Fault diagnosis of power networks applying CE-SVM and fuzzy integral fusion

被引：0
|
作者：
Bian L. [1 ]
Bian C.-Y. [1 ]
机构：
[1] School of Electronic and Information Engineering, Heilongjiang University of Science and Technology, Harbin
来源：
Bian, Chen-Yuan | 1600年 / Editorial Department of Electric Machines and Control卷 / 20期
关键词：
Cross entropy; Distributed; Fault diagnosis; Fuzzy integral; Large power networks; Support vector machine;
D O I：
10.15938/j.emc.2016.02.016
中图分类号：
学科分类号：
摘要：
A fault diagnosis method for power networks based on posterior probability CE-SVM and fuzzy integral dynamic fusion was proposed. The aim was to solve the problem of division fault diagnosis inside the local network and for the tie lines connecting local network. Firstly, a graph partitioning method was used to the large power network into connected sub-networks with balanced working burdens. Historical data was applied to train local CE-SVMs and local CE-SVM modules were selectively triggered according to local alarm information. Fuzzy integral fusion department constructed by fuzzy densities dynamic adjusted algorithm was used to fuse posterior probability of the tie lines fault that outputted by local CE-SVM modules for tie line fault identification. The method can not merely diagnose the faults inside local network, but also solve the fault diagnosis problem of tie lines. The simulation results indicate that the proposed method is effective and have good fault tolerance under action information losing or unwanted operation of protector and circuit breaker. © 2016, Harbin University of Science and Technology Publication. All right reserved.
引用
收藏
页码：112 / 120
页数：8
相关论文
共 19 条
  • [1] Park Y.M., Kim G.W., Sohn J.M., A logic based expert system(LBES) for fault diagnosis of power system, IEEE Transactions on Power Systems, 12, 1, pp. 363-369, (1997)
  • [2] Cardoso G.J., Rolimj G., Applications of neural network modules to electric power system fault section estimation, IEEE Transactions on Power Delivery, 19, 3, pp. 1034-1041, (2004)
  • [3] Zhang Y., Zhang Y., Wen F., Et al., Power system fault diagnosis with an enhanced fuzzy petri net accommodating temporal constraints, Automation of Electric Power Systems, 38, 5, pp. 66-72, (2014)
  • [4] Zheng R., Zhao J., Zhao T., Et al., Power transformer fault diagnosis based on genetic support vector machine and gray artificial immune algorithm, Proceedings of the CSEE, 31, 7, pp. 56-63, (2011)
  • [5] Pellaco L., Costamagna P., Giorgi A.D., Fault diagnosis in fuel cell systems using quantitative models and support vector machines, Electronics Letters, 50, 11, pp. 824-826, (2014)
  • [6] Bi T., Jiao L., Yan Z., Et al., Graph partitioning method for distributed fault section estimation system in power network, Automation of Electric Power Systems, 35, 16, pp. 16-21, (2001)
  • [7] Shi D., Xiong G., Chen J., Et al., Divisional fault diagnosis of power grids based on RBF neural network and fuzzy integral fusion, Proceedings of the CSEE, 34, 4, pp. 562-569, (2014)
  • [8] Vapink W.N., The Nature of Statistical Learning Theory, (1999)
  • [9] Reuven Y.R., The Cross-entropy Method: A Unified Approach to Combinatorial Optimization, (2004)
  • [10] Pieter T.D., Reuven Y.R., A tutorial on the cross-entropy method, Annals of Operations Research, 134, 2, pp. 19-67, (2005)
12