Life Prediction of Residual Current Circuit Breaker with Overcurrent Protection Based on BP Neural Network Optimized by Genetic Algorithm

The service life of the residual current circuit breaker with overcurrent protection (RCBO) is long and its life information is difficult to obtain. In order to predict its life, a life prediction method based on RCBO performance degradation data is presented in this paper. Firstly, the residual operating current is determined as the degradation evaluation quantity of RCBO. An accelerated degradation test with temperature as the acceleration stress is carried out, and the performance degradation data are obtained. Secondly, the Optimized BP neural network (BPNN) model based on test degradation data is established. The genetic algorithm is used to optimize the BPNN model, which prevents the model from falling into the local optimal solution and improves the prediction accuracy. Thirdly, the pseudo failure life of RCBO under various stresses is predicted by the optimized model. After the test, the relative error between the predicted life and the true value is less than 10%, which meets the requirements of engineering applications. Finally, the service life of RCBO at normal temperature (25 °C) is deduced by the Arrhenius equation.