Active learning with single-loop expected improvement for multimodal reliability-based design optimization

doi:10.1016/j.engstruct.2026.123226

Active learning with single-loop expected improvement for multimodal reliability-based design optimization

2026-06-17

0

PRE

AI

DOI：10.1016/j.engstruct.2026.123226

Original

Original request for help

Save

Abstract

En 中文

• A SMAL-RBDO algorithm is developed for efficient multimodal reliability optimization. • A new SLEI criterion integrates EI and reliable feasibility for guided global search. • The method enhances reliability-based exploration and avoids local-optimum trapping. • Benchmarks and an engineering case verify improved accuracy and reduced cost.

Active learning with single-loop expected improvement for multimodal reliability-based design optimization

Abstract

Journal

Organization