| In Industry 4.0 environment,when realizing the intelligent textile production,data are needed to connect the factory as a whole.And textiles from the market planning,design,manufacturing,sales and other life cycle process will produce a large number of structured and unstructured data,forming the textile production big data.The traditional database technology has been difficult to effectively store,manage and analyze these data which not only have large quantities,complex structure but also have a strong time fluctuation.Therefore,it has important theoretical and practical significance to study and apply the textile production big data technology.Based on the large amounts of data generated in the textile production process,this paper presents an overall architecture of the big data application platform for textile production and puts forward a data storage strategy of the big data application platform which based on multi-table schema and No SQL database to achieve separated storage and management of real-time data and historical data.At the same time,under the big data platform it studies the massive historical data distributed processing technology based on Map Reduce technology.On this basis,it develops and realizes a weaving production monitoring and due date forecasting system based on big data platform.The system can real-timely display the running state of the production equipment based on the date in the big data platform,and can through the big date analysis to obtain the information such as the efficiency of the production equipment,therefore,to achieve a more reliable forecasting for the enterprise orders' due date.The main research results and innovations include:1.Research on data storage and management strategy in big data environment.This paper proposes a data storage strategy based on multi-table schema and No SQL database aiming at the problem of huge and complicated data in the textile production process and deficient management of the traditional database.It separates the real-time data from the historical data,and adopts the traditional database multi-table schema to store the real-time data,and uses the data migration technology to transfer the historical data to the No SQL database to realize the storage and management of the mass production data2.Research on the construction technology of distributed big data platform.In big data environment,due to the serious performance problems of the traditional centralized processing method in data processing efficiency and reliability,this paper proposes a distributed storage platform based on the Mongo DB database and builds the distributed storage platform of historical data,and combines the Map Reduce parallel computing technology to realize the calculation and analysis of the massive historical data.Finally it designs and constructs a big data platform for textile production,so as to provide a complete platform environment for enterprises to realize the analysis and management of production process.3.Production process monitoring and management based on big data platform.Based on the actual production environment of weaving enterprises,it gives the overall structure of weaving production monitoring and due date forecasting system under big data platform,and using VS2012 development tool to realize the main function of the system.The system shows the running state of the production equipment in a real-time display,at the same time,according to the historical order information with big date analysis technology it can realize a more reliable forecasting of the enterprise orders' due date,so as to help the enterprise to analyze and manage the production process better.In this paper,a textile production big data platform is proposed and combined with the enterprises' practical production environment,it has verified the feasibility and effectiveness.The big data platform realizes the monitoring and management of the production process in textile enterprise and provides the decision support for the enterprise management. |
PDF Full Text Request