Energy and Power Engineering, 2013, 5, 937-940
doi:10.4236/epe.2013.54B179 Published Online July 2013 (
Research on Fault-feeder Selection Dvice
Zimin g Zhuo1, Haimeng Sun2, Haibin Ji1
1School of Mechanical & Electrical Engineering Xuzhou College of Industrial Technlolgy, Xuzhou, China
2Xuzhou Distribution Station Xuzhou Power Supple Company, Xuzhou, China
Received February, 2013
The compensation current of the arc-suppressing coil makes the phase and amplitude of zero-sequence measurement
current of the earthed fault feeder to vary. It is very hard to detect the fault feeder by using existing detectors based on
single method. In this paper, integrative feeder selection strategy—zero sequence current increment method and the
direction of transient current— is put forward. Based on the integrative feeder selection strategy, the design of
fault-feeder selection device for one-phase-to ground fault on resonance grounding system is presented. For the purpose
of testing and validating the operating principle of the device, the experiment of single-phase-to-ground fault has been
carried out on the simulation of 1.2 kV power n etwork. The resu lts from many repeat experiments show that stability of
the fault selection device is satisfactory.
Keywords: Resonance Grounding System; Earthed Fault Feeder Detection; Zero-Sequence Current Increment Method;
Zero-sequence Transient Currents
1. Introduction
Neutral ineffectively grounding modes especially reso-
nance grounding systems are widely employed in me-
dium and low voltage distribution networks of China.
Because of the compensation current of
ng coil, the fault current is so weak that it is hard to de-
tect fault-feeder. When the single-phase-to-ground fault
occurs in resonance grounding system, it is an important
research subject in resonance grounding system that de-
tecting the ground fault line.[1,2]
Ground-fault is complicated in resonance grounding
system. Zero sequence current may only contain transient
or steady-state or both of them in all kinds of ground-
ault. The existing detectors are based on transient analy-
sis or steady analysis of zero-sequence current. It makes
the result of detecting is not satisfactory.
This paper presents a new method of integrative fault-
feeder selection that is based on the steady-state analysis
current zero-sequence current increment and the direc-
tion of transient current analysis zero sequence transient
current direction.
2. Method of Integrative Fault-feeder
2.1. Zero Sequence Current Increment Method
When single phase grounding fault occurs in resonance
grounding system, zero-sequence transient current of
non-fault line.
jU C
where R1 represents each line’s the resistance to ground,
C1 represents each line’s capacitance to earth.
When single-phase-to-ground fault takes place, the com-
pensation current of arc-suppressing coil makes phase
and amplitude of the zero-sequence measurement current
of earthed fault feeder to vary. 0
represents decrease
of the zero-sequence current of non-fault feeder with and
without compensation .
00201 0201i
ii i
 
 
L1 represents inductance value of arc-suppressing be-
fore adjustment, L2 represents inductance value of arc-
uppressing after adjustment, then 0k1
represents zero
sequence current of fault feeder. 0k
epresents zero
sequence current difference values of fault-feeder before
and after adjusting indu ctance value[4,5,6].
0k1 01i
1, k1, k1
ii ii
IUjc j
 
 
 1
not only relates to parameters , , but also
depends on param et e rs L1. 01
The value of 0102 changed slightly, even entirely
neglect the influence on the result of 0k
. 0k
affected remarkably by L1. Through the discussion of the
Copyright © 2013 SciRes. EPE
Equation (3), (4), it is found that Effective value of 0k
is higher than those of 0k
, which provides operational
judgment basis for fault feeder selection.
02 101 2
0k01 02i
1, k12
-(3 +
 
()t i
2.2. The Transient Current Direction Method
When single-phase earth fault occurs in resonance earthed
distribution system, for some cases the steady state
component of fault current is small or approximates to
zero, it leads to insufficient faulty feeder detection th at is
based on steady state analysis. According to the sign of
integral mean value of mutual dot products of ze-
ro-sequence transient currents of each two feeders, an ap-
proach to judge the direction of fault curren t is proposed.
The integral mean value during a time interval are taken
as the judging quantity
If the ith line and jth line are non-fault line, the fault
judging quantity of every feede was the product of two
lines zero-sequence transient current[1,3,4]
()()()() 0
iji fj
it itt it
  (5)
If one of the ith line and jth line is fault line, i was
the product of two lines of zero-Sequence transient cur-
()()() 0
ijfi gk
it it it
  (6)
The following conclusions are drawn from above
analysis: Only one i is negative integer and the rest is
positive integer. ith feeder is fault feeder. All of i are
positive integer and Reference feeder is fault feeder
All of i are positive integer an d Reference feeder is
fault feeder.
3. Software and Hardware Composition of
Fault-line selection Device
3.1. Hardware Composition
Fault-line selection device for One-phase-to-ground Fault
on resonance grounding system is designed from both
hardware and software. Hardware properties directly
affect the performance of device. According to the design
requests, the detecting equipment use chip TMS-
320LF2812A of TI Company as central processor. Zero
sequence voltage analog signals and zero sequence cur-
rent analog signals of feeders, which is collected through
the signal acquisition module, will be converts into digi-
tal signal, and real-time display of calculated results of
zero sequence current value. Based on integrative line
selection strategy—zero sequence current increment me-
thod and the direction of transient current, the device
judge whether the occurring of the single phase-to-earth
fault. Hardware comprises main components—power
supply, A/D converter, central processor, signal sampling
3.2. Software Composition
Major task for software system of fault-feeder selection
is real-time sampling of zero sequence voltage and zero
sequence current. The data can be used to further analy-
sis and judgment of the operating state of electric net-
The software design adopts the module design thought,
and takes the master routine and the interrupt subroutine
as the main structure. The program consists of one main
program, and four subroutines (system initialization mod-
ule, Signal sampling module, Fault judgment module,
HMI module).
First, system is initialized, then the main loop Proceed
to start sampling. As it can be seen from Figure 1: When
single-phase grounding fault occurs, the system calls
fault judgment subroutines, records the fault information
and the fault-line detection result is displayed on the
LCD screen. Under normal operation condition of the
network, the main program has been in the acquisition
and processing of zero sequence current and zero se-
quence voltage. When there is an interrupt request signal,
DSP deals with corresponding subroutines.
The sampling program adopts time interruption acqui-
sition. a common timer of The EVA (event management
a timer ) is used as sampling timer .For each sampling
interval, EVA generate a timing interrupt. Based on the
timing interrupt, AD converter puts 10-channel analog
signals of zero sequence voltage and current into Digital
Signal. When is called again, AD converter puts module
put analog signals of zero sequence voltage and current
into digital signal by Multip lex[7].
When it is detected that zero-sequence voltage is larger
than setting value, block diagram of fractional subrou-
tines is called subroutine and examine the fault-line by
analysis and judgment of Sampling zero-sequence volt-
age and zero-sequence current program flow chart is
given as follows:
For the method of the transient current fault-feeder se-
lection, notch filter for power-line interference is applied
to sampling data to obtain amplitude of zero sequence
current. Maximum amplitude of zero sequence current
can be considered to be reference feeder. The fault judg-
ing quantity of every feeder i is the product of the
sampling zero sequence current values of reference feed-
er and other feeders. Fault-feeder is then identified ac-
cording to formula 4. Every fault-feeder is given a set of
only numbers after the judgment of fault-feeder, so sam-
pling data array includes a row of numbering of each
Copyright © 2013 SciRes. EPE
Z. M. ZHUO ET AL. 939
4. Experiment of Selection Device
Once the device of fault line selection has been designed
and constructed, its experimental performance has been
4.1. Zero-sequence Measurement Current of
Non-faulted Fault-feeder
For the purpose of validating the selection, the experi-
ment of one-phase-to-ground fault on resonance ground-
ing system has been carried through with the simulation
of 1.2 kV power network. As it has been done in the pre-
ceding experimentation, real wareform of fault and non-
fault zero sequence current is obtained by fault wareform
As it can be seen from Figure 2: From the top to the
end, it is waveform of 1st and 2nd zero-sequence meas-
urement current of non-fault feeder, 3rd zero-sequence
measurement current of non-fault feeder, zero sequence
voltage. MHK series controller of arc suppression coil
start to adjust the inductance of arc suppression co il after
5 cycles.
Fault judgement
interrupt ent ry
The method of transient zero-
sequence current direction
The method of zerosequence
current increment
The comprehensive judgment of
fault feeder
Display and Record Fault
Figure 1. Block diagram of fractional subroutines.
Figure 2. wave recording of one-phase-to-ground fault.
Figure 2 shows amplitude of 2nd zero-sequence cur-
rent is higher those of 1st zero-sequence current for the
difference between the capacitance to earth of 1st and 2rd.
When single-phase grounding fault occurs, the varying the
phase, amplitude and frequency of the zero-se- quence
measurement current voltage of the non-fault feeder is
not remarkable. HoweverThe compensation current of
the arc-suppressing coil makes the phase and amplitude
of the zero-sequence measurement current of fault feeder
to vary significantly. Experimental results verify the va-
lidity of zero sequence current increment method de-
scribed in the previous sections.
4.2. Dynamic Response of the Device
In order to verify the Function of fault-line selection de-
vice based on integrative line selection strategy, the de-
vice is used to detect fault-feeder for one-phase-to-
ground fault is used on the simulation of 1.2kV power
In this experimentation, the behavior of the device has
been studied when single-ph ase grounding fault sudd enly
When a sudden one-pha se-to ground fault hap pensthe
zero-sequence current varies instantaneously. In this case,
it can be observed how zero-sequence current only needs
30 ms to arrive to the step of the compensation of the
arc-suppressing coil from transience. During the time, the
difference between tamplitude of the transient current of
with that of stable current for fault feeder is remarkable.
Tamplitude of the transient cu rrent much higher than th at
of stable current for fault feeder. After MHK series con-
troller of arc suppression coil ach iev ed, the compensation
current of the arc-suppressing, zero-sequence current is
step of stable current in 5 cycle.
Figures 3 and 4 show the direction of zero-sequence
transient current of fault feeder is opposite to that of the
zero-sequence transient current of the non-fault feeder.
Ideally, the result is coincident with the theoretical analysis
Figure 3. Zero voltage and current waves recording of non-
Copyright © 2013 SciRes. EPE
Copyright © 2013 SciRes. EPE
1) Experimental result of the selection device has been
compared with real wave recording of fault wareform
recording. As a result, two values were very matched. So,
we confirmed the feasibility of the proposed fault-line
selection d ev i ce.
2) The integrative line selection strategy—zero se-
quence current increment method and the direction of
transient current— been proposed. Real wareform of
fault wareform recording confirm that the in tegrative line
selection strategy is more stable and accurate method
comparison with the single method.
Figure 4. Zero voltage and current waves recording of fault-
feeder. [1] H. Li, Y. Tang and C. Q. I.Sun, “Line Selection Device fo
Small Current Single-phase Grounding Fault Bsed on In-
tegrative Line Seletion Strategy”Industry and Mine Ato-
mation, pp. 35-38, April 2012.
of detecting earthed fault feeder the transient current di-
Figure 4 shows experimental wareform of the zero-
sequence transient current of the fault feeder. As shown
in this waveform, single-phase grounding fault results in
more higher zero-sequence current increment. The fault
duration was extended to 30 ms to allow the zero-se-
quence current to reach the new operating state. We com-
pared the monitoring results from fault wareform re-
cording with the experimental one. Consequently, two
values were very matched.
[2] X. C. Wang, “Zero Sequence Current Incremental Ratio
Method in Small Current Grounding Fault Line Identifi-
cation,” Vol. 20. Power System Protecetion and control,
October 2011, pp.125-129.
[3] H. C. Su, “Fault Line Seletion of Distribution Power Sys-
tem,” China Machine press, Beijing 2008.
[4] K. Cheng and Y. Tang, “Single-Phase-to-Ground Fault
Protection for Indirect Grounding Power System,” High
Voltage Engineering.
[5] J. LEE, “The Research on Arc Suppression Coil and Line
Selection Control Technology in Distribution Network,”
North China Electric Powe Unversity, 2007.
5. Conclusions
In this paper, integrative line selection strategywhich is
zero sequence current increment method and the direc-
tion of transient current, been proposed. Based on the
integrative line selectio n strategy, the device of fault-line
selection for one-phase-to ground fault on resonance
grounding system is designed. The result obtained from
the experiment carried out on the simulation of 1.2 kV
power network is as follows.
[6] Y. Yorozu, M. Hirano, K. Oka and Y. Tagawa, “Elect ron
Spectroscopy Studies on Magneto-Optical Media and
Plastic Substrate Interface,” IEEE Transl. Journal of
Magnetics in Japan, Vol. 2, No. 8, 1987, pp. 740-741.
[7] M. Young, “The Technical Writer's Handbook,” Mill
Valley, CA: University Science, 1989.