In view of the complex causes of belt tearing fault of coal mine belt conveyor and the difficulty of early accurate early warning of existing diagnostic methods, the cause analysis and fuzzy diagnosis method of belt tearing fault of coal mine electromechanical transportation are studied. The main causes and correlation of tearing fault are analyzed, and the empirical mode decomposition method is used to adaptively extract multi-scale fault features from non-stationary vibration signals of equipment. A fuzzy logic diagnosis model is constructed to deal with the uncertain relationship between features and fault types, so as to realize the quantitative evaluation and early identification of potential tearing risk. Through the long-term application of the main transportation system in Xiayukou Coal Mine, this method can effectively warn the key fault symptoms such as roller damage and foreign body impact, locate accurately, provide clear decision support for on-site maintenance, and significantly improve the preventive maintenance ability and operation reliability of the transportation system.