor Measurements
and Their Applications. New York, NY, USA: Springer-Verlag,
2008.
[10] A. Jain and N. R. Shivakumar, Impact of PMUs in dynamic
state estimation of power systems, in Proc. 40th NAPS, Sep. 2008,
pp. 18.
[11] H. Xua, Q. Jia, N. Wang, Z. Bo, H. Wang, and H. Ma, A
dynamic state estimation method with PMU and SCADA
measurement for power systems, in Proc. IPEC, Dec. 2007, pp.
848853.
[12] N. R. Shivakumar and A. Jain, Including phasor
measurements in dynamic state estimation of power systems, in
Proc. Int. Conf. Power Syst. ACO, Mar. 2008, pp. 16.
[13] E. Farantatos, G. K. Stefopoulos, G. J. Cokkinides, and A. P.
Meliopoulos, PMU-based dynamic state estimation for electric
power systems, in Proc. Power Energy Soc. General Meeting, Jul.
2009, pp. 18.
VI. Conclusion:
This paper presents a new state estimation algorithm based on
the WLS method for accounting for both model errors and
measurement errors. For comparison purposes, both the WLS
method and the NRSE method are evaluated. The estimated
results evince that the WLS method may outperform the
NRSE when the power network model used is not accurate.
The NRSE method is best suited for cases with known
variances for model errors.
[14] A. Jain and N. R. Shivakumar, Phasor measurements in

dynamic state estimation of power systems, in Proc. IEEE Region
10th Conf. TENCON, Nov. 2008, pp. 16.
[15] Z. Huang, K. Schneider, and J. Nieplocha, Feasibility studies
of applying Kalman Filter techniques to power system dynamic state
estimation, in Proc. Int. Power Eng. Conf., Dec. 2007, pp.
376382.
[16] Chawasak Rakpenthai, Sermsak Uatrongjit and Issarachai
Ngamroo WLAV state estimation for power system containing
multitype FACTS devices IEEJ Transactions on Electrical and
Electronic Engineering Volume 8, Issue 3, pages 207214, May
2013.
2023
Raja and Ramesh

Synchronized Phasor Measurements Based Power System Dynamic State Estimation

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Synchronized Phasor Measurements Based Power System Dynamic State Estimation

Uploaded by

Copyright:

Available Formats

International Electrical Engineering Journal (IEEJ)

Vol. 6 (2015) No.8, pp. 2018-2023