An adaptive hybrid single-loop method for reliability-based design optimization using iterative control strategy

摘要: Single-loop approach (SLA) is one of the most promising methods for solving linear and weakly nonlinear reliability-based design optimization (RBDO) problems. However, since SLA locates current approximate probable point (MPP) by using gradient information previous to reduce computational cost, it may lead inaccuracy when nonlinearity probabilistic constraints becomes relatively high. To overcome this limitation, a new adaptive hybrid single-loop method (AH-SLM) that can automatically choose search MPP or accurate proposed in paper. Moreover, get more efficiently alleviate oscillation process, an iterative control strategy (ICS) with two criteria developed. In each step, KKT-condition performance measure (PMA) introduced check validity MPP. If infeasible, ICS will be further carried out The are used update step length, then converge fast both highly functions. Besides, numerical examples presented verify efficiency robustness our method.