The Research Of Process Quality Control Method Based On SOV Theory

The problems about variation transmission in multi-processes are investigated and studied systematically in this dissertation. The research in dissertation is focused on quality control and analysis methodology for the whole process.Based on discussions of state-of-art and approaching direction of researches in quality control technique, a key point is put forward that quality control methodology must be oriented to the process. At the same time, some shortcomings of the traditional quality control techniques and stream of variation (SOV) are summarized. The dissertation is focused on the process oriented quality control model in machining, quality control methods and fault analysis. Technological breaks and valuable research results have been obtained.Firstly, this dissertation has analyzed the factors impacting the feature size of product and introduced the structure and characteristics of the multi- processes manufacturing process which has been modeled by state space equation. The specific mathematic relations between product and process have been obtained by analysis the effect of variation sources, and the detail forms of each item in model can be achieved. So the 3D variation transmission model, the academic foundation of system optimization and variation diagnosis, can be established..The quality stability of multi- processes manufacturing process is investigated for the transformation from production to process in quality control. According to the variation transmission model, the conditions to judge if the process is stable and the index-"stability degree"to show the degree of stability are proposed under the concept of BIBO stability of system.The processing design in multi-processes manufacturing process has been researched on the basis of variation transmission model, including: the oriented process tolerance optimized allocation, the sensitivity analysis of process parameters. The research has integrated process parameters and quality characteristic of product closely and expanded the content of traditional processing design. For considering the tolerance of process parameters, the oriented process tolerance optimized allocation can not only assure the quality of product and establish the more rational service scheme of equipment. Similarly, the concept of sensitivity, as the quantificational evaluate index of physical structure of multi- processes manufacturing process, provide the theoretic rules to compare and improve the physical structure.In addition, the conception of diagnosability is proposed to quantitate the efficiency of measure strategy and the method to design the optimal measuring strategy having completely diagnosability can get in order to adapt to the development of enterprise information and quality control.For the problems of variation diagnosis and improvement, the fault process can be detected by the concept of virtual process. Then, the fault process parameters can be also founded according to the pattern recognition and match. Finally, the variation causes have been analyzed and improved using the rough set theory.The methodology in this dissertation has been validated. The results show that the proposed model has a good performance for quality control and analysis in multi-processes manufacturing process.