Research On Risk Assessment And Quality Control Method For The Assembly Process Of Numerical Control Machine Tool

The CNC machine tool is the "mother machine" of manufacturing industry,and its quality and reliability largely reflect the development level of a nation's industry and market competitiveness.the assembly process is the most costly stage in the whole product design and the quality of the assembly process has the key influence on product quality.the final performance of the product was directly determined by the quality of the assembly.Therefore,the quality control of the assembly is a necessary condition to ensure the quality of CNC machine tools.In order to explore a set of reasonable and effective assembly quality improvement theories and methods,this article conducts a comprehensive study on the risk assessment and quality control methods of CNC machine assembly process.The paper mainly includes the following aspects:(1)The background and significance of the research on risk assessment and the quality control method for the assembly process of numerical control machine tool were analyzed.Then the basic concepts about assembly quality was introduced.The research status of risk assessment and the quality control method for CNC machine tool and its deficiencies were discussed.the source and main contents of the paper were expounded.(2)Probabilistic language PFMEA risk assessment method for assembly process.In order to identify and prevent high-risk potential failure modes in assembly process in advance,a probabilistic language PFMEA risk assessment method for assembly process was proposed.Firstly,the probabilistic linguistic term sets(PLTS)was used to characterize the ambiguity and uncertainty of experts' risk assessment information;Secondly,the risk factor weights were determined by using comprehensive weighting method;Thirdly,taking experts' psychological behaviors into account,regret theory and Complex Proportional Assessment(COPRAS)were adopted to prioritize failure modes.the failure modes with higher risks in the assembly process were determined,and corresponding preventive measures were taken in process design and process control to improve the assembly quality of the CNC machine tool.(3)Key assembly quality control points judgment method in the CNC machine tool based on Partial Least Squares Regression(PLSR).There are many assembly control points in the assembly process of CNC machine tools.If all quality controls were performed without screening,it is difficult to grasp the key elements of improving the assembly quality of the CNC machine tool.A key assembly quality control points judgment method in the CNC machine tool based on Partial Least Squares Regression(PLSR)was proposed First,based on ‘‘Function-Motion-Action'' functional structural decomposition,the definitions of meta-action and meta-action chain are introduced,Second,the kinematics parameters of the terminal meta-action in meta-action chain were taken as the assembly quality analysis object,the power-transmitting model for metaaction chain was developed based on the state space model and the hierarchical iteration method.Third,the PLSR(Partial Least Squares Regression)was used to solve the multicollinearity problem and few sample data problem of the influential factors,and the variable importance projection index was utilized to measure the importance degree of each influencing factor,and corresponding preventive measures were taken.(4)Prediction model of assembly process quality based on support vector machine optimized by genetic algorithm(GA-SVM).In order to predict the assembly quality in advance,optimize the assembly process with quality problems,and improve the assembly qualification rate,a prediction model of assembly quality based on GA-SVM was proposed.Firstly,the basic principles of genetic algorithms and support vector machines was briefly introduced.Secondly,an assembly quality prediction framework and process were built,Finally,the Manual trial method are mainly used for the main parameter selection in traditional SVM,which is not only slow in efficiency but also difficult to obtain the optimal parameter combination,and easily causes poor local search or slow convergence,genetic algorithms was introduced to optimize its parameters,improving the learning ability and prediction accuracy of the SVM model.