For the problem that the failure rate of related electronic equipment in operation is uncertain,resulting in difficult to identify the reliability.In this paper,an improved reliability evaluation system for electronic equipment is designed.The Mean Time Between Failure (MTBF-T) model is used to monitor the time characteristics of the power equipment,and the MTBF-T curve of the equipment is fitted by least squares.The operating life and failure probability of the equipment are predicted by analyzing the reliability development trend of the equipment.The MTBF-T model is fused with AMSAA technology to obtain the corresponding stage of the whole cycle of the equipment.The Long Short Term Memory (LSTM) algorithm is adopted to improve the accuracy of the model's identification and prediction of electronic equipment faults.The experiment results show that the accuracy of data quality verification by this algorithm is more than 90%,which indicates that the research system has strong practicability to solve the problem of improving the accuracy of fault discrimination.