Life prediction of non-polar Dc circuit breaker based on SSA-Elman neural network

As an important switching element, non-polar Dc circuit breaker will cause serious consequences to Dc power network and electrical equipment once it cannot be reliably switched off. It is important to study the life prediction model of non-polar Dc circuit breaker to improve the safety of using electricity. To further improve the forecast accuracy of the model, the Sparrow Search Algorithm (SSA) is proposed to optimize the Elman neural network. Firstly, the data at 32A and 63A were processed and analyzed by MATLAB, and it was concluded that the cumulative arc burning energy and the cumulative burning of fixed contact amount acts as input characteristic quantities to the model. Secondly, Elman and SSA-Elman neural network are constructed to predict the electrical life of non-polar Dc circuit breakers. Finally, absolute error, relative error, mean absolute error, root mean square error and mean absolute percentage error are used as evaluation indexes of Elman neural network and SSA-Elman neural network. The results show that the error of SSA-Elman neural network is smaller than that of Elman neural network. The results show that the SSA-Elman neural network prediction model can improve its accuracy.