Study On Dynamic Preventive Maintenance Method Of Body Welding Line Based On Performance Degradation

With the development of industrial technology and the expansion of manufacturing enterprises,manufacturing technology has gradually evolved from semi-artificial and semi-mechanized to the all-round development of mechanization and intelligential.Highly sophisticated and expensive equipment was increasingly utilized in enterprise production lines.The operation of manufacturing equipment in the flow-type enterprise had the high safety performance and it kept continuous production.The performance degradation or the equipment failures cost much to maintenance equipment and it may also had the great impact on the production efficiency.Therefore,timely mastering the performance of each equipment and adopting countermeasures to maintain good production performance played a pivotal role for enterprises to sustain their competitive superiority in the fierce market competition.This thesis was based on Sichuan Science and Technology Projects "The Remote Monitoring System,Life Circle,Management,and Intelligent Maintenance of the Automatic Flexible Production Line "(project number: 2017GZ0060).The serial body welding equipment acted as the research object.For the lack of standard and quantitative models,which had selection of performance evaluation indicators,the relatively low accuracy of the mainstream methods for division and evaluation,and the major impact on the weights obtained in a single way which were affected by the subjective and objective factors,this thesis analyzed the partitioning method of the equipment’s performance.Combining AHP and information entropy theory,this thesis proposed an approach to get the combination weight of each index,and then further improved the conventional K-means clustering algorithm.For the historical monitoring data,a scientific and rational evaluation of the experiment’s performance during the entire cycle was performed,the maximum cluster radius and data collecting window were defined,and the evaluation results of the equipment were divided into intervals to provide a basis for evaluating the real-time performance of the equipment.Secondly,the development of dynamic preventive maintenance methods for welding equipment defined the initial variance factors.According to the different performance intervals,this thesis proposed a combination of preventive maintenance strategies for the single welding equipment under the constraint of performance degradation domain and maintenance cost.The dynamic programming methods used to formulate a dynamic preventive maintenance method for the serial welding production line.Finally,the monitoring data provided by the enterprise was adopted in the modeling and simulation experiment so as to verify the rationality and accuracy of the method.Based on the development of the dynamic preventive maintenance methods for the performance degradation of equipment,this thesis had the theoretical guidance and practical application value for the serial production line of similar equipment.