Diagnosis, early warning and maintenance of mechanical and electrical equipment is an important means to ensure the continuity of production in coal preparation plant. With the reduction of mechanical and electrical maintenance personnel and the uneven technical level, the rapid upgrading of equipment, the lack of maintenance work, and the equipment detection methods mainly rely on manual detection, Luling Coal Preparation Plant has installed and implemented a set of intelligent on-line monitoring system to improve the reliability of key equipment operation. The system uses wireless temperature vibration sensors to collect data from 73 core devices, aggregates through wireless intelligent monitoring stations and uses edge computing for preliminary processing, and finally uploads to the SuperCare cloud platform. The platform integrates the adaptive alarm threshold model based on statistics and the equipment fault feature database, and realizes the real-time evaluation and intelligent diagnosis of the equipment running state. The application results show that the system runs stably and successfully warns early bearing faults of multiple equipment. After the system was put into operation, the unplanned downtime caused by equipment failure was reduced by 77.3 % year-on-year, which verified the effectiveness of the technical scheme in realizing predictive maintenance of equipment and ensuring production safety.