Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《High Voltage Engineering》 2014-08
Add to Favorite Get Latest Update

Distributed Diagnosis Algorithm for Transformer Fault by Dissolved Gas-in-oil Parameters Analysis

ZHONG Yuanchang;WAN Nengfei;XIA Yan;ZHANG Liang;QIAO Jing;College of Communication Engineering,Chongqing University;College of Accounting,Chongqing Technology and Business University;  
To improve the speed and accuracy of fault diagnosis for power transformer, we proposed a distributed diagnosis algorithm based on the hydrocarbon parameters in transformer oil. The first diagnosis step utilizes the quantum behavior of support vector machine(SVM), namely, the support vector machine(SVM) is used to classify power transformer faults, and the SVM parameters are optimized by an improved genetic algorithm with the behavior of quantum in the classification process. Moreover, adopting results obtained from the first step, we used K-nearest cluster analysis to further classify the samples in the suspicious areas. Simulation results show that the proposed algorithm requires only 50 generations to obtain the best classification model, in contrast the ordinary genetic algorithm needs 170 generations. Plus, the combination of cluster analysis and SVM increases the accurate classification rate from 97.5% to 100%. It is concluded that the proposed algorithm can effectively improve both speed and accuracy of the fault diagnosis, and it is worth being applied to the fault diagnosis of power transformers.
【Fund】: 国家重点基础研究发展计划(973计划)(2012CB21520);; 第四届国家大学生创新性实验项目(101061118)~~
【CateGory Index】: TM41
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved