摘要
轿车车身是轿车的重要组成部分，是影响整车质量的重要因素之一。车身质量水平不仅反映了车体设计水平，而且也反映了制造水平和管理水平。轿车车身主要由冲压金属薄壁件组成，具有柔性大、刚度小等特点，其装配偏差的分析与综合对装配质量有很大影响。现有装配偏差的分析与综合研究主要沿用了刚体装配误差理论，不适合柔性金属薄壁件，因此，寻求适用于具有柔性特点的薄板装配偏差研究方法，探明引起装配偏差的各类偏差源，对控制轿车车身等薄板装配偏差具有重要意义。
本课题是在上述背景下开展的，目的是通过对薄板装配偏差的计算，寻求适合金属薄板装配的公差分析方法；通过对引起装配偏差的各类偏差源的分析，发现不同偏差源对综合偏差的不同影响，有利于提高金属薄板的装配质量。
本文以金属薄板为研究对象，利用有限元软件MSC/PATRAN对其进行建模、计算与分析，利用数论方法产生的伪随机点与有限元法相结合计算薄壁件的装配公差。这种方法的优点在于可以同时考虑零件装配过程中的弹塑性变形。通过与Monte-Carlo模拟法的比较可以发现，在计算精度基本相同的条件下，该法比Monte-Carlo模拟法计算量小，计算时间短，效率高。
同时，利用正交试验设计法对薄板的装配偏差进行正交分析。分析结果可以表明：在柔性金属薄板装配中，较厚的薄板零件对装配的综合偏差起主导作用。
Auto-body assembly is one of important parts of the whole car. Its quality is a key factor that affects the quality of the whole car. The quality of auto-body reflects not only the design level of automobile factory, but also its manufacturing level and management level. The auto-body assembly requires hundreds of sheet metals. For accounting the effect of low stiffness of sheet metals, their tolerances based on analysis and synthesis have an influence on the quality of the assembly. The existing assembly tolerance analysis and synthesis are mainly applied to the rigid assembly, and not suited to the deformable sheet metal assembly. It is necessary to find a new method that applied to the deformable sheet metal assembly. There is great meaning that we use the new method to make clear the variances sources and to improve the quality of the sheet metal assembly.
This study is started according to the above background. The purpose is to find a best variance analysis method for the deformable sheet metal assemblies. We can gain useful information to improve the quality of assembly by analyzing the variance sources.
In this paper, a sheet metal assembly is as to the study object. MSC/PATRAN is used to modeling, calculating, and analyzing. The assembly variation is calculated using finite element method and pseudo random numbers which are produced by number theoretic method. This method can account for the effect of elastic and plastic deformation. It can be concluded that this method needs simpler computation and shorter time than Monte-Carlo Simulation for the same precision. In a word, this method is more efficiency than the method of Monte-Carlo Simulation.
In this paper, we analyze the variations of sheet metal assembly according to the method of orthogonal design. It can be proved that the thicker sheet metals leading the variance of assembly in the deformable sheet metal assemblies.
引文
[1]国斌．汽车车身制造工艺．吉林：吉林大学出版社，1984．10-15
[2]胡仕新．美国汽车车体装配与焊接研究现状．中国机械工程，1997，8(1)：24-26
[3] Ceglarek,D, Shi,J, Zhou,Z. Variation Reduction for Assembly: Methodologies and Case Studies Anlasysis. Technical Report of the "2mm" Program. University of Michigan, Ann Arbor
[4] Greenwood,W.H, Chase,K.W. A New Tolerance Analysis Method for Designers And Manufactures. ASME Journal of Engineering for Industry, 1987, 109(5): 112-116
[5]胡敏．车身点焊装配偏差分析的建模方法研究：博士论文．上海：上海交通大学，2001
[6]武一民，周志革，杨津，等．公差分析与综合的进展．机械设计，2001，2(2)：4-5
[7] Nigma, S.D, Turner, J.U. Review of Statistical Approaaches to Tolerance A analysis. Computer-Aided Design, 1995, 27 (1): 6-15
[8] Greenwood,W.H, Chase,K.W. Worst Case Tolerance Analysis with Nonliner Problems. ASME Journal of Engineering for Industry, 1988, 110(88): 232-235
[9] Kusiak,A, Feng,C.X. Deterministic Tolerance Synthesis: A Comparative Study. Computer-Aided Design, 1995, 27(10): 795-768
[10] Medland,A.J, Mullineux,G, Rentoui,A.H. A Strategy for the Verification of Tolerance by Coordinate Measuring Machines. Proceedings of the Institute of Mechanical Engineers, Journal of Engineering Manufacture, 1996, 210(B4): 377-383
[11] Ngoi,BoK.A, Tan,K.C.K. Tolerance Stack Analysis for Assembly. Proceedings of Institute of Mechanical Engineers. Journal of Engineeering Manufacture, 1996, 210(B3): 279-289
[12] Tsai,PL, Fu,C.C, Cheing,E.H. Analysis and Synthesis of Geometric Models Using Tree- Structure Relations. Computer-Aided Design, 1997, 29(9): 607-615
[13] Greenwood,W.H, Chase,K.W. Root Sum Squares Tolerance Analysis with Nonliner Problems. ASME Journal of Engineering for Industry, 1990, 112(5): 382-384
[14] Lee,W.J, Woo,T.C. Tolerance: Their Analysis and aynthesis. ASME Journal of Engineering for Industry, 1990, 112:113-121
[15] Skowronski,V.J, Turner,J.U. Using Monte Carlo Variance Reduction in Statistical Tolerance synthesis. Computer-Aided Design, 1993, 29:63-69
[16] Nigma, S.D, Turner, J.U. Review of Statistical Approaches to Tolerance Analysis. Computer-Aided Design, 1995, 27(1): 6-15
[17]李良巧，顾唯明．机械可靠性设计与分析．北京：国防工业出版社，1998．30-35
[18] Takezawa.N. An improved method for establishing the Process Wise Quality Standard. Reports of Statistical and Applied Research. Union of Japanese Scientists and Engineers (JUSE), 1980, 27(3): 63-76
[19] Dariusz, J, Ceglarek. Multivariate Analysis and Evaluation of Adapative Sheet Metal Assembly Systems. Annals of the CIRP, 1998,47(1); 17-23
[20] Nigma, S.D, Turner, J.U. Review of Statistical Approaches to Tolerance Analysis. Computer- Aided Design, 1995, 27(1): 6-15
[21] Liu,S.C, Hu,S.J. An Offset Finite Element Model and Its Applications in Predicting Sheet Metal Assembly Variation. International Journal of Machine Tools Manufacturing, 1995, 35(11): 1545-1557
[22] Liu,S.C, Lee,H.W, Hu,S.J. Tolerance Analysis for Deformable Sheet Metal Assemblies. Journal of Mechanical Design Engineering, 1996(3), 118:62-67
[23]胡敏，来新民，林忠钦，等．多柔性件并联点焊装配误差的数学模型．上海交通大学学报，2001，35(1)：138-143
[24]沈利冰．轿车车身装配偏差分离与控制技术研究：博士论文．上海：上海交通大学，2001
[25] Hu, S.J. Stream-of-Variation Theory for Automotive Body Assembly. Annals of the CIRP,1997, 46(1):1-6
[26] Liu.S.C, Hu.S.J. Variation Simulation for Deformable Sheet Metal Assemblies Using Mechanistic Models. Trans. of NAMRI, 1995, ⅩⅩⅡ(5):235-240
[27] Liu,S.C, Hu,S.J. Variation Simulation for Deformable Sheet Metal Assemblies Using Finite Element Method. ASME Journal of Manufacturing Science and Engineering, 1997,119(8): 368-374
[28] Liu S.C, Hu.S.J. A Parametric Study of Joint Performance in Sheet Metal Assembly. Int. J. Machine Tools & Manufacturing, 1997, 37(6): 873-884
[29] Liu S.C, Hu.S.J. Sheet Metal Joint Configurations and Their Variation Characteristics. Transactions of ASME Journal of Manufacturing Science and Engineering, 1998, 120(2): 46 !-467
[30] 来新民，林忠钦，陈关龙．轿车车体装配尺寸偏差控制技术．中国机械工程，2000，11(11)：1215-1220
[31]林忠钦，胡敏．轿车车体装配偏差研究方法综述．机械设计与研究，1999，3：58-60
[32]王守信．有限元法教程．哈尔滨：哈尔滨工业大学出版社。1994．15-20
[33]张国瑞．有限元法．北京：机械工业出版社，1991．32-35
[34]王立辉．铝合金车轮的强度分析及优化设计：硕士论文．天津：河北工业大学，2002
[35]韩旭．灯泡贯流式水轮发电机组主轴的静动态有限元分析：硕士论文．天津：河北工业大学，2003
[36]郄艳辉．铝合金车轮的结构分析及优化：硕士论文．天津：河北工业大学，2003
[37]MSC公司．MSC／NASTRAN-世界公认的CAE工业标准．计算机辅助设计与制造，2002(5)：59-62
[38] MSC/NASTRAN Version 70 Quick Reference Guide. The MacNeaI-Schwendler Corperation, 1997
[39] Basic MSC/NASTRAN Linear Static and Normal Modes Analysis, NAS101 Exercise. The MacNeaI-Schwendler Coroeration, 1997
[40]MSC／PATRAN-基于并行工程的框架式前后处理及有限元集成分析系统．The MSC．Software Corperation
[41]阿木古愣，吴晓薇，李建福．Monte Carlo方法在计算物理中的应用．内蒙古农业大学学
报2001，22(3)：113-118
[42]童继平，韩正姝．蒙特卡罗方法与计算机模拟研究．计算机与农业，2000：7-23
[43]黎锁平．运用蒙特卡罗方法求解随机性问题．甘肃工业大学学报，2001，27(3)：95-97
[44]宫野．计算多重积分的蒙特卡罗方法与数论网格法．大连理工大学学报，2001，141(1)：20-24
[45] Liu Hongqian, Yuan Xigang, Fang Kaitai. Application of Evolution Sequential Number Theoretic Optimization in Global Optimization. Transactions of Tianjin University, 2002, 8(4): 221-226
[46]颜松远．数论及其应用．数学的实践与认识，2002，32(3)：486-508
[47]陈志云，程敬荣．同余关系在初等数学中的应用．2001，高等函数学报．14(6)：9-13
[48]唐翠娥，曹喜旺．线性同余方程组解的存在性及其算法．黄冈师范学院学报，2002，22(6)：17-19
[49] Zhige Zhou, Wenzhen Huang, Li Zhang. Sequential Algorithm Based on Number Theoretic Method for Statistical Tolerance Analysis and Synthesis. Journal of Manufacturing Science and Engineering. 2001, 123(3):490-493
[50]方开泰．王元，数论方法在统计中的应用．北京：科学出版社，1996，1-25
[51]周志革，黄文振，张利．数论方法在统计公差分析中的应用．机械工程学报，2000，36(3)：69-72
[52]周志革，黄文振，张利．一种计算随机变量函数均值和标准差的方法．机械强度，2001，23(1)：107-110
[53]周海波．正交试验设计在整机试验中的应用．拖拉机与农用运输车，1999，6：22-25
[54]陈兆能．试验分析与设计．上海：上海交通大学出版社，1991．250-270
[55]高允彦．正交及同归试验设计方法．冶金出版社，1988，50-63