A Research on Emergency DC Power Support Based on DCOI

doi:10.4236/epe.2013.54B166

Paper Menu >>

Journal Menu >>

Energy and Power Engineering, 2013, 5, 869-872

doi:10.4236/epe.2013.54B166 Published Online July 2013 (http://www.scirp.org/journal/epe)

A Research on Emergency DC Power Support

Based on DCOI

Fan Fan, Yingmin Zhang, Xingyuan Li, Wei Ma

School of Electrical Engineering and Information, Key laboratory of Smart Grid Sichuan University, Chengdu, China

Email: [email protected]

Received October, 2012

ABSTRACT

An integrated scheme of emergency DC power support which is based on dominant center of Inertia DCOI is proposed

to improve power system transient stability. The inter-area speed differences equivalent by dominant center of inertia

alleviates the main weakness o f the traditional signal which is made of the freq uencies at both sides and is represen t of

abnormal information of the nonlinear dynamic behavior of the power system based on a model in which two AC sys-

tems are connected by a HVDC Link. Sichuan Power Grid connected with Southwest Power Grid by Deyang-Baoji

HVDC project is used to testify the method. The simulation results show that the control strategy can significantly ad-

vance the transient stab ility of AC/DC system through extended equal area criterion (EEAC).

Keywords: DCOI; Speed Differences; EEAC

1. Introduction

The main appeal of HVDC (High Voltage Direct Current)

with long-distance and large capacity competence for

power transmission is the prominent ability to regulate

reactive and active power. HVDC improves an efficient

and robust approach to transient stability by modulating

DC power into AC grid rapidly, which will remedy power

imbalance between sending-grid and receiving- grid in

order to advance transient angle stability and improve

last-low-voltage level or voltage vibrating conditions [1].

Emergency control is a most significant measure

against large disturbances which are the gravest threat to

the stability to transient security of power systems. Thus,

it becomes necessary to choose the most useful DC

modulation signal from the huge amount of in- formation

for the system stability control services [2]. This paper

identifies and selects the “dangerous generators” to make

the dominant center of inertia while discarding all “harm-

less” ones by appraising the kinetic energy as the index

in the during-fault period respectively. The inter-area

speed differences equivalent by dominant center of iner-

tia can effectively reflect the abnormal information of the

nonlinear dynamic behavior of the power system based

on a model in which two AC systems are connected by a

HVDC link. Accordingly it is as the controller input sig-

nal [3-6].

DC power lift / drop-back can provide certain transfers

of power for the AC system to ensure the stability of the

system by modifying the instructions of the DC system

power to increase or reduce the DC transmission power.

For instance, when there happens a generator loss in the

sending-grid system, DC power drop-b ack can reduce th e

DC power to balance the system's active power shortfall

[7].

In this paper, an emergency DC power support com-

prehensive strategy is devised by combining DC power

lift / drop-back with the dominant center of inertia speed

deviation signal to modify the power set value jointly.

Taking Sichuan power grid for background, it indicates

that the signal deriving from the dominant center of iner-

tia is more efficient than the traditional signal made of

the frequencies at both sides after comparative analysis

of the system characteristics of various operating condi-

tions under typical faults.

2. Mechanism of Edcps Comprhensive

Strategy

In the transient process, the separation of the system is

not decided by the energy of the whole system. The un-

stable situation of the system is caused by a small num-

ber of out-of-step generators seriously deviating from the

system and most of the generators with each other can

still maintain synchronous operation. If all seriously dis-

turbed crew in the system is stabilized, the system is sta-

ble; Otherwise, the system is unstable. Thus, th e stability

of the system is converted to determine the stability of

the severely disturbed generators. During a failure, tran-

sient energy injected into the system is converted into the

F. FAN ET AL.

870

absolute kinetic energy increment of each generator set.

Kinetic and potential energ y convert into each other dur-

ing the process of power angle swing [8,9]. Under the

center of inertia (COI) coordinates each generator rotor

equations of motion are shown as follow:



me COICOI

COI

iii ii

MPPPD











 







(1)

where

MT : inertia time constant of system

Mi : inertia time constant of generator



I : power angel of generator

ωi : angular velocity deviation

Pmi : input mechanical power

Pei : output electromagnetic power

PCOI : accelerating power of COI

Transient energy of the system is expressed as follow:

COI

2COI

iimi eii

ii T

VMPPP

MPPP



























































(2)

Take the derivative of V as the change rate of the

transient energy:

 

me COI

COI COI

iiii i

iiiii

VMPPP

ttM





 























 













 





















(3)

V Consists of two parts which are the rate of relative

kinetic energy into potential energy and the rate of

potential energ y into relative kinetic energy. If change the

rate of the transient energy to be less than zero or

fluctuate around zero and gradually decreases over time,

the system has a larger stability margin. The kinetic

energy of the relative motion between the center of

inertia of the critical group and the remaining group can

truly reflects out of sync information under dual-machine

equivalents mode:



 (4)

Kinetic ene rgy is expres sed by th e speed and according ly

the equivalent of two machine speed deviation can be

used as a feedback signal to realize emergency control.

And the DC power modulation amount generated by the

feedback signal is represented as follow:





PK K





 

(5)

Early in the large disturbance, the generators absorb-

ing kinetic energy deviate from the synchronous speed

and gradually evolve into the relative movement of the

two generator group. The greater the relative kinetic en-

ergy of the two generator groups, the greater the likely-

hood of system instability [10]. After large disturbance,

all generators can be divided into two groups, namely

severely disturbed groups (S groups) and remnant groups

(R groups), and then the whole system can be equaled

into dual-machine-unstable model. For the time-varying

two-machine system, the NRP (not return point) which

decides the stability is the DSP (dynamic saddle point)

where the value of image acceleration power is zero.

During the process of the first swing, first determine the

moment when acceleration kinetic energy of the single

machine for critical group reaches a maximum [11,12].

And then selecting the generator maximum acceleration

kinetic of this moment as the benchmark, participation

factor is defined as the ratio of each generator accelera-

tion kinetic energy with the reference value；For remnant

groups, first calculate each generator deceleration kinetic

energy the moment disturbed trajectory get through the

DSP and take the same critical group generator accelera-

tion kinetic energy maximum value as a reference,

therefore the participation factor is defined as the ratio of

each generator deceleration kinetic energy with the ref-

erence value. To make the formation of the dominant

center of inertia (the Dominant Center of Inertia, DCOI),

selected generators of which participation factors are

greater than 0. 5. According to formula (1), formula (2),

formula (12), the control input signal of the emergency

power support is integrated by speed deviation informa-

tion extracted from the two DCOIs.











(6)











(7)

where:

F. FAN ET AL. 871

Mi : inertia time constant of generator in A group

Mj : inertia time constant of generator in B group

MA : inertia time constant of the equivalent genera-

tor A

MB : inertia time constant of the equivalent genera-

tor B



i : power angel of generator in A group



j : power angel of generator in B group



A : power angel of the equivalent generator A



B : power angel of the equivalent generator B

ωi : angular velocity deviation in A group

ωj : angular velocity deviation in B group

ωA : angular velocity deviation of the equivalent

generator A

ωB : angular velocity deviation of the equivalent

generator B

3. Control Effect

Taking Sichuan power grid in the summer peak load pe-

riod for background, while Debao HVDC system operat-

ing in monopolar mode, this pap er focuses on the impact

of EDCPS strategies on transient stability in Sichuan

power grid. EDCPS strategies are as follows: 1) DC

modulation based on the traditional inertia center (TIC);

2) the speed deviations modulation based on the dominant

center of inertia (DCOI); 3) the comprehensive EDCPS

strategy based on the dominant center of inertia the speed

deviations modulation cooperating with DC power

upgrade / drop back(CEDCPS).

Consider the following fault: the fault-free outage of

generating unit in Ertan Hydropower Station. Investigate

the kinetic energy of the generator of the critical group

and the remaining gr oups respectively. Kinetic energy of

Ertan unit increases the maximum, which is followed by

the order of the Jialing, Jintang factory, Jiangyou,

Pubugou. The detailed information is shown in Table 1.

It shows that Ertan, Jialing, Panzhihua and Jintang plant

units satisfy the conditional and take them to form the

dominant center of inertia of group A. Compare the

effect of the three emergency power support strategies

for improving the system transient stability through the

following simulation [13].

Table 1. The largest accelerating power of generator ruing

the fault period.

Unit name Acceleration power Unit name Acceleration power

Ertan

Jialing 12.282 2

10.782 0 Tianwan

Ziyili 1.666 5

1.057 7

Panzhihua 6.647 2 Shiziping 0.981 1

Jintang 6.267 5 Xiaotianhu 0.825 2

Pubugou 4.198 5 Tianlonghu 0.618 8

Fuxi 4.089 1 Weizhou 0.518 8

Baozhou 3.799 0 Xuecheng 0.313 8

Baozhu 2.599 8 C hib usu 0.265 7

Figure 1 shows the degree of oscillation of curve 2 is

stronger than curve1, which is also consistent with the

fact that the kinetic energy of generator units disturbed

seriously is lager than the other s .

The fault results in not on ly the serious system oscilla-

tion but also the power shortage which is up to 320 MW

in Sichuan Power Grid. The power shortage is made up

by droping the DC power. The three DC power control

strategies all can effectively drop the DC power to en-

hance the transient stability of the system. Analyzing

Table 1, Figure 2 and Figure 3 it can be obviously seen

that the improving effect of policy 3 is the best and

Strategy 2 is better than Strategy 1.

Figure 2 tells the trend of the Ertan unit power angle

and Figure 3 tells the trend of the Huangyan - Wanxian

active power by different control strategies in the post-

fault.

The oscillation in system was caused by the serious

fault during the angle-swing-up course. Strategy 2 taken

the DOCI speed deviation signal which can effectively

20151050

-0.01

-0.02

-0.03

-0.04

-0.05

-0.06

-0.07

-0.08

2 DCOI

1 TIC

Figure 1. Moudulation SIGNAL Curve s(HZ).

20151050

Figure 2. Power angle curves of ertan under different mod-

ulation signal for EDCPS (Degree).

F. FAN ET AL.

872

20151050

2,000

1,500

1,000

500

Figure 3. Active power of Huangyan-Wanxian lines (MW).

reflect the inter -area oscillatio n information provides more

damping compared to strategy 1 which is limited by lo cal

information. Strategy 3 with the same ability of providing

damping as strategy 2 makes up the power shrotage to

restore the system to a stable operating point close to the

pre-failure stable equilibrium point by droping DC

power.

4. Conclusions

This paper proposes a wide area information filtering and

integration method ba sed on participation factor which is

defined by relative kinetic energy increase during the

transient pro cess. It provides an effective wa y for screening

and selection of the feedback control signal from the

massive information. Compared with traditionas signal

modulation, the selected signal with higher accuracy can

not only reflect the global features of the system, but also

greatly reduce the amount of computation. After a com-

parative study of the system characteristics of various

EDCPS under typical faults, it indicates that the DOCI

speed deviation as a control signal can significantly en-

hance the inter-area damping. The emergency power

support comprehensive strategy based on the dominate

center of inertia speed deviation modulation cooperating

with DC power increase / back drop further develop and

utilize the potential of DC modulation to maintain system

transient stability. It provides a new way for detection

and effective correction of instability threat caused by

severe disturbance.

5. Acknowledgements

This work is supported by National Natural Science

Foundation of China (No. 51037003) and National High

Technology Research and Development Program of

China (863 Program). At the same time, the authors

ould like to appreciate all research members of the re-

search team for valuable discussion and investigation.

REFERENCES

[1] X.-Y. Li, “High Voltage Direct Current Transmission

System Operation and Control” Beijing: Science Press，

1998.

[2] E. Bristol, “On a New Measure of Interaction for Multi-

variable Process Control,” IEEE Transaction on Automa-

tion Control,Vol. 11, No. 1, 1966, pp. 133-134.

doi:10.1109/TAC.1966.1098266

[3] P. X. Zhang, Y. J. Cao, H. F. Wang, et al., “Application

of Relative Gain Array Method to Analyze Interaction of

Multi-Functional FACTS Controllers,” Proceedings of

the CSEE, 2004, Vol. 24, No. 7, pp. 13-17．

[4] J. Y. Zhang and Y. Z. Sun, “A Vulnerable Cut Sets Based

Sensitivity Analysis Method for Locating Emergency

Control,” Power System Technology, Vol. 31, No. 11,

2007, pp. 1-7.

[5] X. M. Mao, Y. Zhang and L. Guan, “A Novel WAMS

Based HVDC Modulation Method,” Automation of Elec-

tric Power Systems, Vol. 31, No. 7, 2007, pp. 45-49.

[6] J. B. He, L. C. Li, H. X. Chen, et al., “Selection of Feed-

back Signal for Power System Damping Controller Based

on Wide Area Measurements,” Automation of Electric

Systems, Vol. 31, No. 9, 2007, pp. 6-10.

[7] W. D. Yang, Y. S. Xue, Y. Jing, et al., “Emergency DC

Power Support to AC Power System in the South China

Power Grid,” Automation of Electric Power Systems，

Vol. 27, No. 7, 2003, pp. 68-72.

[8] Y. Zhang, L. Wehenkel, P. Rousseaux, et al., “SIME:

Ahybrid Approach to Fast Transient Stability Assessme nt

and Contingency Selection,” Electrical Power Energy

Syst, 1997, Vol. 19, No. 3, pp. 195-208.

doi:10.1016/S0142-0615(96)00047-6

[9] Y. S. Xue, “A Critical Comparison of Various Methods

for Transient Stability Assessment-Part Two Barrier Point

and the Observation Point,” Automation of Electric

Power Systems， Vol. 25, No. 11, 2001, pp. 1-7.

[10] M. C. Liao, G. L. Cai and Y. J. Zhang, “Transient Volt-

age Stability of Received Power Grid in AC/DC Hybrid

Power Systems,” Power System Protection and Control,

Vol. 37, No. 10, 2009, pp. 1-4.

[11] H. J. Wu, Y. H. Wang, X. Y. Li, et al., “Study on DC

modulation for AC/DC hybrid transmission system based

on comprehensive performance index,” Power System

Protection and Control, Vol. 38, No. 22, 2010, pp. 68-73.

[12] M. M. Xu, X. Y. Li, Y. H. Wang, et al., “Impact on the

Damping Characteristics of Sichuan Power Grid by De-

bao DC Modulation,” Power System Protection and Con-

trol, Vol. 38, No. 23, 2010, pp. 141-146.

[13] T. S. Xu, B. J. Li, Y. H. Bao, et al., “Optimal Parame-

ter-Setting of Under Frequency and Under Voltage Load

Shedding for Transient Security,” Automation of Electric

Power System, Vol. 27, No. 22, 2003, pp.12-15.