Robust fixture design of compliant assembly process based on a support vector regression model
- 9 Downloads
Fixtures are used to locate and clamp parts and subassemblies during the compliant assembly process of automotive body parts. As a result, the assembly quality of compliant sheet metal parts is mainly determined by fixture locating layouts and variation sources, such as manufacturing variation of parts and fixture locating variation. This paper proposes a robust fixture design method of the compliant assembly process based on a support vector regression model. Firstly, an assembly variation modeling method of compliant sheet metal parts is discussed during the assembly subprocesses of fixture locating, welding gun clamping, weld finishing, and welding gun releasing. The combined assembly variation model describes a numeric relationship between the assembly variation of key measurement points and all variation sources under a certain fixture locating layout. Further, to solve the difficulty of directly obtaining the parameter model between the assembly variation and fixture design layouts under a certain variation source level, a response surface modeling method based on a support vector regression model is presented. And a whole assembly robust objective function is presented to realize the combining consideration of the whole assembly quality control with different assembly quality controls of different areas. Finally, a taillight bracket assembly is used to demonstrate the robust fixture design method. Results show that the proposed method is feasible and the assembly quality under the robust fixture locating layout has been greatly increased.
KeywordsFixture design Robust design Assembly variation modeling Response surface model
Unable to display preview. Download preview PDF.
The work was supported by the Shandong Provincial Natural Science Foundation, China (Grant No. ZR2016EEM31).
- 7.Fan L, Senthil Kumar A (2010) Development of robust fixture locating layout for machining workpieces. P I Mech Eng B-J Eng 224(12):1792–1803Google Scholar
- 12.Camelio JA (2002) Modeling and diagnosis of dimensional variation for assembly systems with compliant parts. University of Michigan, Ann ArborGoogle Scholar
- 15.Vapnik V, Golowich S, Smola A (1997) Support vector method for function approximation, regression estimation, and signal processing. Adv Neural Inf Proces Syst 9:281–287Google Scholar