A CNN Based Approach with Identity Mapping Module for Mechanical Fault Diagnosis of High Voltage Circuit Breaker

Abstract

In order to improve the accuracy and generalization of mechanical fault diagnosis of high voltage circuit breaker, a CNN with identity mapping module based approach is proposed. Six acceleration sensors are installed at specific positions of the circuit breaker to collect comprehensive vibration signals. A mechanical fault diagnosis model is established based on convolution neural network with identity mapping module. After preprocessing such as down sampling and data splicing, the input signals are analyzed to extract feature information and identify mechanical fault. The experimental results show that the proposed method has better performance in mechanical fault detection compared with traditional CNN method.