GIS Fault Diagnosis Method Based on Multi-Source Feature Dual Decision Hybrid Classification Network

GIS high-voltage disconnect switches are important equipment in power grid operation. In order to avoid safety and economic losses caused by high-voltage disconnecting switch faults, this paper proposes a multi-source feature dual-decision hybrid classification network for GIS disconnecting switch fault diagnosis. First, the voltage, current and angle signals of the control motor are feature extracted and the model is trained. Then the multi-source feature dual decision hybrid classification network is utilized for fault diagnosis. The experimental results show that the method can effectively identify two common faults: unidirectional loosening faults and biphasic loosening faults.