Research On Product Quality Control System Based On Manufacturing Process Data

In the trend of continuously pursuiting high quality,the manufacturing enterprises are paying more and more attention to the quality control of the products during the industry competition.Accompanying the advancement of information technology,JOHN SON ELECTRIC enterprise is exploring the quality control method combining with information technology actively.In view of the existing problems of quality control methods being used in the magnetic tile manufacturing shop,this paper carried out the research and the design of quality control system and aimed at the stability and the improvement of product quality.The overall architecture and the business process of this system were designed from the aspects of process control,quality improve and information reform.To enhance the ability of quality control in the process of molding,a process framework of process monitoring and fault diagnosis was designed and a method for batch process monitoring and fault diagnosis based on Multi-way Principal Component Analysis was proposed.In this method,real time process monitoring and fault diagnosis were implemented by using the contribution chart and the improved Multi-way Principal Component Analysis method based on time-varying covariance.Besides,T squared chart’s ability of monitoring the small amplitude or potential fault was improved by recognizing the T squared statistic’s track based on Support Vector Machine classification algorithm.The method of process improve based on association rules was researched to meet the requirement of improving the quality of products in the system.To assist technologists to adjust and improve the process parameters,the regular pattern was mined by the FP-growth algorithm and the influence of process parameters on product quality was explained with the help of the technology of data cleaning and visualization.Finally,the database,functional module and the structure of network were designed.Then key modules of the prototype system were programed by using Java EE.