Development of Condition Diagnosis Method for Gas Circuit Breaker with 1D-CAE

Abstract

Gas circuit breakers (GCBs) are commonly used equipment in transmission and distribution systems. GCBs are designed to protect the system from over voltage and over current. Conducting operations with damaged GCBs can cause expansion of failure parts and delay their restoration, therefore diagnostic methods for GCBs are essential to ensure a stable power supply. In this study, a one-dimensional computer aided engineering (1D-CAE) model was used to generate pseudo sensor data under different conditions. A multiclass classifier based on shapelets, which are waveform patterns discovered by machine learning, was developed, and generated sensor data was used for training and test. The results showed that the proposed method can accurately classify the location of anomalies in GCBs and provide explanations based on shapelets.