The Research And Implementation Of Industrial Manufacturing Equipment Monitoring System

Fault monitoring of manufacturing equipment is a particularly important part of manufacturing systems.Pre-judging the possible failures of equipment during the manufacturing process is an important means to effectively improve the reliability of the manufacturing system and reduce the cost of downtime.Modern equipment health monitoring relies on the PHM(Prognostics and Health Management)system,and equipment failure prediction is usually achieved by an expert-based mechanism-based model approach.Although the system has been verified with a large amount of data,due to different process requirements and manufacturing processes,the production system may have unobservable factors that make the monitoring system unable to adapt to the new manufacturing environment.At present,the fault maintenance of most domestic manufacturing equipment is based on regular maintenance and fault maintenance.The research of PHM system is only designed for the fault prediction of military equipment and aerospace equipment.It has not entered the civil manufacturing stage and cannot meet the large-scale batch.Fault prediction driven by mass equipment feature status data in a production environment.This thesis analyzes the common failure modes of industrial manufacturing equipment,and studies the fault prediction of industrial manufacturing equipment based on PHM technology,and designs an industrial manufacturing equipment monitoring system that uses the characteristic state data of massive equipment to analyze the equipment failure.The performance requirements of industrial big data driven PHM systems are designed to meet the required system architecture.At the same time,the modular design of the industrial manufacturing equipment monitoring system is implemented and elaborated.This thesis implements a data-driven industrial manufacturing equipment monitoring system,enabling the manufacturing system to self-learn based on historical experience data,spontaneously discovering equipment anomalies,realizing real-time fault prediction of equipment,thereby reducing accident rate and downtime costs,and changing the manufacturing system to equipment in China.The protection strategy stays in the situation of regular fault maintenance.At the same time,the hierarchical architecture of industrial manufacturing monitoring system for large-scale mass production environment is proposed.Combining the advantages of distributed structure and hierarchical structure,the cloud platform and edge computing hybrid network architecture are used to design the industrial manufacturing equipment monitoring system.The system can handle large-scale data analysis while improving the system's applicability and operational efficiency.Finally,the research results are applied to the fault prediction of the servo drive,and the servo drive can only be changed after the fault occurs.