Research On MES System Of Automoblie Powertrain Production Line Based On Lightweight Process Data

In the production process,the automobile powertrain machining production line will generate a large amount of production process data,production quality data and production path data including equipment sensor data and RFID production information label data.The characteristics of its data are sometimes correlation,massive,uncertain,and real-time.The system integration,data management,data storage of large-scale streaming data in modern factories,and data redundancy and data transmission in the system are difficult.It is an urgent problem to be solved in the intelligent automatic production line manufacturing execution system.In view of the above problems,this paper summarizes the research progress and existing shortcomings in related fields at home and abroad.Introduced path GID coding compression and process data clustering method to lightly optimize the massive process data to improve the efficiency of manufacturing execution system.The main research content of this paper is divided into the following three parts:(1)Construction of manufacturing execution system(MES)and establishment of data model.Establish an automated workshop system architecture,and design a system information exchange process based on RFID data to complete online automated production.On the basis of production process data,production quality data and production path data,a production process data cube model is established,which provides a basis for the subsequent data lightweighting method.(2)Data weighting method based on path wgid coding and clustering.The path coding based RFID path data coding method and the unsupervised high-dimensional data clustering method are introduced to solve the reference compression problem of high-dimensional data.After comprehensively comparing K-means clustering method,Hierarchical algorithm and self-organizing neural network algorithm,it is confirmed that K-means is used in the best inter-group difference and SOM algorithm for reference compression.The advantage of class separation is used,and the clustering result is optimized by SOM algorithm.(3)Automotive powertrain MES system data lightweight engineering application.Applying the lightweight process data method to the intelligent manufacturing line MES system,constructing the MES system process data lightweight framework,establishing an optimized database,and demonstrating the effectiveness of the lightweight process data method through examples.In summary,this paper introduces the K-means clustering method of machine learning and the deep learning self-organizing neural network(SOM)into the process data optimization of the vehicle powertrain manufacturing execution system,and carries out data weight reduction and database optimization.the study.This method is also useful for the field of industrial information system data management in aerospace,train and ship discreet manufacturing industries.