Efficient method for predicting fatigue of four-linked structural based submodel

Abstract

An efficient method is proposed in this paper to improve the minimum fatigue safety factor of the four-link system using shape optimization. In the meanwhile, the probabilistic response of the structure on the minimum fatigue safety factor is studied. This paper takes the four-link system as the object of study. The time-consuming simulation model were often employed in conventional method for such complex system that including a various of parameters and decision variables, and this would increase a mass of time cost. In the proposed method, submodel techniques are used to replace the original uneconomical simulation model and reduce the computational burden in complex system performance analysis and shape optimization. In order to certify the accuracy of the submodel, the result of shape optimization based such method was tested by replacing it back into the original HYPERMESH models and recalculate in OPTISTRUCT. The result of which indicated that it was attractive in finding out the optimum shape for improving minimum fatigue safety factor. Probabilistic analysis was employed for the purpose as well with vast experiments. In this paper, utilizing submodel technique can save approximately two-thirds of the time, and the result of which is practically the same with the original model.