Mechanical remaining Life Prediction of Circuit Breaker based on long-term and short-term memory Neural Network

Abstract

The prediction of the residual life of the mechanical operating mechanism of the circuit breaker is of great significance for the real-time evaluation of the working state of the circuit breaker and the management, operation and maintenance of the transmission and distribution system. In this paper, a method for predicting the remaining service life of circuit breaker is established based on grey correlation analysis(GRA) and long-term and short-term memory neural network(LSTM). Firstly, the mechanical characteristic parameters such as overrun, opening distance and synchronous opening and closing of the operating mechanism are screened by GRA, and the characteristic variables which are strongly related to the number of actions of the circuit breaker are obtained. Furthermore, the characteristic variables obtained are used as the input of the LSTM prediction model to predict the remaining life of the circuit breaker. The example shows that the accuracy of the prediction model can reach 95% or more. This study can provide reference for the management, operation and maintenance of circuit breakers.