Abstract: To enhance the reliability of mine mechanical and electrical systems,detect and prevent potential failures,improve mining production efficiency,ensure worker safety,and maintain normal equipment operation,a fault information data diagnostic system was established to collect fault data from mine mechanical and electrical systems.A Bayesian fault network was constructed and converted into a binary decision diagram using the ITE structure for qualitative and quantitative analysis of system faults.The results show that the constructed model achieved a diagnostic accuracy rate of over 98 % across different mining software platforms.In an actual evaluation of a mine mechanical and electrical system,insulation aging or damage,overload or overheating,electronic component failures,and internal short circuits were found to have the highest impact on fault,approximately 0.972.This significantly improved the accuracy and efficiency of fault diagnosis in mechanical and electrical systems and provided an effective assessment of underlying fault nodes.The findings can serve as a reference for fault diagnosis and reliability assessment in similar mechanical and electrical systems.