An active learning Kriging‐based multipoint sampling strategy for structural reliability analysis

Abstract

Abstract In order to effectively and accurately assess the failure probability of mechanical structures, this paper proposes a multi‐point sampling active learning reliability analysis method called AKMP. First, a GA‐Halton sequence is introduced to make the initial samples well dispersed and homogeneous in the design space. Second, a new learning function FELF is constructed to efficiently update the Kriging model, which takes into account the relationship between the location of the sampling points and the performance fun. Then, a combination of NCC criterion and multipoint sampling strategy is proposed to further improve the convergence efficiency, which can effectively terminate the active learning process. Finally, numerical and engineering cases are tested to verify the application performance of the proposed AKMP. The results show that the method has superior performance in terms of both accuracy and failure probability efficiency, and can reduce the computational resources of the active learning process.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 7 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.