Design And Implementation Of Plant Production Process Control System Based On MES

MES (Manufacturing Execution System) is the management information system between the program management level and the industry control level for workshop production, performing as the basis to achieve the goal of agile manufacturing within an enterprise. MES provides the plant management personnel with current status of the implementation and tracking of production plan together with other resources. MES acts as the intermediary between the business management and manufacturer’s shop floor control, filling the gap in between, which means that it is an effective tool to manage and control plant production.With the growth in production scale and the increase of its complexity, effective performance monitoring, error diagnosis and quality control has become the key to ensure production safety, to improve product quality and to increase the profit. Due to the fast growth of information technology, more and more variables could possibly be measured and processed during the plant production process. In other words, the problem is now how to filter the massive amount of information and find the most relevant data, so that the safety and reliability of the execution of the process could be increased. The statistical performance monitoring is developed in this context, which attracts a lot of attention.The statistical performance monitoring is developed based on the multivariate statistical theory. Through the analysis and interpretation of measured data, it recognizes the current status of the process, meanwhile, the abnormal conditions are monitored and identified, which assists the production by minimizing the loss from production error and increasing production efficiency.This article mainly focuses on the following aspects:1. Based on KPCA (Kernal Principal Component Analysis), analyze the advantages of non-linear performance monitoring and the similarity analysis is then introduced into the field of error diagnosis, suggesting the performance monitoring and error diagnosis methods combining both KPCA and pattern matching technology. Besides, improvements are made by targeting the existing problems in KPCA and similarity analysis. 2. By analyzing the characteristics and the process of plant production, a series of statistical performance monitoring and quality estimation software are developed, which are the implemented into the actual production process, with positive testing results, forming the foundation for the successful implementation of enterprise information platform and advanced control.