| Assembly dimensional variation of auto-body has a significant impact on the body overall appearance, wind noise, ride performance, and so on. There are many factors that are affecting the assembly quality of auto-body, such as structural design of parts, manufacturing variation and assembly process. As the production level has reached the limit of manufacturing, the dimension quality of auto-body is more dependent on the assembly process design in the body design. Therefore, it is important for us to conduct a study of body assembly variation analysis. As the characteristics of multi-station, multi-constraint features and multi-assembly sequence exist in the body assemble process, it is too complicated and difficult to analyze and model the analysis process of assembly variation in the body assembly process. The traditional method is inappropriate for body to conduct the assembly variation analysis. And the efficiency of Monte Carlo method, which is used in the traditional model, is inefficient. It is difficult to solve the process optimization of multi-part assembly. Therefore, in order to solve the above problem, the body variation analysis model of multi-station assembly and the process optimization of multi-part assembly are studied.Considering the part deviation, fixture deviation and relocator in the part assembly process, the assembly deviation analysis model, which is based on linear deviation equations, is established in this paper. Then, considering multi-stage and assembly sequence characteristics, assembly dimension chain is estalished to get the complete rigid deviation analysis model of multi-station assembly. Then two kinds of evaluation methods in the assembly deviation analysis model, the Monte Carlo method and RSS method, are studied. The two methods are solved through different analysis means for light assembly case and the results are compared. Finally, the optimization of assembly control features is operated for automobile body case based on the deviation analysis model and solving method. The mainwork is as follows:Variation analysis model of multi-station rigid is established according to linear deviation equation in this section. For the characteristics of assembly sequence, constraint and multistage of body assembly process, it is difficult to use traditional variation model to analyze assembly variation.This paper presents a new method to eatablish multistage vaviaton model based on rigid body. Based on the single-station assembly variation analysis model, this paper uses linear variation model to analyze single-station assembly variation and the parts variation, which is caused by relocation variation. The preliminary variation model is based on multi-stage rigid body.Considering the characteristics of assembly sequence and multistage in the assembly process, the dimension chain model is established to put all influential factors together and provide theoretical basis for subsequent RSS calculation method. First of all, the assembly relation model is established by using graph theory and the assembly directed graph is got. Then, based on the improved Floyd algorithm, dimension chain is automaticlly generated on the basis of assembly directed graph. Fanally the application of lateal assembly is presented to complete statement of multistage variation model of the rigid body.Considering the efficiency of Monte Carlo method is low, RSS calculation method, which is based on rigid analysis model of multistage variation, is proposed. First the autobody case is calculated by Monte Carlo method in the software 3DCS. Then, the same case is calculated by RSS method based on variation analysis model of rigid body by MABLAB programming. Finally, this part illustrates the feasibility and effectiveness of RSS method by comparing the results of the above calculation with each other.Based on the variation analysis model of rigid body and RSS calculation mothod, in order to improve the quality of the assembly, the assembly control features of the part is optimized. First of all, through analyzing of the assembly control feature in assembly process, this section analyzes the effect of specific control feature to the assembly variation. Then, based on ant colony algorithm, the lateral assembly case is used to optimize theassembly control features. The objective function of this case is RSS calculation result. |
PDF Full Text Request