doi:10.16186/j.cnki.1673-9787.2021010112

Received:2021/01/24

Revised:2021/03/19

Published:2022/03/15

Research on fault diagnosis method of MCCB based on multi data fusion

LI Kui1,2, LIANG Qiming1,2, ZHAO Chengchen1,2, HU Bokai1,2, MA Dianliang1,2, ZHAO Weizhuo1,2

1.Key Laboratory of Electromagnetic Field and Electrical Apparatus Reliability of Hebei Province, Hebei University of Technology, Tianjin  300130 , China；2.State Key Laboratory of Reliability and Intelligence of Electrical Equipment, Hebei University of Technology, Tianjin  300130 , China

Abstract: In order to improve the accuracy of MCCB fault diagnosis,a fault diagnosis method of MCCB based on multi information fusion was proposed. Firstly,empirical mode decomposition(EMD) was used to decompose the MCCB closing sound and vibration signals. Then,the IMF envelope energy entropy could be extracted as eigenvector and be put into LIBSVM(library for support vector machines) to obtain the basic probability assignment. The total classification accuracy of LIBSVM test samples was taken as a fixed weight to form the weighted probability assignment of sound and vibration signals. Finally,the weighted probability assignment of sound and vibration signals was integrated by using D-S evidence theory to obtain the fault diagnosis results of MCCB. Under laboratory conditions,the test data of MCCB in three different states including normal,loose installation and main spring fracture were obtained,and the diagnostic analysis was carried out. The results showed that the accuracy of fusion diagnosis was higher than that of single information diagnosis.

Key words:MCCB;empirical mode decomposition;information entropy;LIBSVM;D-S evidence theory

 基于多信息融合的塑壳断路器故障诊断方法研究_李奎.pdf