A novel two-stage Dissolved Gas Analysis fault diagnosis system based semi-supervised learning

作     者：Xuemin Tan Chao Guo Ke Wang Fu Wan Xuemin Tan;Chao Guo;Ke Wang;Fu Wan

作者机构：College of Control EngineeringChengdu University of Information TechnologyChengduChina State Grid Chengdu Power Supply CompanyChengduChina China Electric Power Research InstituteBeijingChina State Key Laboratory of Power Transmission Equipment and System Security and New TechnologyChongqing UniversityChongqingChina 

出 版 物：《High Voltage》 (高电压（英文）)

年 卷 期：2022年第7卷第4期

页      面：676-691页

核心收录：

学科分类：080801[工学-电机与电器] 0808[工学-电气工程] 08[工学] 

基　　金：Research Initiation Project of Introducing Talents of Chengdu University of Information Technology,Grant/Award Number:KYTZ201902 National Natural Science Foundation of China,Grant/Award Number:51977017 

主　　题：fault diagnosis system 

摘      要：Dissolved Gas Analysis(DGA)is an important method for oil-immersed transformer fault ***,collecting labelled DGA data is difficult because the determi-nation of the transformer fault is time-consuming and expensive in the transformer substation,but DGA data without labels is easier to ***,the paper pro-posed a semi-supervised two-stage diagnostic system based DGA by using less labelled *** two-stage system includes a novel semi-supervised feature selection based Genetic Algorithm(GA)and Support Vector Machine(SVM)model(SSL-FS-GASVM)for selecting optimal features and a novel semi-supervised transformer fault diagnosis model based improved Artificial Fish Swarm Algorithm(AFSA)and SVM(SSL-IAFSA-SVM)for optimising the SVM ***,the performances of SSL-FS-GASVM and SSL-IAFSA-SVM models are tested and compared with traditional supervised diagnostic models combined with other optimisation methods,*** results show that the proposed two-stage system works in optimising features and parameters and has strong robustness in solving small sample classification problems.