Online Fault Classification Method for High Voltage Electronic Switchgear Based on Meta Learning

This article mainly studies the fault classification problem of high-voltage electronic switchgear. During the working process, high-voltage circuit breakers operate at high speeds and generate high-strength impacts, causing the mechanical structure between circuit breakers to experience increased friction between transmission components, mechanism movement blockage and jamming, and circuit breaker shortage after a large number of opening and closing operations. This will lead to problems of circuit breaker misoperation, refusal to operate, and refusal to close. This article proposes an online fault classification method, equipment, and storage medium for high-voltage electronic switchgear. The monitoring parameters of high-voltage electronic switchgear are randomly selected and segmented, the dataset is reconstructed, one-dimensional subsequences are obtained, a training set is constructed using one-dimensional subsequences, and a convolutional neural network is trained to obtain a classification model. The model uses a sliding window to recognize subsequences and performs pruning processing, improving data quality and solving the problem of persistent model training time and inability to improve classification accuracy caused by insufficient data volume in existing technologies.