IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 41, NO.

2, MARCH/APRIL 2005 507

Integrated High-Speed Intelligent Utility Tie Unit for
Disbursed/Renewable Generation Facilities
Worakarn Wongsaichua, Student Member, IEEE, Wei-Jen Lee, Senior Member, IEEE,
Soontorn Oraintara, Senior Member, IEEE, Chiman Kwan, Senior Member, IEEE, and Frank Zhang
AbstractFollowing the steps of the gas industry, the traditional
paradigmof the vertically integrated electrical utility structure has
begun to change. In the United States, the Federal Energy Regula-
tory Commission has issued several rules and Notices of Proposed
Rulemaking to set the road map for the utility deregulation. The
crisis in California has drawn great attention and sparked intense
discussion within the utility industry. One general conclusion is to
rejuvenate the idea of integrated resource planning and promote
the distributed generation via traditional or renewable generation
facilities for the deregulated utility systems. Fuel cell and photo-
voltaic are the most promising renewable generation technologies
for the residential and small commercial users. It is desirable for
these facilities to be interconnected with the utility grid to perform
peak shaving, demand reduction, and to serve as emergency and
standby power supply. However, the mismatch between the utility
tie protection and the equipment protection makes it impossible
for the fuel cell and/or photovoltaic to serve as emergency and
standby power supply when the utility supply is lost due to nearby
external faults. To overcome this issue, this paper discusses the de-
velopment of an integrated high-speed intelligent utility tie moni-
toring, control, and protection system to replace the traditional tie
breakers for those residential and small commercial facilities with
disbursed/renewable generation facilities.
Index TermsDistributed generation (DG), fault detection, fault
identication.
I. INTRODUCTION
U
NDER the restructuring phase of the electrical power in-
dustry, the traditional vertically integrated utility environ-
ment is inevitably being changed. The power system operation
will become more competitive and many challenges will arise
[1]. After experiencing the price hikes and rotating blackouts in
California, the disbursed or distributed generation (DG) via re-
newable generation facilities becomes one of the most attractive
alternatives for the future utility industry. In addition to the de-
velopment of large-scale wind generations at remote sites, fuel
cell and photovoltaic are the most promising technologies for
the urban residential and small commercial users. For economic
and reliability purposes, it is desirable for these facilities to be
Paper ICPSD-05-02, presented at the 2004 Industry Applications Society An-
nual Meeting, Seattle, WA, October 37, and approved for publication in the
IEEETRANSACTIONS ONINDUSTRYAPPLICATIONS by the Energy Systems Com-
mittee of the IEEE Industry Applications Society. Manuscript submitted for re-
view October 15, 2004 and released for publication January 27, 2005.
W. Wongsaichua and W.-J. Lee are with the Energy Systems Research
Center, The University of Texas at Arlington, Arlington, TX 76019 USA
(e-mail: worakarn_w@hotmail.com; lee@exchange.uta.edu).
S. Oraintara is with the Electrical Engineering Department, The University
of Texas at Arlington, Arlington, TX 76010 USA (e-mail: oraintar@uta.edu).
C. Kwan and F. Zhang are with Intelligent Automation, Inc., Rockville, MD
20855 USA (e-mail: ckwan@i-a-i.com).
Digital Object Identier 10.1109/TIA.2005.844856
Fig. 1. System block diagram.
Fig. 2. IUT hardware block diagram.
interconnected with the utility grid to performpeak shaving, de-
mand reduction, and to serve as emergency and standby power
supply [2]. However, the third desirable function cannot be ac-
complished due to the coordination mismatches on the protec-
tion between the local renewable power facility and the utility
supply when a nearby external fault occurs. Asimplied system
as shown in Fig. 1 is used to explain the above-mentioned issue.
Generally speaking, the fuel cell is equipped with a high-speed
power electronics disconnecting mechanism whereas the oper-
ation speed of the tie breaker is 10 cycles or more. If an external
fault happens, breaker B1 will be activated immediately to sep-
arate the fuel cell from the system. The breaker B2 will then
operate to isolate the fault. As a result, the user lost both power
supplies.
This paper discusses the development of an integrated high-
speed intelligent utility tie unit (IUT) to monitor, control, and
protect the system for residential and small commercial users to
enable their disbursed/renewable generation facilities to serve as
emergency and standby power supply. The proposed algorithm
of the IUT is veried with computer simulation.
II. IUT
The hardware block diagram of the proposed IUT is shown
in Fig. 2. It will replace the traditional tie breaker and act as a
0093-9994/$20.00 2005 IEEE
508 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 41, NO. 2, MARCH/APRIL 2005
smart coordinator between the fuel cell and the utility supply.
The operating principles of the IUT are described below.
A. Response for the External Fault
When an external fault occurs, the IUT will rst detect any
nearby external fault by analyzing the voltage and current sig-
nals. Then, it will initiate a trip signal immediately to separate
the utility supply and the local system. This will consequently
prevent the fuel cell from tripping. In the mean time, the circuit
breaker at upperstream will be activated to isolate the external
fault. The IUT will reconnect the customer back to the system
after the fault is cleared and the service is restored.
B. Response for the Internal Fault
When an internal fault occurs, the protection mechanism of
the fuel cell will operate to protect the fuel cell. The fuel cell can
be reconnected after the fault is cleared. In this case, the IUT
will perform the coordination tasks similar to the traditional tie
breaker.
C. Hardware Conguration
The design considerations of the four hardware subsystems
of the IUT, as shown in Fig. 2, are discussed below.
Voltage and Current Sensing Devices: High-speed Hall-ef-
fect sensors will be used for monitoring the system voltage and
current. The response time of these sensors is 520 s. High-
speed sensing is critical for this type of application because the
response of the IUT for the nearby external faults should be
faster than the built-in protection of the fuel cell. The target op-
eration speed of the IUT for the nearby external faults is 8 ms.
Signal Conditioning Circuit: The voltage and current signals
are contaminated during fault conditions. It is necessary to use a
signal conditioning circuit to extract desired information for fast
and accurate decision making. Analog lters will be employed
to lter out disturbances and measurement noise. Later, the sig-
nals will be ltered further through a digital ltering algorithm.
Embedded Control Processor: The embedded control pro-
cessor is the brain of the IUT. It acquires the signals from the
signal conditioning circuit, then analyzes and sends out the nec-
essary control signal for systemprotection. Aoating-point dig-
ital signal processor (DSP) TMS320C6713 is chosen to process
the data. The cycle time of this processor is 4.44 ns.
Tripping Unit: An insulated gate bipolar transistor (IGBT)
is selected as a tripping unit [4]. The turn-on and turn-off times
of the selected IGBT are 0.4 and 0.8 s, respectively. The IGBT
tripping unit, as shown in Fig. 3, acts as an ac switch. During
each half cycle, one IGBT and a freewheeling diode of the op-
posite IGBT are conducted. For example, in the positive half
cycle, the current ows through IGBT1 and the diode D2 to the
load. This ac switch is controlled by the gate-to-emitter voltage
of each IGBT.
In order to handle the short-circuit current of the internal
faults, two IGBT ac switches are connected in parallel with one
of them in series with a reactor to limit the fault current [5] as
depicted in Fig. 4. The IGBT switch 1 is normally on and the
IGBT switch 2 is normally off. For an external fault, the em-
bedded control processor sends a control signal to turn off IGBT
Fig. 3. AC switch using IGBT.
Fig. 4. Tripping unit.
switch 1 within a half cycle; IGBT switch 2 remains off. When
the external fault is cleared, IGBT switch 1 will be turned on
again. For an internal fault, the IGBT switch 1 is turned off im-
mediately by a short-circuit protection circuit and then the IGBT
switch 2 is turned on for 10 cycles to comply with the DG inter-
connection requirements and to improve the service continuity
of the customers in the case of momentary faults. When the in-
ternal fault is cleared, IGBTswitch 1 will be turned on and IGBT
switch 2 will be turned off.
D. Software Functions
The functions of the software in the embedded control pro-
cessor are to detect and determine the fault type as soon as pos-
sible. The amplitude and phase of a sinusoidal signal will be es-
timated for fault detection and identication by using weighted
least-square method [6][8] with the addition of decaying dc
component [9].
Weighted Least-Square Estimation: Assume that a sinu-
soidal signal with a decaying dc component in a power system
can be dened as
(1)
where and are the amplitude and time constant of the de-
caying dc component. , , and are the amplitude, frequency,
and phase of the sinusoidal signal.
By using Taylor series expansion to extract the rst two terms
of the decaying dc component, (1) can be rewritten as
(2)
WONGSAICHUA et al.: INTEGRATED HIGH-SPEED IUT FOR DISBURSED/RENEWABLE GENERATION FACILITIES 509
This equation can also be expressed in the matrix form as
(3)
where
(4)
(5)
By applying the weighted least-square error criterion
(6)
with a forgetting factor , the estimate of can be cal-
culated as
(7)
where
gain factor (8)
covariance matrix
(9)
The initial conditions are and ,
. The estimates of the amplitude and phase of the sinusoidal
signal can be computed as
(10)
(11)
where is the th element of .
The fault can be detected from the step change of the current
signal by using
(12)
where is the current signal, is the estimated of the cur-
rent signal, and is a threshold value.
After the fault is detected, the covariance matrix will be reset
to the initial condition [6]. The fault can be identied by using
the amplitude estimation of the current signal , the phase es-
timation of the current signal , and the phase estimation of the
voltage signal . The decision process is described as follows.
1) If , it is an internal fault.
2) If ,
a ) if , it is an external fault, and
b ) if , it is a normal condition.
III. SIMULATION RESULTS
A sample radial distribution system, as shown in Fig. 5, is
used to verify the developed algorithm. The system is simulated
by using the Alternative Transient Program (ATP) of the Elec-
tromagnetic Transient Program (EMTP) with sampling rate of
Fig. 5. System simulation in ATP-EMTP.
Fig. 6. Voltage and current in normal condition.
16 samples/cycle. The 7-kVA fuel cell is considered as a con-
stant current source of 58.33 A and the internal load is assumed
to be 10 kVA with 0.95 lagging power factor. The steady-state
system information is shown in Fig. 6.
The internal faults at F1 and external faults at F2F4 are sim-
ulated with different inception angles. Figs. 7 and 8 show some
of the voltages and the corresponding tie unit currents in one in-
ception angle when the fault happens at s.
As shown in the Figs. 7(a) and 8(a), the voltage drops to zero
while the current jumps up to 4956 A when a solid ground
fault happens at F1. The selected IGBT switch is not able to
tolerate this high fault current. Therefore, the parallel connec-
tion arrangement as described in Fig. 4 is necessary to protect
the IGBT. For the external fault at F2, the voltage has fallen to
93 V while the current has risen to 50 A in the neg-
ative cycle. The results for the external fault at F3 are similar
to those at F2. For the external fault at F4, the voltage has de-
creased to 2 V while the current has increased to 105 A
in the negative cycle. Due to the current-limiting capability of
the fuel cells inverter circuit, the current owing through the
510 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 41, NO. 2, MARCH/APRIL 2005
Fig. 7. Voltage prole for fault at F1F4. (a) Voltage prole for fault at F1. (b)
Voltage prole for fault at F2 or F3. (c) Voltage prole for fault at F4.
tie unit during the external faults is much smaller than the in-
ternal faults. One can also observe that there is a decaying dc
Fig. 8. Tie unit current for fault at F1F4. (a) Tie unit current for fault at F1.
(b) Tie unit current for fault at F2 or F3. (c) Tie unit current for fault at F4.
component when the fault took place at different angles of the
sine wave. In order to estimate the phase difference between the
WONGSAICHUA et al.: INTEGRATED HIGH-SPEED IUT FOR DISBURSED/RENEWABLE GENERATION FACILITIES 511
TABLE I
ESTIMATED AMPLITUDE OF CURRENT WITHOUT NOISE (A)
TABLE II
ESTIMATED PHASE DIFFERENCE WITHOUT NOISE (DEGREE)
voltage and current during fault, the pre-fault voltage signal has
been used as the reference.
The fault detection and identication algorithms with
, , and are implemented in Matlab with
the consideration of possible zero-mean unit-variance N(0,1)
White Gaussian Noise (WGN) and harmonics. The magnitude
of WGN is limited to 5% of the rated amplitude. The harmonics
are composed of 6% of the third harmonic and 4% of the fth
harmonic. The WGN and harmonics are blended with the re-
sults from ATP-EMTP simulations. Three types of input data:
1) internal fault at F1 (iut2aiut2h); 2) external fault at F2 or F3
TABLE III
ESTIMATED AMPLITUDE OF CURRENT WITH WGN (A)
TABLE IV
ESTIMATED PHASE DIFFERENCE WITH WGN (DEGREE)
(iut3aiut3h); and 3) external fault at F4 (iut4aiut4h) are used
for algorithm verication.
The simulation results are shown in Tables IVI. The shad-
owed area indicates the denite identication point of the esti-
mation scheme. One can see that the fault can be detected at the
moment it happened in every studied case.
By setting , the internal fault can be
identied with two samples delay for all cases as seen from
Tables I, III, and V. Similar results are obtained for external
faults (Tables II, IV, and VI). Since the actual signals are highly
contaminated during system faults, longer time delay may be
512 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 41, NO. 2, MARCH/APRIL 2005
TABLE V
ESTIMATED AMPLITUDE OF CURRENT WITH HARMONICS (A)
TABLE VI
ESTIMATED PHASE DIFFERENCE WITH HARMONICS (DEGREE)
needed to achieve more reliable fault identication. However,
high-speed sensing is critical for the proposed application
because the response of the IUT for the nearby external faults
should be faster than the built-in protection of the fuel cell, and
the tradeoff between speed, sensitivity, selectivity, and accuracy
may be needed for the fault identication.
IV. CONCLUSION
After experiencing the price hikes and rotating blackouts in
California, the disbursed or DG via renewable generation facil-
ities is considered as one of the most attractive alternatives for
the future utility industry. Fuel cell and photovoltaic are the most
promising technologies for the urban residential and small com-
mercial users. One of the major motivations for this type of in-
stallation is to improve the service continuity while the external
supply is not available. However, the current interconnection ar-
rangement has made it difcult, if not impossible, to achieve this
goal. This paper has discussed the development of an integrated
high-speed intelligent utility tie monitoring, control, and protec-
tion system that enables local distributed generations to serve as
emergency and standby power supply. The computer simulation
shows that the proposed development is able to perform correct
actions within the desired time frame.
REFERENCES
[1] C. Chompoo-inwai, W. J. Lee, P. Fuangfoo, M. Williams, and J. Liao,
System impact study for the interconnection of wind generation and
utility system, in Proc. IEEE I&CPS Tech. Conf., 2004, pp. 2732.
[2] IEEE Recommended Practice for Emergency and Standby Power Sys-
tems for Industrial and Commercial Applications, IEEE Standard 446,
1995.
[3] Distributed Generation Interconnection Manual, Public Utility Com-
mission of Texas, Austin, TX, May 1, 2002.
[4] R. R. Boudreaux and R. M. Nelms, A comparison of MOSFETs,
IGBTs, and MCTs for solid state circuit breakers, in Proc. IEEE
APEC96, vol. 1, 1996, pp. 227233.
[5] L. Palav and A. M. Gole, On using the solid state breaker in distribu-
tion systems, in Proc. IEEE Canadian Conf. Electrical and Computer
Engineering, vol. 2, 1998, pp. 693696.
[6] H. S. Song, K. Nam, and P. Mutschler, Very fast phase angle estimation
algorithm for a single-phase system having sudden phase angle jumps,
in Conf. Rec. IEEE-IAS Annu. Meeting, vol. 2, 2002, pp. 925931.
[7] L. Ljung, System Identication: Theory for the User. Upper Saddle
River, NJ: Prentice-Hall, 1999.
[8] S. M. Kay, Fundamentals of Statistical Signal Processing: Estimation
Theory. Englewood Cliffs, NJ: Prentice-Hall, 1993.
[9] M. S. Sachdev and M. Nagpal, Arecursive least error squares algorithm
for power system relaying and measurement applications, IEEE Trans.
Power Del., vol. 6, no. 3, pp. 10081015, Jul. 1991.
Worakarn Wongsaichua (S03) received the B.S.
and M.S. degrees in electrical engineering from King
Mongkuts Institute of Technology Ladkrabang
(KMITL), Bangkok, Thailand, in 1997 and 2001, re-
spectively. He is currently working toward the Ph.D.
degree in electrical engineering at the University of
Texas at Arlington (UTA) in the Energy Systems
Research Center (ESRC).
His research interest is application of digital signal
processing in power systems.
Wei-Jen Lee (S85M85SM97) received the B.S.
and M.S. degrees from National Taiwan University,
Taipei, Taiwan, R.O.C., in 1978 and 1980, respec-
tively, and the Ph.D. degree from the University
of Texas at Arlington in 1985, all in electrical
engineering.
He joined the University of Texas at Arlington,
where he is currently a Professor in the Electrical
Engineering Department. He has been involved
in research on power ow, transient and dynamic
stability, voltage stability, short circuit, relay coor-
dination, power quality analysis, and deregulation of utility industries. He is
also involved in research on the design of integrated microcomputer-based
monitoring, measurement, control, and protection equipment for electric power
systems.
Prof. Lee is a Registered Professional Engineer in the State of Texas.
WONGSAICHUA et al.: INTEGRATED HIGH-SPEED IUT FOR DISBURSED/RENEWABLE GENERATION FACILITIES 513
Soontorn Oraintara (S97M00SM04) received
the B.E. degree (with rst-class honors) from King
Mongkuts Institute of Technology Ladkrabang,
Bangkok, Thailand, in 1995, the M.S. degree in elec-
trical engineering from the University of Wisconsin,
Madison, in 1996, and the Ph.D. degree in electrical
engineering from Boston University, Boston, MA,
in 2000.
He joined the Department of Electrical Engi-
neering, University of Texas at Arlington (UTA), as
an Assistant Professor in July 2000. From May 1998
to April 2000, he was an Intern and a Consultant with the Advanced Research
and Development Group, Ericsson Inc., Research Triangle Park, NC. His
current research interests are in the eld of digital signal processing: wavelets,
lterbanks, and multirate systems and their applications in data compression,
signal detection and estimation, communications, image reconstruction, and
regularization and noise reduction.
Dr. Oraintara received the Technology Award from Boston University for the
invention of Integer DCT (with Y. J. Chen and T. Q. Nguyen) in 1999. In 2003,
he received the College of Engineering Outstanding Young Faculty Member
Award from UTA. He represented Thailand in the International Mathematical
Olympiad competitions and, respectively, received the Honorable Mention
Award in Beijing, China, in 1989, and the Bronze Medal in Sigtuna, Sweden,
in 1990.
Chiman Kwan (S85M93SM98) was born in
1966. He received the B.S. degree in electronics
with honors from the Chinese University of Hong
Kong, Hong Kong, in 1988, and the M.S. and Ph.D.
degrees in electrical engineering from the University
of Texas at Arlington in 1989 and 1993, respectively.
From April 1991 to February 1994, he was with
the Beam Instrumentation Department of the Super-
conducting Super Collider Laboratory (SSC), Dallas,
TX, where he was heavily involved in the modeling,
simulation, and design of modern digital controllers
and signal processing algorithms for the beam control and synchronization
system. Between March 1994June 1995, he was with the Automation and
Robotics Research Institute, Fort Worth, TX, where he applied intelligent
control methods such as neural networks and fuzzy logic to the control of
power systems, robots, and motors. Since July 1995, he has been with Intel-
ligent Automation, Inc., Rockville, MD. Currently, he is the Vice President,
leading research and development efforts in signal/image processing and
controls. He has served as Principal Investigator/Program Manager for more
than 50 different projects, such as modeling and control of advanced machine
tools, digital control of high-precision electron microscopes, enhancement of
microscope images, adaptive antenna arrays for beam forming, automatic target
recognition of FLIR and SAR images, fast ow control in communication
networks, vibration management of gun-pointing systems, health monitoring
of electromechanical systems, high-speed piezoelectric actuator control,
fault-tolerant missile control, speech enhancement, and underwater vehicle
control. His primary research areas include fault detection and isolation, robust
and adaptive control methods, signal and image processing, communications,
neural networks, and fuzzy logic applications.
Dr. Kwan is listed in the New Millennium edition of Whos Who in Science
and Engineering and is a member of Tau Beta Pi. One of his projects related
to Space Shuttle Auxiliary Power Unit monitoring received a NASA Space Act
Award. He also received an invention award for his work at SSC.
Frank Zhang received the Ph.D. degree in electrical
engineering from the University of Cincinnati,
Cincinnati, OH, in 2001.
In 2001, he joined Intelligent Automation, Inc.,
Rockville, MD, as a Research Scientist. Two years
later, he was promoted to Senior Research Scientist.
His research areas include fault diagnostics and
prognostics, underwater sensor networks, fault-tol-
erant control, and agent-based health monitoring.
He has authored more than 15 published journal and
conference papers on these topics.