Surrogate model based uncertainty analysis and key process parameter determination for product reliability in assembling process

As an indispensable stage of product manufacturing, assembly process plays an important role in assuring product reliability by curbing the variation of assembly quality characters. And the characters, mainly affected by the uncertainty components quality and assembly process parameters, are formed by a complex process. This paper approaches the uncertainty analysis of the assembly quality characters. Firstly, by product assembly process data and the finite element method(FEM), the support vector regression (SVR) method is used to establish the surrogate model between the influencing factors and assembly quality characters. Secondly, on the basis of surrogate model, Monte Carlo Simulation(MCS) is used for the uncertainty analysis of assembly process, and then the sensitivity analysis is carried out to determine the key process parameter. Finally, a bolt assembly is used as case study to verify the effectiveness of the proposed method, which shows that the above method can express the propagation of the uncertainty in assembly process effectively, and the surrogate model can greatly increase the efficiency of uncertainty analysis with acceptable accuracy.