Variation Propagation Modeling And Design Evaluation Of Locating Datum System In Sheet Panel Multi-station Assembly Processes

Auto-body assembly is a representative sheet panel multi-station assembly processes(MAP) which is very complex and the quality of assembly dimensional quality influences vehicle performance and appearance directly. The design phase of a product development cycle has the largest impact on life-cycle cost and quality of the product, and the lack of product dimensional variation control at this phase is a primary cause of design changes. The locating datum system is an important content to uniform positioning throughout the design, manufacture, assembly, and inspection in dimensional management of sheet metal product development, which is a decisive factor of product robustness. The accurate controlling of product dimensional quality process-oriented during early design stage is an essential method to improve product quality, to shorten the development cycle, as well as to reduce costs. Establishing an accurate variation propagation mathematical model in multi-station assembly processes during early design stage is the most important premise to control the sheet panel product assembly variation.This paper studies the variation propagation in multi-station assembly process, and proposes a discrete time nonlinear state space model to describe the variation in the process of multi-station assembly on the basis of traditional state space model. The proposed model provides a more accurate mathematical representation between Key Control Characteristics(KCC) and Key Product Characteristics(KPC) in which deviation of the locating datum on product is considered,using design parameters for locating datum system and the motion vector of the component-oriented reference system as the parameters and state variable of the model respectively.Combining with the multi-station assembly processes of an case of inner panel of an A-class car body, using Variation System Analysis(VSA) software to analysis the virtual assembly simulation for multi-station assembly of sheet panels, and forecast products deviation of multi-station assembly using the established nonlinear state space model, the simulation software and the existing model respectively, the result of the comparison shows that the accuracy of the proposed mathematical model is adequate whose accuracy is much higher than the existing model.Based on nonlinear state space model and method of Sobol’ global sensitivity analysis method, propose a multi-tiered design evaluation methods of locating datum system for total deviation and KPC points deviation, and the evaluation method is applied to the engineering example analysis. Assembly system level design evaluation method can be directly used for locating datum system robust design optimization, and the station level evaluation method and fixture level design evaluation can be used to find the key stations or key fixture pins in multi-station assembly processes to simplify design parameters of needed to optimize in locating datum system and improve the efficiency of robust optimization.