Quality Process Statistics System Based On Internet Of Things

After years of development,the scale of China’s manufacturing industry has increased rapidly,but most manufacturing enterprises are in the traditional processing mode,and there are problems such as over reliance on manual fault diagnosis.In order to promote the rapid transformation of China’s manufacturing industry as soon as possible and improve the digital and intelligent level of discrete manufacturing industry.Therefore,it is of great significance to study the application of statistical process control technology in quality process statistical system.Combined with SPC technology and support vector machine,this paper designs a quality process statistics system based on Internet of things,which realizes the statistical analysis of production process data and on-line fault diagnosis.The main research work of this paper is as follows(1)According to the needs of the enterprise and the actual situation of the production line,the overall architecture scheme of the quality process statistics system is designed.It is determined that the system architecture includes three main components: data acquisition side,server side and client side,and the technical schemes of each part of the system are proposed to realize the functions of data acquisition,storage,analysis and fault diagnosis.(2)The three-tier C/S architecture is adopted to develop the system client software to ensure the security of production process data.Aiming at the problems of real-time quality control and quality data persistence,the master-slave database architecture is designed to realize the client monitoring of quality process statistics system based on Internet of things.(3)The quality data analysis method based on SPC control chart is used to analyze the controlled state of the production process and the capacity of the production process.In order to solve the problem that traditional control chart discrimination needs to consume a lot of labor cost,this paper proposes an on-line fault diagnosis method that applies multi classification support vector machine to control chart pattern recognition.In this method,the feature of control chart is extracted by histogram of oriented gradient,and the SVM model based on Gaussian radial basis kernel function is trained.Finally,the case-based reasoning method is used to map different abnormal patterns to the corresponding fault points to complete the online fault diagnosis of the production process.Finally,in order to verify the system performance,the system is tested.The test results show that the quality process statistics system designed in this paper can realize the functions of quality data query and export,control chart generation,production process monitoring,fault diagnosis and alarm,and can achieve the expected design goal.