A non-probabilistic reliability-based design optimization method for structures based on interval models

摘要

A non-probabilistic reliability optimization method for structures with uncertain-but-bounded variables was presented. Based on the interval model description for the uncertain-but-bounded parameters, the non-probabilistic reliability index was used to measure the structural reliability. The optimal design was formulated as a nested two-loop optimization problem. For a nonlinear function, a number of constraints that satisfied the non-probabilistic reliability index were treated as the serial system, and the feasible region of the uncertain parameters was determined by the target non-probabilistic reliability index. The inner optimization method for solving a number of non-probabilistic reliability indices was transformed into the problem of minimizing the functions within the feasible range. For a linear function, the non-probabilistic reliability optimization problem can be transformed into a deterministic optimization problem. The high computational cost of determining the non-probabilistic reliability was greatly reduced. Three numerical examples were presented to illustrate the efficiency of the proposed method.