Global Sensitivity Analysis Method Based on Support Vector Machine

Aiming at the difficulties of implicit limit state function and high-computational cost in structural sensitivity analysis, a new method is proposed for structural global sensitivity analysis which is based on support vector machine(SVM) method and Monte Carlo simulation(MCS). In this study, SVM is used to construct the mapping relationship between the input random variables and the output response. In order to obtain a certain number of failure samples, the uniform mapping is applied, which benefits to obtain an accurate SVM surrogate model. The proposed method only requires a small number of samples to explicitly express the implicit limit state function, and its efficiency is much higher than that of Monte Carlo simulation. Numerical and engineering examples show that the proposed method has the potential value of engineering application.

Abstract licence: CC BY

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.