Available online at www.sciencedirect.com

Energy

Procedía

ELSEVIER

Energy ¡Procedía 14 (2012) 937 - 942

2011 2nd International Conference on Advances in Energy Engineering

A Comprehensive Decision-Making Method for Wind Power Integration Projects Based on Improved Fuzzy AHP

Siyuan Liu a*, Jianhua Zhang, Wenxia Liu, Ying Qian

North China Electric Power University, No. 2 Beinong Road Zhu Xinzhuang Changping District, Beijing, 102206, China

Abstract

Given the blindness in present wind power integration decision-making, the paper constructed complete index system considering the characteristics of the wind power integration. By means of introducing the tolerance matrix into fuzzy AHP, the judgment matrix is easy to test consistency. The paper used the improved fuzzy AHP to make decision on the wind power integration schemes. The calculation results show that the method is practical.

Keywords: wind power integration; scheme decision; index system; tolerance matrix; fuzzy AHP

1. Introduction

Because the wind power is intermittent and random, the wind farm integration has influence on the power flow. The influence degree depends on the type and the layout of the units. The impact on the power flow are also different when the wind farms choose the same unit type and the same installed capacity but their access to different position [1]. Because of the lack of a unified index system for wind power integration, the decision on the access point relies on experts experience or the "proximity principle". Thus, the access point is often not the best one. The establishment of index system for wind power integration decision-making has become a serious problem.

In recent years, a variety of decision-making mathematical method is widely used in the power system. The document [2] used the fuzzy comprehensive decision-making in the power transmission project, the disadvantage is that the determination of index weights are based on completely subjective experience. The document [3] proposed a method based on entropy weight fuzzy comprehensive evaluation, and applied it in the transmission network planning decision-making. This method can determine the weight

* Liu Siyuan. Tel.: +86-18600012626; E-mail address: liusy01309@yahoo.cn.

1876-6102 © 2011 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of the organizing committee of 2nd International

Conference on Advances in Energy Engineering (ICAEE).

doi:10.1016/j.egypro.2011.12.887

more objectively. But using the point value in the evaluation process can not accurately reflect the ambiguity and uncertainty of the decision-making factors. The document [4] introduced the triangular fuzzy number into the evaluation progress. But this method has failed to check the consistency of the judgment matrix, which cannot be proved real and effective.

The paper constructed a complete index system considering the impact of the wind power integration. Given the existing deficiencies, the tolerance matrix was introduced into fuzzy AHP, which is a fundamental solution to the problems of the matrix consistency test. Finally, the actual project recently completed show that the decision-making method is theoretical and practical.

2. Decision-making index system of the wind power integration

2.1. Index system

Considering the impact of the wind power integration, the paper simulated the steady-state power flow distribution, the transient stability, the N-1 line static security analysis and the short-circuit current levels, and considered some interested factors in the project. The decision-making index system of the wind power integration is as shown in Figure 1.

CoaiprriuasLve OMisiia-MLiinj afAJitraim* ]a[r;:a;ioa

4. A-

V.i\!n;u::i puurr SlKlJv Vnli

Dm diiULtiiica

A;; Volop

A;;. :cid rut

Civrrjoad !snr rumbfi

An Ati A); Af!

?oH-fai£r fuil-fllllL jociv iiatv

völlig* iüail ralr JHjdr* niJfi

^luljhflllKtt flUsJm TTUTftlD ni^^Jl

Fig. 1. The decision-making index system of the wind power integration

2.2. Index

• Maximum capacity of wind power integration^)

The wind farm integration has impact on the operation of the regional power grid scheduling, the reactive power compensation and the power quality to some extent. With increased access capacity, the stability of the system may also be affected. Therefore, the maximum capacity should be one of the evaluation factors. When such capacity of the power is integrated into the grid, the system should ensure that the power flow is convergent, and all nodes should be consistent with the corresponding operating voltage requirements.

• Steady-state flow distribution(A2)

Because the local power grid is weak, the power flow distribution will be affected by the wind power integration. It will cause the voltage fluctuation of the access point and nearby nodes, near-link

transmission power to be off-limit and other problems [5]. Therefore, the factor A2 can be divided into three quantifiable factors: voltage qualification rate(A21), net damage rate(A22) and load rate(A23) .

Among them, A21 is the voltage level of the nodes with the reactive power adjustment in the actual configuration after the wind power integrates into the grid.

A22 is the net damaging rate of the active power. Practical operation shows that the proper amount of the wind power often can optimize the power flow distribution, so the net damaging rate will be decreased. Thus, A22 can be used as a sub-factor in the decision-making progress. The smaller net damaging rate, the higher optimization levels of the power flow.

A23 can be divided into the underlying factors: average load rate(A231) and overload line number(A232). These two factors can fully reflect the transmission power change after the wind power integration. Calculations only consider the important lines in different voltage levels. Moreover, the economic load rate is from 30% to 70%.

• Transient Stability (A3)

Because China's wind farm location is further away from the load center, and the regional network structure is weak, so wind power integration decision-making must be taken into account for the transient stability analysis.

In present, most of the wind turbines use doubly-fed induction generator, wind turbine and the power system is a flexible coupling between, so appropriate integration can improve the transient stability. With the wind power capacity increasing, the stability may not be improved but deteriorated. Critical clearing time improvement rate is proposed, and calculated as follow:

1 T t -1

n =1 £ ^cm lccn0 xioo% (1)

T i=1 tCCTi0

Where, tCCTi9tCCTi1 are the critical clearing time before and after the wind power is integrated into the grid respectively. T is the number of the main lines in the local power grid. If n is positive value, it indicates that A3 is improved by the wind power integration. The higher value the more obvious improvements. On the contrary, it indicates that the stability is deteriorated.

• Reliability(A4)

Evaluation factor A4 is calculated by the N-1 static security analysis. When any line exits at run time, the substation bus voltage will be changed, the other routes of the transmission system will change the power, and the load shift is likely to lead to other lines overload, a serious fault will trigger a chain. Consider these aspects, A4 is divided into two indicators: post-fault voltage qualification rate(A41) and post-fault average load rate(A42). Index value is calculated the same with A21 and A231.

• Short-circuit current levels(A5)

The wind power integration makes a great contribution to the short-circuit current of the lines nearby, and it will be decreased with the distance [5]. The calculation of the short-circuit current can be used to check the rated breaking current of the switch and the dynamic stability of the disconnector and other electrical equipment.

Y is proposed to evaluate A5, the calculation formula is:

10 1 (2;

Where, I9 is the maximum limit of the current; Ii is the short-circuit value of the nodes i. T is the number of the selected bus nodes. • Economy(A6)

A6 is not only an important index for the investing party, but also is essential to judge different schemes for decision-making. A6 contains the cost of the construction and the operation. The cost of the

expanded or the new-built transmission lines and transformers and the switching equipment upgrade should be included.

3. Index weights determination

Fuzzy AHP based on the triangular fuzzy number can effectively improve the past method which rates the qualitative evaluation of the scheme by a point value, cannot reflect the vagueness and uncertainty of the factors accurately[5]«

The tolerance matrix is introduced into fuzzy AHP, not only can reduce the workload of the repeated re-construct the judgment matrix, but also ensure the consistency of the comparison matrix, which simplifies the process of the index weights determination.

3.1. The tolerance matrix

The basic idea of the tolerance matrix analysis method is to determine the matrix A= (aij) and amend the elements to make it satisfy the consistency conditions of the judgment matrix. The calculation steps are as follow:

• Construct matrix A: A = (a )nXn, aij = L aij = —;

• Let b.. = n na.,a ■ , obtain the tolerance matrix: B = (b..) , b = 1, b, = —; bH = bk ■ b,,;

\i=l j Vjlnxn' lj ' lj b j lk kj

Seek the index weight (Bj :

Bj =-j (j = 1,2, •••, n) (3)

Where, Cj = n fib]k (j = 1,2 •••, n)

3.2. Improved fuzzy AHP calculation steps

T is the number of the experts; Q is the number of the alternative schemes. Expert scoring uses the triangular fuzzy numbers, namely the numbers are based on the mean value m. Set up n is the number of the sub-factors associated with a decision-making factor.

• Construct the decision-making hierarchy based on the importance of different factors;

• Evaluate the index and schemes by experts comparison, determine the judgement matrix with different factors( N = (Nj )nxn) or schemes( F = (Fy )qxQ );

• T experts judge each element of the matrix, calculate the average and obtain the comprehensive judgment matrix, such as: a^ = 1/ T ® (a! + a2 +-----+ a[ij);

• Calculate the adjustment matrix Q:

Q = M x E (4)

Where, E is the fuzzy evaluation factor matrix; M is the matrix constructed by the middle elements:

E = ( eij )nxn =

1- (u21 - l21)/ 2m2

1- (u12 -l12)/ 2m1; 1

1 - (un1 - ln1)/2mn1 1 - (un2 - ln2 ) / 2mn

1- (u1n - L )/2m1n 1- (u2n l2n )/2m2n

Transform ß to a judgment matrix with diagonal 1: A = (ar )nxn, atj = 1, ar = ;

• Seek the index weight (Bj by the tolerance matrix method;

• From the bottom, calculate (Bj step by step to get the overall weight;

• Sort weights and choose the best scheme.

4. Example analysis

4.1. Project overview

Take the recent actual wind farm project in Jilin province as an example, the new index system and the improved fuzzy AHP are applied into decision-making. The regional power grid is shown in Figure 2:

Fig. 2. The regional power grid

The construction of wind farm C contains four phases. Phase I, II have installed 66 units, 1.5 MW wind turbines, a total of 99MW, which has been boosted to 220kV through the center booster station, and accessed to substation QA through a single line.

Phase III, IV plan to install 66 units, 1.5MW wind turbine. The wind power will increase by 99MW. Because the structure of the region is weak and there are lots of wind farms, the original transmission

line substation QA ~ substation SY has been overloaded. Therefore, three schemes are proposed considering the practical power system.

Scheme I: the wind power of phase III, IV is boosted to 220kV by the central booster station and sent to substation QA through the original 220kV line.

Scheme II: the wind power of phase III, IV is boosted to 220kV by the central booster station and sent to the 220kV bus of station CL through a new-built line, from wind farm C to station CL.

Scheme III: build 500/35kV transformer. The wind power of phase III, IV is boosted to 500kV by the central booster station and integrated into the transmission line, wind power base XY~ 500kV bus of substation CL by T-connection.

4.2. Application of improved fuzzy AH/ for comprehensive decision-making

In accordance with the calculation steps in 3.2, the comprehensive weight is obtained:

Mj = [0.33, 0.37, 0.30]

The calculation results show that the scheme II is the optimal one. Compared with the other two schemes, scheme II has the largest wind power integration capacity, the lowest net damaging rate and the greatest critical clearing time improvement rate. The decision is reasonable.

5. Conclusion

The paper constructed an index system for the comprehensive decision-making of the wind power integration. The system is proved effective and practical. The selected indicators are accessible and easy to quantify.

The improved fuzzy AHP, not only can reduce the workload of the repeated re-construct the judgment matrix, but also ensure the consistency of the comparison matrix, which simplifies the process of index weight determination.

Practical case analysis results show that the proposed index system is clear in theory and convenient to calculate as well as its results are intuitive to react the characteristics of different schemes, avoiding the blindness and the empiricism in decision-making progress, so it is practical.

References

[1] Hannu Laaksonen, Pekka Saari, Risto Komulainen. Voltage and Frequency Control of Inverter Based Weak LV Network Microgrid [C]. 2005 International Conference on Future Power Systems, 16-8 Nov. 2005 Page(s): 6PP.

[2] Wu Li, Wu Jie, Zhou Lerong. Fuzzy multi-objective comprehensive decision on transmission projects[J]. Power System Technology, 1999, 23(3): 19-22(in Chinese)

[3] Nie Hongzhan, LüPan, Qiao Yi, Yao Xiuping. Comprehensive fuzzy evaluation for transmission network planning scheme based on entropy weight method [J]. Power System Technology, 2009, 33(11): 60-4(in Chinese)

[4] Chen Dayu, Xiao Jun, Wang Chengshan. A FAHP-Based madm method in urban power system planning [J]. Power System Technology, 2003, 15(4): 83-8 (in Chinese)

[5] Chi Yongning, Liu Yanhua, Wang Weisheng. Study on impact of wind power integration on power system[J]. Power System Technology, 2007, 31(3): 77-81 (in Chinese)