Research On Product Tolerance Analysis And Optimal Allocation In Virtual Assembly

Parts products tolerance attributes is a direct reflection of product processing and assembly precision,not only affects the level of product performance,but also largely determines the manufacturing processes and costs.Therefor,in the design stage,a reasonable tolerance design have great significance in improve product assemblability,ensure product quality level,reduce product manufacturing costs.Computer aided tolerance design(CAT)as one of the key technologies to achieve the integration of computer aided design(CAD)and computer aided manufacturing(CAM),but the development of CAT is relative lagging,the research work on CAT is still in urgent need to solve for domestic and foreign scholars and enterprises.This paper based on the study of CAT domestic and foreign papers,mainly studies and analyzes the problem of tolerance analysis and tolerance optimum allocation in the field of computer aided tolerance design.The main contents are as follows:Firstly,summarize and elaborate the theory and method of computer aided tolerance design,focus on analysis and comparison tolerance analysis common methods and tolerance allocation theory which based on optimization technology.Then,analyzed the basic idea and characteristics of Monte Carlo method,aiming at the disadvantage of slow convergence rate of traditional Monte Carlo method,a quasi-Monte Carlo method is proposed by construct quasi random number sequences instead of pseudorandom number sequences,gives the specific process based on quasi-Monte Carlo method to calculate the closedloop size and tolerance,use Matlab do the tolerance analysis of the gear subgroup assembling chain,make the results compared with the traditional Monte Carlo method,probabilistic method and extremum method,to verify the applicability of the Monte Carlo method and quasi-Monte Carlo method on tolerance analysis problems,and the advantages of the quasi-Monte Carlo method in the computational efficiency to the traditional Monte Carlo method.Finally,on the basis of the quality loss cost,introduce quality assurance costs to form the comprehensive quality cost,bulid tolerance multi-objective optimization allocation model,which takes minimal processing cost and comprehensive quality cost as the optimization goal,with constraints of the assembly function requirement and economic processing capacity;two improvements of the basic gravitational search algorithm(GSA)are carried out.The realization process of the improved gravity search algorithm(IGSA)is designed.The optimization calculation of the axial dimension chain tolerance distribution of the gear assembly is realized.Through calculation,And the comparison of the results shows that the improved model and the IGSA algorithm are superior in solving the problem of optimal allocation of tolerances.