Typical Fault Simulation and Fault Diagnosis of High Voltage Circuit Breaker

High-voltage circuit breakers are among the most crucial equipment in power system. With the advancement of artificial intelligence, numerous advanced methods have been applied to identify fault types in high-voltage circuit breakers. This paper designs and acquires vibration waveform for common faults in circuit breaker operating mechanism, such as reduced spring pre-tightening force, excessively low operating voltage, loose bolts, and the stiction of iron core. Utilizing the denoising and feature extraction capabilities of the Channel-Wise Residual Shrinkage Module (DRSN-CW) with different thresholds in a deep residual neural network, a fault diagnosis model based on deep residual neural networks is proposed. This approach provides a novel perspective for advancing the application of fault localization in high-voltage circuit breakers.