Quality Big Data Analysis Of Intelligent Manufacturing Process And Collaborative Optimization

Since the 1970 s,there are increasingly scarce of non-renewable energy,such as gas,coal,oil.In today's high speed development,global energy demand is growing fast which means that the reasonable development of existing energy and utilization of new energy become urgent.Solar energy is one of the most widely used new energy,experts predict that by 2050,more than 35% of the human energy will be occupied by the solar energy.While,manufacture industry as the world's measure of national comprehensive strength,directly embodies the productivity level of a country or a region.On behalf of a country's overall competitiveness,manufacture industry is an important factor of the difference between developed countries and developing countries.In recent years,as to big data,Internet of things,and cloud manufacturing as the main driving force of the rise of the new revolution of manufacture industry.Represented by Germany "Industry4.0" projects,the United States(Industrial Internet),Japan(Industrial Value Vhain Initiative,IVI),the European Union,push the manufacturing revolution with intelligent.China also started in 2015 "Made in China2025" strategy,and put forward to depth fusion between the new generation of information technology and manufacturing.In this paper,the research emphasis is focus on the quantification,optimization and intelligent of the intelligent production line of solar cells,which satisfies the trend of energy development,and stays in line with the key areas in the "Made in China 2025".This study introduces the research background of Big Data Analysis of Intelligent Manufacturing Process and Collaborative Optimization,include the industry big data,data correlation analysis,modeling technology,the quality of calibration,quality prediction and forecasting oriented manufacturing services adaptive collaborative optimization.This research can be summarized as the following sections:Firstly,the modeling of the intelligent production line system based on the Place Refinement Timed Petri Net(PTPN)is constructed after the research of Discrete Event Dynamic Systems and intelligent production line system.In the case study,the solar cell intelligent production line systems are divided into several levels(characteristic /unit/system level)and the Boundedness Properties,Liveness Properties and Reachability Analysis are used to verify the validity of the modeling method.Secondly,the Multi-level welding quality fault discovery of an intelligent production line by using Taguchi Quality Loss Function and Signal-Noise Ratio(QLF-SNR).Through the study of the measure the quality of service features in intelligent production line based on QLF,the feature selection methods are chosen to build the intelligent production line Multi-level welding quality fault discovery.By calculating the deviation degree of the quality features and making the threshold to distinguish the abnormal and normal samples,this method makes the fault discovery more targeted which can reduce the noise interference and improve the accuracy of the discovery to some extent.A case study has been test the method,and in the case study,the Unit level and System level to feature selection technology(Relief Algorithm)are introduced to calculate the threshold of abnormal and normal samples.And then,the mulit-kernel Support vector machine and D-S theory(SVMs-DS)method to evaluate service quality is proposed in the paper.The BPAs are put into the D-S method as independent evidence.The Dempster rule and thresholds are introduced to the algorithm to evaluate the predict result.The SVMs-DS is found to be superior and more accurate for process service quality evaluation.This method combines three SVMs in parallel for service quality evaluation based on kernel functions.Take each SVM as independent evidence;the evaluation results can be obtained by combining the three individual SVM results and making the ultimate decision by employing D-S evidence theory.The experimental results show that SVM-DS elevates the accuracy and stability of the service quality evaluation.After that,the research on the Adaptive Collaborative Optimization is carried out.Through the analysis of the constraints of the welding process operation pattern and the difficulties of the adaptive collaborative optimization,Augmented Lagrangian Coordination(ALC)is used to the optimization mechanism.And by improving the Genetic Algorithm,the solving progress based on ALC is optimized.The validity and accuracy of the algorithm are verified by the case study.And the last,this study puts forward ATWDP big data analysis platform for one well-known solar cells series welding machine enterprise to deal with the problems in data storage and processing.The ATWDP is designed based on the specific application requirements and addressed the storage,the precondition,the analysis and the decision making of the big data.Test results show that the ATW platform meets the enterprise demands on the data storage and processing from the feasibility and the latency time.The contribution of this chapter can be summarized as follows:(1)This chapter proposes a new hybrid service-driven architecture ATWDP which combines both online and batch offline methods by processing the real-time data in Real-Time Layer and dealing with the large scale hypo-real-time data in the Batch Layer.(2)We have designed a parallel algorithm for quality evaluation which can handle gather the massive sensor data efficiently.Such a distributed mechanism contributes to the high dependability of the ATWDP.