Data-driven accurate design of variable blank holder force in sheet forming under interval uncertainty using sequential approximate multi-objective optimization

Abstract

In this paper, a sequential approximate multi-objective optimization method is employed to obtain the data-driven optimal variable blank holder forces in sheet forming under interval uncertainty. Uncertain parameters driven by data are modeled as intervals, bounds of which can be easily obtained from a small number of uncertainty information. Through a nonlinear interval number programming method, the multi-objective optimization problem under interval uncertainty is converted into a deterministic one. Considering the problem of low efficiency and local optima caused by traditional optimization methods, sequential approximate optimization approach is introduced to improve the efficiency and the ability to find global optimal solutions during variable blank holder forces (VBHF) uncertain multi-objective optimization process. The presented method is applied to accurate design the variable blank holder force for a certain space large size thin-walled part forming under uncertainty. The forming quality is compared with deterministic multi-objective optimization of VBHF, and its effectiveness is proved.