A Novel Support-Vector-Machine-Based Grasshopper Optimization Algorithm for Structural Reliability Analysis

Aiming at the characteristics of high computational cost, implicit expression and nonlinearity performance functions corresponding to large complex structures, this paper proposes a support-vector-machine- (SVM) based grasshopper optimization algorithm (GOA) for structural reliability analysis. With method, problem is transformed into an problem. On basis using finite element method (FEM) generate small number samples, SVM model used construct surrogate function, explicit nonlinear function under condition samples realized. Then, GOA search most probable point (MPP), reasonable iterative constructed. The MPP information each iteration step dynamically improve reconstruction accuracy in region that contributes failure probability. Finally, with after as sampling center, importance (ISM) further infer feasibility verified by four numerical cases. applied long-span bridge. results show has significant advantages efficiency suitable solving problems engineering.