As a key component in the power system,the safe and stable operation of gas-insulated combined switch(GIS)is a necessary foundation to ensure the safety and stability of the power system.Aiming at the problem of poor per-formance of existing diagnostic methods in the face of unbalanced small sample data,this paper proposes a GIS mechani-cal fault diagnosis model based on WKELM.In this scheme,the GIS sound signal is preprocessed and the short-term en-ergy spectrum is calculated,and then the short-time energy and energy entropy as features are extracted,the WKELM mechanical fault diagnosis model is constructed,and the mapping relationship between the feature vector and the GIS run-ning state is mined.Experiments show that the proposed method can effectively improve the diagnostic performance in the face of unbalanced small sample data.
关键词
加权核极限学习机/GIS/故障诊断/不平衡小样本
Key words
WKELM/GIS/fault diagnosis/unbalanced small samples
维普
万方数据