The mine AC motor works in the harsh environment of high dust, high humidity and strong vibration for a long time. The faults occur frequently (such as start-up failure, three-phase current imbalance, etc.) and the traditional inspection diagnosis lags behind, so it is difficult to find hidden dangers in time. In view of the above problems, this study developed an intelligent fault diagnosis system for mine AC motors. Based on multi-sensor data fusion monitoring, combined with edge computing + cloud collaborative architecture, the system realizes online continuous monitoring of key parameters such as motor vibration, temperature and current, and uses deep learning models such as convolutional neural network (CNN), graph neural network (GNN) and Transformer to extract and classify fault features. The edge end of the system architecture is responsible for real-time preprocessing and lightweight diagnosis. The cloud platform performs big data analysis and digital twin modeling to provide global fault assessment and life prediction. At the same time, a comprehensive test platform for high and low voltage motors was developed to test and verify the system. The diagnosis system can give early warning in the early stage of failure, improve the efficiency, accuracy and reliability of fault detection, reduce downtime loss and maintenance cost, and provide effective support for intelligent sensor monitoring and predictive maintenance of coal mine equipment．