Parameter Design And Optimization Based On Online Adjustment

With the rapid development of global economy and the continuous progress of science and technology,product and service quality has become the focus of competition among industrial enterprises around the world.It is also the key factor to gain competitiveness in the global market.With the development of information technology,the manufacturing processes are becoming more and more integrated,intelligent and complex,which put forward higher requirements for the quality control of manufacturing process.Based on the manufacturing product,the concerned problems in manufacturing processes are systematically studied by means of robust parameter design,online adjustment and statistical process control.Response surface method,Kriging model,quality loss function and other theories and methods are synthetically utilized in research process by means of system modeling,simulation analysis and case study.The main contents of the thesis are summarized as follows:(1)A robust parameter design strategy with uncertainty in input parameters of processes is proposed.Most of the existing robust parameter designs are based on the assumption that the input parameters of processes are constant and do not take into account the influence of the variation of the measurement system on the process input parameters.To solve this problem,this thesis proposes a robust parameter design strategy with uncertain input parameters.The unbiased estimation of regression parameters and the solution method of optimal control variables are given.The effectiveness of the proposed strategy is verified by an industrial example.(2)A robust parameter design strategy based on online adjustment is proposed.Most of the quality control methods carry out offline parameter design and online process control separately.In this thesis,on the basis of considering the uncertainty of input parameters,online adjustment and offline robust parameter design are systematically integrated,and the corresponding control strategy is designed.The effectiveness of the proposed strategy is verified by a case study.(3)A robust parameter design strategy based on variable selection is proposed.Considering quality loss and controllable variables' adjustment cost in manufacturing processes,a robust parameter design strategy is proposed based on variables selection.This approach aims to reduce the adjustment cost as much as possible on the premise of ensuring the product quality.The effectiveness of the proposed strategy is verified by an industrial example.(4)A robust parameter design strategy for multistage manufacturing processes is proposed.According to the characteristics of the multistage manufacturing processes,the transfer model is constructed by using the Kriging model,and the parameter setting of the multistage manufacturing processes is systematically optimized,including the design of offline control variables and realtime adjustment of online control variables.The approach integrates offline robust parameter design and online adjustment,and can effectively improve the efficiency of quality control in multistage manufacturing processes by a simulation study.(5)An online adjustment strategy based on pattern recognition of control chart is proposed.For the manufacturing processes which have adopted robust parameter design based on online adjustment,an online adjustment strategy based on pattern recognition of control chart is proposed.Firstly,the adaptive control chart is established,according to the quality characteristic distribution parameters of the manufacturing processes.Then,control chart pattern recognition is carried out by an existing pattern recognition method,such as BP neural network.Lastly,online adjustment strategy is proposed based on pattern recognition to improve the quality of manufacturing processes.Finally,we also discuss some challenging topics which deserve further research in the future based on the above research results.This thesis integrates robust parameter design,automatic process control,statistical process control and various modeling technology to improve product quality based on systematically combining the design and manufacture of manufacturing product.These research not only enrich and develop the research contents of quality engineering,but also have important application value for improving the product quality of manufacturing industry.