Research on Performance Degradation Prediction of Power Driver Considering Multiple Stresses

Power driver is one of the key components in electronic system, its performance degradation and function failure will affect the normal operation of the whole system. Therefore, it is important to study the performance degradation prediction of power driver. In this paper., the multi-stress accelerated degradation test of power drive device is designed., performance degradation parameters are collected., data are preprocessed by mean filtering., and a performance degradation prediction model of power drive device based on PSO-BP neural network is constructed., and the model is trained and tested by the test data. The results show that the root mean square error of PSO-BP neural network prediction is 4.3e-4., the prediction accuracy is high., and the prediction trend fits well with the test data., indicating that the prediction model is suitable for the accelerated degradation performance prediction of power drive devices., and provides a method for the health state prediction of electronic system.