1

An Adaptive Distance Protection Scheme for

Distribution System with Distributed Generation
Jing Ma, Member, IEEE, Jinlong Li, and Zengping Wang, Member, IEEE

costs of DG technologies are falling. This makes it more

Abstract—The value setting of the traditional distance economical to meet an increase of load by connecting DGs to
protection has become very difficult for the distribution system distribution feeders rather than expanding transmission and
with distributed generators (DGs) owing to the randomness of distribution (T&D) facilities [7]. Therefore, these technologies
the DG output power. This paper proposes a new adaptive are entering a period of rapid expansion and
distance protection scheme to solve this problem. Firstly, on the
basis of the connected point of DGs, the protected feeders are commercialization and studies have predicted that DGs may
divided into different zones, with adaptive distance protection account up to 20% of all new generation going on line by year
configured at the upstream side of DG. Then, based on a fault 2011 [4]. In such a system, DGs would feed loads around its
characteristic analysis of the distribution system with DG, the location, thus relieving the burden on the source. This clearly
adaptive distance protection scheme for the distribution system suggests that the basic assumption of distribution system being
with DG is studied to solve this problem. The results show that radial is not likely to hold in near future. One would then be
this scheme can automatically calculate the settings according to
the operation mode and the output power of DG with no need for looking at a multisource unbalanced system. It is a well-
communication. Compared with the traditional adaptive established fact that protection devices in a multisource
protection scheme, the performance of this method is greatly system have to be direction sensitive [1]–[3]. Hadjsaid et al. [9]
improved. Simulation results on a 10 kV distribution system have show through a simple example that fault currents through
verified the validity of the scheme proposed. protective devices would change after introduction of DG.
They further suggest checking protection selectivity and
Index Terms—Power system, Distributed generation (DG),
Distribution network, Adaptive distance protection sensitivity for each new connection of DG. Thus the study of
adaptive protection for distributed system with DGs is
I. INTRODUCTION demonstrated in this paper. The protected feeders are firstly
divided into different zones, with adaptive distance protection
D ISTRIBUTED generation (DG) is by definition
generation which is of limited size (few kilowatts to few
configured at the upstream side of DG. Then, the adaptive
distance protection scheme for the distribution system with
megawatts) and interconnected at substation, distribution
DG is proposed. Simulation results on a 10 kV distribution
feeder or customer load level [4]---[8]. DG technologies include
system have verified that the scheme is able to clear the fault
photovoltaics, wind turbines, fuel cells, micro turbines, gas
with high sensitivity and selectivity.
turbines and internal combustion engines [5], [6], [8]. DGs can
be divided into two categories: traditional rotating machine
II. ADAPTIVE SCHEME
DG and inverter interfaced DG (IIDG). Rotating machine DG
can be equivalent to series connection of voltage source and In the transient process of fault, the output power will
impedance, whereas IIDG is connected to power system via increase in short time [12]. Then the IIDG output P and Q
power electronic devices [10]-[11]. Even wind power and return to the setting value very quickly, whereas the output
micro gas turbine are also connected to power grid via current becomes larger than the pre-fault value due to voltage
rectifying-inverting process, to avoid the problems brought by reduction. Furthermore, the three-phase output currents of
direct connection of traditional generator. IIDG usually adopts IIDG are always equal during the transient process. Thus,
DC-type PQ-mode control, namely makes DG output P and Q IIDG only provides positive sequence current no matter what
achieve the setting value by adjusting output current of the type of fault occurs. Zone I of distance protection is not
inverter. affected by the IIDG and can be calculated by conventional
The cost of transmission and distribution is rising, but the method. However, Zone II should coordinate with Zone I of
distance protection of next line [13]-[14]. Therefore, this paper
mainly focuses on value setting of Zone II of distance
This work is supported in part by the National Science Foundation of China protection.
(No. 50907021, 50777016, 50837002), and in part by the “111” project
(B08013).
Distribution system containing an IIDG is shown in Fig. 1.
Jing Ma and Zengping Wang are with the Key Laboratory of Power Relay 2 has an infeed current provided by IIDG. For three-
System Protection and Dynamic Security Monitoring and Control under phase fault, the setting value of Zone II of relay 2 is given by
Ministry of Education, North China Electric Power University, 102206,
Beijing, China (e-mail: [email protected]).
II
Z set (
II
)
I
.2 = K rel Z BC + K b Z set .3 (1)
Jinlong Li is with the Chengdu Electric Power Bureau, 610021, Chengdu, where Kb is the branch coefficient. Kb is calculated as
Sichuan Province, China

978-1-4244-8081-4/10/$26.00 ©2010 IEEE

 2

I is used to demonstrate the effectiveness of the adaptive setting

Kb = 1 + DG (2) method. The structure of the system is shown in Fig. 2. The
I2
base capacity is 500MVA, and base voltage is 10.5kV. Branch
AB, BC and AF are overhead lines. The parameters of these
lines are: r1=0.27Ω/km, x1=0.347Ω/km. Branch CD, DE and
FG are underground cables. The parameters of these lines are:
E S I1 I2 I3 I4 r1=0.259Ω/km, x1=0.093Ω/km. Each load are connected to
each bus with the nominal capacity 6MVA and nominal power
factor 0.85. IIDG with P-Q control mode is connected to bus C
and its nominal capacity is 10MVA. The simulation is carried
out on PSCAD/EMTDC.
Fig. 1. Schematic diagram of power system containing IIDG
A F
For the two-phase short circuit fault, branch coefficient can G
E S ZS 4km 6km
not be calculated by (2) directly. If B-C phase fault occurs at
~ 5 6
the end of zone I of relay 3, currents at relay 2 and relay 3 is
B C D
calculated as E
⎪⎧ I2(1.b) = I3(1.b) − IDG.b 5km 8 2 5km 7 3 7km 4 14km
⎨ (1) (3) 1
⎪⎩ I 2.c = I3(1.c) − IDG.c
⎧ I2( 2.b) = I3(.2b) ~ IIDG
⎨ ( 2 ) ( 2 ) (4) Fig. 2. Structure of example power system in Tianjin
⎩ I 2.c = I 3.c
where I2(1.b) is positive sequence current of phase B at relay 2, A. Analysis of Zone II Protection
I2( 2.b) is negative sequence current, I2(1.c) is positive sequence Integration of IIDG makes the area between buses A and C
a double-terminal power system, thus relays 7 and 8 are added
current of phase B at relay 2, I2( 2.c) is negative sequence to clear fault between buses A and C. Zone I of relays 1 to 8 is
current; I3(1.b) is positive sequence current of phase B at relay 3, not affected by the IIDG, and can be calculated by
conventional method. However, Zone II of relays 2 and 7 are
I3( .2b) is negative sequence current, I3(1.c) is positive sequence affected by infeed branch coefficient of generator current.
current of phase C at relay 3, I3( .2c) is negative sequence Setting of relay 2 is carried out with the method motioned
above. Considering infeed current of the traditional power
current, IDG.b is phase B current of IIDG, IDG.c is phase C source E S , setting value of relay 8 is calculated as
current of IIDG.
Voltages of phase B and phase C at relay 2 are given by
II
Z set II
( I
.8 = K rel Z BA + K b Z set .5 ) (8)
where Kb is the branch coefficient.
⎧⎪U 2.b = U F .b + I 3.b Z CF + ( I 3(1.b) − IDG.b ) Z BC + I 2( 2.b) Z BC
⎨ (5) When three-phase fault occurs, Kb is given by
 (1)  ( 2)
⎩⎪U 2.c = U F .c + I 3.c Z CF + ( I 3.c − I DG.c ) Z BC + I 2.c Z BC I
K b = 1 + S (9)
where U and U are voltages of phase B and phase C at
2.b 2.c I 8

relay 2, respectively. U F .b and U F .c are voltages of phase B where IS is the output current of E S , and I8 is the measured
and phase C at fault point, respectively. ZCF is the positive current of relay 8.
sequence impedance of the line between relay 3 and fault When two-phase fault occurs, branch current at relay 8 only
point, ZBC is positive sequence impedance of the line between contains positive sequence, so Kb can be calculated by:
relay 2 and relay 3. I − I S .c
Measured impedance of protection 2 is calculated as K b = 1 + S .b (10)
I8.b − I8.c
U − U
Z 2 m = 2b  2c = Z BC + Kb Z CF (6) B. Simulation Results
I 2b − I 2 c
where Kb is the branch coefficient, and it can be expressed as: Faults at the end of Zone I of the relays 3 and 5 are carried
out to test the selectivity of Zone II of the relays 2 and 8,
I − IDG.c
K b = 1 + DG.b (7) respectively. The amplitude comparison operation equations
I − I 2.b 2.c of relays 2 and 8 are shown in Fig. 3 and Fig. 4, respectively.
Substituting (7) into (1), setting value of zone II of relay 2 1
during two-phase fault can be obtained. In both figures, “1” represents the amplitude Z m − Z set ,
2
III. TESTING RESULTS AND ANALYSIS 1
and “2” represents the amplitude Z set . From both figures,
A 10kV distribution system [15] in Tianjin power network 2
 3

1 1 the protected line.

Zm − Z set is much bigger than Z set . Therefore, Zone TABLE II
2 2 SENSITIVITY OF RELAY 8
Ⅱof relays 2 and 8 will not exceed the protection range of
Zone Ⅰof the relays 3 and 5, respectively. Selectivity of the Two – phase Fault Three – phase Fault
S(MVA)
two relays can be guaranteed. Setting Value Sensitivity Setting Value Sensitivity

2.5 2.5 10 42.0619 19.1335 42.0811 19.1422

1 1 8 52.3898 23.8315 52.4142 23.8425

6 69.5983 31.6594 69.6408 31.6787

Impedance/

2 Impedance/ 2 4 103.9998 103.9998 104.0834 47.3462

2 206.8427 94.0902 207.2046 94.2548
2 2
0 Exit Exit Exit Exit
1.5 1.5
0.4 0.6 0.8 0.4 0.6 0.8
Time/S Time/S
Two phase fault Three phase fault IV. ACKNOWLEDGMENT
Fig. 3. Amplitude comparison operation equation of protection 2. Financial support from the National Science Foundation of
China (50907021, 50607006, 50837002) and “111” project
8 8
(B08013).is gratefully acknowledged.
1
6 1 6
V. CONCLUSIONS
Impedance/
Impedance/

4 4 This paper presents an adaptive distance protection scheme

of distribution system with DG. Firstly, with adaptive distance
2 2 protection configured at the upstream side of DG, the
2 2 protected feeders are divided into different zones. Then, on the
0 0
0.4 0.6 0.8 0.4 0.6 0.8 basis of fault characteristic analysis of the distribution system
Time/S Time/S with DG, the adaptive distance protection scheme is proposed.
Two phase fault Three phase fault The results show that this scheme can automatically calculate
Fig. 4. Amplitude comparison operation equation of protection 8. the settings according to the operation mode and the output
power of DG with no need for communication. Simulation
Two-phase and three-phase fault are set on bus C results on a 10 kV distribution system have verified scheme
separately to check the sensitivity of relay 2. Different has highly acceptable selectivity and sensitivity.
capacity of the IIDG is considered. Simulation results are
shown in Table I. VI. REFERENCES
[1] M. A. Anthony, Electric Power System Protection and Coordination.
TABLE I
New York: McGraw-Hill, 1995, pp. 109–121, 342-346.
SENSITIVITY OF RELAY 2
[2] J. L. Blackburn, Protective Relaying Principles and Applications. New
York: Marcel Dekker, 1998, pp. 383–408.
Two – phase Fault Three – phase Fault [3] P. M. Anderson, Power System Protection. New York: IEEE Press, 1999,
S(MVA) pp. 201–240, 249-257.
Setting Value Sensitivity Setting Value Sensitivity [4] P. Barker and R. W. de Mello, “Determining the impact of distributed
generation on power systems: Part 1—Radial power systems,” in Proc.
10 3.8309 1.7426 3.8561 1.7541
IEEE Power Eng. Soc. Summer Power Meeting, 2000, pp. 1645–1658.
8 3.6639 1.6666 3.6616 1.6656 [5] U.S. Department of Energy’s distributed power program homepage,
Online Available:, http://www.eren.doe.gov/der/basics.html.
6 3.4799 1.583 3.4826 1.5842 [6] Portfolio of DG Technologies. [Online]http://www.distributed-
4 3.3031 1.5025 3.2996 1.5009 generation. com/technologies. Html
[7] H. L. Willis and W. G. Scott, Distributed Power Generation Planning
2 3.1208 1.4196 3.1182 1.4184 and Evaluation. New York: Marcel Dekker, 2000.
[8] “Role of distributed generation in competitive energy markets,” Gas Res.
0 2.9497 1.3418 2.9497 1.3418
Inst., Rep. GRI-99/0054, Chicago, IL, http://griweb.gastechnology.
org/pub/solutions/dg/distgen.pdf.
From the simulation results, it can be concluded that zone [9] N. Hadjsaid, J. Canard, and F. Dumas, “Dispersed generation impact on
distribution networks,” IEEE Comput. Appl. Power, vol. 12, pp. 22–28,
II of protection 2 has enough sensitivity to clear fault at the Apr. 1999.
end of the protected line. [10] C. V. Dobariya and S. A. Khaparde, "Decoupled power controller for
Then faults on bus A are simulated to verify the sensitivity inverter-interfaced distributed generation system", in 2007 IEEE PES
General Meeting, pp.1-6.
of protection 8. Different capacity of the IIDG is considered.
[11] G. Venkataramanan and M. Illindala ,"Small signal dynamics of inverter
Simulation results are shown in Table II. interfaced distributed generation in a chain-microgrid", in 2007 IEEE
From the simulation results, it can be concluded that Zone Power Engineering Society General Meeting, pp.1-6.
II of relay 8 has enough sensitivity to clear fault at the end of
 4

[12] Shengwei Li, Yongli Li, Jingliao Sun, Qiang Jin, and Xuguang Li, "A
novel control algorithm for inverter-based distributed generation in
unbalanced three-phase power systems", in 2009 Sustainable Power
Generation and Supply Int. Conf.,pp.1-6.
[13] Tarlochan S. Sidhu, David Sebastian Baltazar, Ricardo Mota Palomino,
and Mohindar S. Sachdev, "A new approach for calculating zone-2
setting of distance relays and its use in an adaptive protection system",
IEEE Trans. Power Delivery, vol. 19, pp.70-77, Jan. 2004.
[14] M.M. Saha, J. Izykowski, E. Rosolowski, M. Bozek, "Adaptive line
distance protection with compensation for remote end infeed", in 2008
IET Developments in Power System Protection Int. Conf., pp.321-326.
[15] Sun Jingliao, Li Yongli, Li Shengwei, and Jin Qiang, "Study on
adaptive current instantaneous trip protection scheme for distribution
network with inverter interfaced DG", Automation of Electric Power
Systems, vol.33,pp.71-76, Jul. 2009.(in Chinese).

Jing Ma (S’06-M’08) was born in Hebei Province,

China on February 25, 1981. He received his B.S.
and Ph.D. degree from North China Electric Power
University, China, in 2003 and 2008, respectively.
He is currently a visiting scholar in the Bradley
Department of Electrical and Computer Engineering,
Virginia Polytechnic Institute and State University,
USA, on leave from the School of Electrical and
Electronic Engineering, North China Electric Power
University, China. His major interests include power
system stability, control and wide area backup protection (e-mail:
[email protected]).

Jinlong Li is currently an assistant engineer working

in the Chengdu Electric Power Bureau, Chengdu,
Sichuan Province, China. He received M.S. degree
from the School of Electrical and Electronic
Engineering, North China Electric Power University
in 2010. He received his B.S degree from Zhengzhou
University in 2007. His interests mainly include
wide area power system measurement, dynamic
analysis, and backup protection.

Zengping Wang (M’05) was born in Hebei

Province, China, on November 3, 1964. He received
the B.S. and M.S. degree in Electric Engineering
from North China Eletric Power University, China,
in 1985 and 1988, respectively. He received the
Ph.D. degree from Harbin Institute of Technology,
China, in 1997. He is a professor and the dean of the
School of Electrical and Electronic Engineering at
North China Eletric Power University. His special
fields of interest include power system equipment protection, fault analysis
and wide-area protection.