Fault Model of IIDG Considering LVRT and Its

Application in Fault Analysis of Active Distribution

Networks

Zhihui Dai, Chuan LI Xi Chen

School of Electrical and Electronic Engineering， Development and Plan Department
North China Electric Power University Baoding Power Supply Company,
Baoding 071003, China Baoding 071003, China
[email protected]

Abstract—According to the low voltage ride through (LVRT) the accuracy of network division is hard to be guaranteed.
requirement stipulated by the distribution grid standards, an
LVRT strategy based on reactive power support of the inverter The fault current expression of IIDGs is obtained in [7],
interfaced distribution generation (IIDG) was first presented. taking IIDGs as constant power sources. The error may be
The maximum energy output, which achieves the optimal great since the fault current also depends on the control
efficiency of distributed power supply, was also taken into strategies. A fault current calculation model was deduced in [8],
account. Then the fault characteristics of IIDG with LVRT considering the relation between DG and sequence voltages of
strategy were analyzed. Considering the impact of LVRT PCC.
controlled IIDG on power distribution, the short circuit current
analysis is discussed and a short circuit calculation model of In this paper, firstly, an LVRT strategy aiming at supplying
distribution networks with IIDG was established. Case study in the reactive power support is proposed through reactive current
Digsilent shows the effectiveness of the LVRT strategy and the control. Secondly, considering the LVRT strategies, the current
fault calculation method. output of IIDG under system fault is analyzed to obtain the
equivalent fault model. Thereafter, fault calculation of
Keywords—inverter interfaced distributed generation, low distribution networks with IIDG is given. Case study based on
voltage ride through, fault analysis, short circuit model DigSILENT verifies the effectiveness of the proposed models.

I. INTRODUCTION II. LVRT STRATEGY CONSIDERING ACTIVE POWER

In compliance with several requirements, DG sources must OPTIMIZATION AND REACTIVE CURRENT SUPPORT
remain connected to the grid during voltage sags, following a When a fault occurs in the distribution network with DGs,
predefined time/sag-depth profile before disconnection, which the DG experiences the transient stage and quasi-stable state. In
is known as low-voltage ride-through (LVRT)[1,2]. the transient stage, the fault current, mainly relating to grid-
connected control strategy, changes rapidly. However, this
Unlike wind generation, the LVRT technology of the
stage has less impact on the protection system.
inverter integrated DG (IIDG) is still in its infancy, which
mainly depends on storage devices, Var Compensator and During the quasi-stable state, the fault current tends to be
reactive current and voltage [3]. stable, and decided mainly by grid-connected control strategy
and LVRT strategies. Especially when the fault in the
What is more, the integration of DGs changes classical radial
distribution network results in serious voltage sag at the PCC
topology of distribution networks, and results in the change of
point of large-capacity DGs [9].
both the amplitude and direction of power flow and fault
currents. The variation of the capacity and PCC of DGs is also In China, the reactive power support capability of PVs is
a big disadvantage to the safe operation of protection systems. required [10]. The reactive current injected by PV should
follow the change of voltage of PCC, , while there exists
According to the function of LVRT, DGs are supposed to
voltage sag. The injected reactive current is specified according
supply reactive power support. Therefore, the fault current
to different voltage sags as follows.
generated from IIDG could be regulated by the LVRT
strategies, and the fault analysis of distribution networks with  I T  0;(U pcc  0.9U N )

DGs should take into account the impact of LVRT strategies.  I T  1.5  (0.9  U pcc / U N ) I N ;(0.2U N  U pcc  0.9U N ) (1)

In [4], an ideal voltage source with constant impedance was  I T  1.05I N ;(U pcc  0.2U N )
used as the equivalent fault model of DGs. It didn’t consider
the impact of control strategies of DGs. A controlled current Where, is the p.u. voltage of PCC. is the rated current
source model of IIDGs was presented in [5]. In [6], a fault of PV station.
calculation based on the node split method is given. However,
Based on the double-loop control strategy, an LVRT strategy
aiming at supplying reactive power support is proposed. The d-
Supported by the National Natural Science Foundation of China (51307059),
Hebei Provincial Natural Science Foundation of China (E2014502065), and the
Fundamental Research Funds for the Central Universities (2014MS86). .
axis current component of AC-side active power is controlled
(7)
to implement the regulation of DC-side voltage, and the q-axis
current component depends on .
Hence the voltage interval is extended to four from the three
Specifically, when the system operates normally, the defined in [10].
capacitance voltage on the DC side is stable and . III. FAULT ANALYSIS MODEL OF DISTRIBUTION NETWORKS
The MPPT makes the PV system output maximum power, and WITH IIDG
the PV source could be deemed as constant power sources.
Hence the output current would be , where is Based on the equivalent fault model of IIDG, the fault
the maximum output power and is the rated voltage. analysis of distribution networks with IIDG is implemented in
this section. The symmetrical component method and the
When faults occur, to retain the DC-side capacitance voltage, voltage-controlled current source model of IIDG are employed.
the AC-side d-axis active current will remain unchanged, i.e. The iterative method was used to solve the nonlinear equations.
. The q-axis current is then determined by the . A simplified distribution shown in Fig.1 was used for the
fault calculation. Where, G is the main grid. Zs is the system
1) , it is not necessary to apply LVRT to DGs. impedance, Z12, Z23, Z34 and Z15 are the line impedance. Zload1,
The output current is the same as the current before the Zload2 and Zload3 are equivalent load impedances. ZT is the
fault. impedance of the grid-connected transformer.
2) , the LVRT is employed. Assume per Assume a Phases BC fault occurs on Bus 4. The equivalent
positive and negative sequence networks are shown in Figs. 2
1% voltage sag needs x% reactive current, i.e. and 3, respectively. The IIDG output only positive sequence
(2) current.
Additionally, the output current of inverters should be less 1 2 3 4
than 1.2 times rated current.
LV
(3) G
5 IIDG
When remains normal and the voltage of PCC dips to
0.2p.u., it could be obtained that Fig. 1. Diagram of a distribution network with IIDG

Zs Z12 Z23 Z34

(4) Z15 ZT
Zload2 Zload3
Zload1

However, it might not meet the requirement of (4). The

minimum voltage of PCC, , corresponding to the Fig. 2. Positive sequence network

possible maximum injected active current is obtained by Zs Z12 Z23 Z34

+ +
Z15 Zload3
(5) Zload1
Zload2
_ _

. Fig. 3. Negative sequence network

The use of helps the PV meet the requirement of
The negative sequence impedance in Fig.3 is calculated by
reactive power support, as well as keep the output active
power unchanged as much as possible. (8)
The composite sequence network is shown in Fig.4.
Zs 1 Z12 2 Z23 3 Z34 4 Zf
(6) + +
ZT
Z15
Zload2 Zload3
Zload1
_ _

The LVRT strategy while the voltage belongs to the interval

of [0.2pu, 0.9pu] is shown in (10). Fig. 4. Composite sequence network
3) , the PV injects only reactive current to the Based on the nodal voltage method, the following equation
distribution network could be obtained.
 IV. CASE STUDY
(9) A. Verification Of The LVRT Strategy
Simulation models, based on a 10kV distribution network
Where, ， ， and are voltages of Nodes 1-4, shown in Fig.1, were built in DIGSILENT to verify the
effectiveness of proposed models. The network parameters are
respectively. Additionally,
shown in Table1. The scenario of PCC voltage dipping to
0.3pu is analyzed.
(10) 1
0.6 Id Idref
0.2
According to the boundary conditions of phase to phase

pu
-0.2

faults, i.e. -0.6 Iq Iqref

-1


U fb  U fc 0 0.1 0.2 0.3 0.4 t/s 0.5 0.6

 (11)
 I fb   I fc
 Fig. 5. reference and actual value of d and q-axis fault currents

The IIDG is handled as a voltage controlled current source, 1.2

Positive sequence current
which could be represented by 1

0.8
(12)

pu
0.6

Equation (9) could be solved by iteration method by 0.4 Positive sequence voltage

0.2
integrating the voltage-controlled current source of IIDG in 0 0.1 0.2 0.3 0.4 t/s 0.5
0.6

each iteration. The initial value of PCC voltage after fault

occurrences, , could be obtained by using (9)-(12). Fig. 6. Positive sequence current and voltage
Thereafter, could be substituted in (9)-(12) 3.5
to obtain . The process should be
2.5
repeated until meets the accuracy requirement. 1.5
MW

A. Fault Analysis Model Of Distribution Networks with 0.5

Mvar
multiple IIDG -0.5
0 0.1 0.2 0.3 0.4 t/s 0.5 0.6
The branch current equation of distribution networks with
m IIDG could be written as Fig. 7. Active and reactive power
 I DG1   Y11 Y1i Y1n  U1 
    
1200

    
, I n   Gnm  I DGi    Yi1 Yii Yin  U i 
T 1000
 I1 , I 2 , (13)
    
Udc
kV

    
800

  Y Y Y   
 DGm   n1
I ni nn  U n  600

0 0.1 0.2 0.3 0.4 t/s 0.5 0.6

Where, is the injection current from main grids to Node i,
is injection current from jth DG to Node i, means Fig. 8. DC side voltage
the voltage of PCC i connecting the jth DG, is the
incidence matrix of m DG and n nodes. It can be seen that, when the PCC voltage dips to 0.3pu, the
q-axis current increases from 0 to 1pu and d-axis current
Substituting (12) into (13), then deceases to 0.66pu, to support the reactive power. The output
current of the PV is 1.05pu, which meets the requirement of
 F (U DG1 )   Y11 Y1i Y1n  U1  less than 1.2pu and implements the LVRT. Meanwhile, the
    
     active-power output of PV decreases from 8.96MW to 2MW,
, I n   Gnm  F (U DGi )    Yi1 Y
? ii Yin  U i 
T
 I1 , I 2 , (14) whereas the reactive power increases from 0 to 3MVar. The
    
   DC-side voltage of PV increases from 700V to 860V.
 
  Y Y Y   
 F (U DGm )   n1 ni nn  U n  B. Verification of fault analysis model of distribution
networks with IIDG
This equations could be solved by using iteration method, to
obtain all node voltages and branch currents. According to the fault analysis model shown in Section II,
symmetric and asymmetric faults on Bus 4 are analyzed to test
(15) the validity of the proposed model.
Comparison results of theoretical and simulation values
when two-phase fault and three-phase fault occurred at the proposed fault analysis method, and the positive-sequence
downstream bus of IIDG are shown in Tables 2 and 3, voltage controlled current source model of IIDG under fault
respectively. scenarios are effective. They could be used for fault
calculation of distribution networks with IIDG.
It can be seen from Tables 2 and 3 that, the calculation
values are quite close to simulation values. It indicates that the

TABLE.1 NETWORK PARAMETER

System System
Branch impedance / Transformer / Load /
capacity /MVA impecance /

500 j0.27
0.233+j 1.683+j 0.267+j 0.642+j 0.928+j 0.894+j 0.949+j
5+j5
0.187 1.068 0.377 0.283 0.371 0.446 0.316

TABLE.2 COMPARISON OF THEORETICAL AND SIMULATION VALUE WHEN INTERPHASE FAULT OCCURS AT DOWNSTREAM BUS OF SINGLE IIDG
Current of main grid /kA IIDG current /kA Fault current /kA IIDG voltage /p.u.
Positive Negative Positive Negative Positive Negative Positive Negative
sequence sequence sequence sequence sequence sequence sequence sequence
Simulation 0.402∠
1.392 -45.52 0.786 159.65 1.085 -129.5 0 1.016 -16.91 0.856 160.46 0.542 15.784
value 5.45
Calculation
1.390 -45.63 0.755 160.02 1.065 -130 0 1.012 -16.93 0.866 160.40 0.553 15.523 0.4∠5.50
value

TABLE.3 COMPARISON OF THEORETICAL AND SIMULATION VALUE WHEN THREE-PHASE FAULT OCCURS AT DOWNSTREAM BUS OF SINGLE IIDG
Current of IIDG current Fault current Voltage of IIDG voltage Voltage of Voltage of
main grid /kA /kA /kA main grid /kV /kV Node 2 /kV Node 3 /kV
Simulation
2.225 -35.862 0.606 -87.62 1.511 -21.62 0.985 -1.161 0.152 2.101 0.329 8.177 0.184 2.152
value
Calculatio
2.21 -35.78 0.612 -87.55 1.50 -21.73 0.991 -1.08 0.16 2.35 0.344 8.15 0.186 2.18
n value

V. CONCLUSIONS [2] Kumar ． DS ， Srinivasan ． D ， Reindl ． TA ． ”Fast and scalable

protection scheme for distribution networks with distributed generation,”
Considering the impact of IIDG on the fault current of IEEE Transactions on Power Delivery，2016，vol. 31, pp.67-75．
distribution networks, an LVRT strategy is presented and a [3] Nuutinen, Pasi, Pasi Peltoniemi, and Pertti Silventoinen. "Short-circuit
protection in a converter-fed low-voltage distribution network." IEEE
fault analysis is given. Transactions on Power Electronics, vol.28, no.4, 2013, pp. 1587-1597.
1) An LVRT strategy is proposed, which meets the [4] Qianggang Wang. “The LVRT control and fault current calculation,”
requirement of active power output and low voltage ride Chongqing University, 2015.
through to improve the operation efficiency of IIDGs. [5] Chengshan WANG, Xiaojing Sun. “An Improved Short Circuit
Calculation Method for Distribution Network with Distributed
2) Combining the LVRT strategy and grid-connected control Generations,” Power system automation, 2012, vol.36, no.23, pp.54-58.
strategy, the equivalent positive-sequence voltage controlled [6] WU Zhengrong , WANG Gang , LI Haifeng , PAN Guoqing , GAO
current source model of IIDGs is proposed. Xiang. “Analysis on the Distribution Network With Distributed
Generators Under Phase-to-phase Short-circuit Faults,” Proceedings of
3) Taking into account the nonlinear characteristic of the the CSEE, 2013, vol.33, no.1, pp.130-136.
equivalent circuit of IIDG, the symmetrical component method [7] Xiujuan Yang, Xiaohe Liu, Fang Zhang. “Model of Large PV and Its
and iteration method are employed to solve the fault current Applications on Power System Analysis,” Proceedings of the CSEE,
2011, S1, pp.19-22．
and voltage of distribution networks with IIDGs.
[8] Zhilong An. Research on the integration control strategy and low voltage
ride through technique of PV. North China Electric Power University.
REFERENCES 2012
[1] Jorge Luis Sosa, Miguel Castilla, Jaume Miret, “Control Strategy to [9] BI Tianshu, LIU Sumei, XUE Ancheng, YANG Qixun. “Fault
Maximize the Power Capability of PV Three-Phase Inverters During Characteristics of Inverter-interfaced Renewable Energy Sources,”
Voltage Sags.” IEEE Trans. on Power Electronics, Vol. 31, No. 4, April Proceeding of the CSEE, 2013, vol.33, no.13, pp.165-171.
2016, pp.3314-3323. [10] The Standardization Administration of China, “ Technical requirements
for connecting photovoltaic power station to power system,” Beijing:
Standards Press of China. 2012