The non-probabilistic reliability-based design optimization based on imperialistic competitive algorithm and interval model

Probabilistic and fuzzy reliability-based design optimization (RBDO) all need a large amount of experimental data as the basis. The non-probabilistic RBDO, which need less date information, is a good supplement for the above two methods. In view of the phenomenon that the conventional optimization method may lead to the “interval expansion”, and then make the non-probabilistic reliability inaccurate, in this study a new method was proposed to solve this problem. In this method, a one-dimensional optimization algorithm is introduced into the non-probabilistic RBDO for the first time, and then based on the imperialistic competitive algorithm (ICA) and the interval model a non-probabilistic RBDO model and its sub-model are established. This model adopts a double nested optimization structure, in which the non-probability reliability index is calculated by the iterative method in the inner layer, and in the outer layer the global optimal scheme meeting the reliability requirements is searched using the ICA. In the end, examples demonstrated that the proposed method is fast, accurate and effective.