Structural reliability analysis based on improved wolf pack algorithm AK-SS

To improve the accuracy and efficiency of solving structural reliability problems with highly nonlinear black box functions, this paper proposes an active learning Kriging approach for structural reliability analysis based on an improved learning strategy of the Improved Wolf Pack Algorithm (IWPA). Firstly, we optimize the kriging surrogate model by IWPA. Secondly, a parallel plus point strategy named IWPA-NLF which is based on IWPA and a new learning function (NLF) is used to update the kriging surrogate model. Finally, the subset simulation (SS) method is used to calculate the reliability. A numerical example and three engineering examples are used to verify the proposed method in this paper. The results show that the proposed method in this paper has a strong function fitting ability and can calculate the accurate failure probability with fewer performance function calls.