Optimal Assembly Tolerance Design Based On Monte Carlo Method And Modified PSO Algorithm

As the rapid development of computer technology,computer aided technologies have been paid more and more attention in the area of manufacturing and processing industry .These technologies such as CAD/CAM have been significant breakthroughs and remarkable achievements in recent years, but the technology of computer aided tolerancing (CAT) ,which occupies an important place in the designing and manufacturing of products, lags far behind correspondingly because of the late start. Since the 90s of last century, domestic and foreign scholars have worked hardly in the current twenty years nearly, and obtained some advantages of tolerance theories which are used in practical engineering. Based on adequately investigation and research of prevenient CAT theories, this paper does some study for tolerance analysis and tolerance allocation which have been seemed as hard problems in CAT.Firstly, theories and methods in tolerance analysis and tolerance allocation are cleared and induced comprehensively including the merits and faults as well as application. Through the contrast, it is detected that Monte Carlo simulation and tolerance allocation method based on optimization technologies bear out the superiority for settling tolerance problems .Secondly, how Monte Carlo simulation has been applied in tolerance analysis is illuminated. Then a random sample is carried out for each component loop of the assembly chain , with that dimension and tolerance of the closed loop is calculated. In this paper Monte Carlo simulation is studied in three kinds of tolerance design function( linear/ nonlinear/ geometric tolerance added) by assembly chain applications in reduction gear box,connecting rod structure and piston structure of diesel engine and horizontal lathe. Compared with traditional worst case method and Root Sum Squared method ,it is indicated that the processing precision can be reduced advisably to get lower processing cost in the same assembly precision requirement .Thirdly,tolerance allocation based on minimum process cost method is studied and discussed contrasted with traditional equal precision method and equal tolerance method. This technique not only summarizes tolerance cost modeling according the comprehensive consideration between economical efficiency and precision of processing, but also establishes two kinds of tolerance optimization allocation model with constraint conditions based on worst case method and Root Sum Squared method; a modified PSO algorithm is proposed to solve this constrained Single objective optimization problem, and subsequently applied in the gear- shaft assembly chain of a reducer. Results indicates that modified PSO algorithm is superior paralleled with original PSO algorithm .And under the same assembly precision requirement, tolerance optimization allocation model based on Sum Squared method constraint condition can save more process cost of the whole assembly chain which satisfies the economic requirement.Lastly, this paper gives a summarize of the entire research work ,and indicates the further study direction .