An efficient approach for evaluating the reliability of engineering structures using support vector machine with clustering algorithm

ABSTRACT

An implicit limit state function is predicted with a support vector machine (SVM) for reliability analysis of engineering structures to reduce the number of finite element analyses. The accuracy and predictability of the SVM method are reduced considerably by noise in data. In this paper, density-based spatial clustering of applications with noise (DBSCAN) is applied to reduce the noise in training samples for SVM regression. Then, the SVM model is linked with Monte Carlo simulation (MCS) to find out the reliability of the engineering structures. Four different examples of static and dynamic problems are solved to show acceptability and efficiency of the proposed method. It is observed that the proposed method is suitable for a smaller number of performance function calls. Direct MCS, artificial neural network-based MCS and response surface methods have been used to examine the effectiveness of the algorithm.

KEYWORDS:

• Failure probability of engineering structures
• clustering algorithm
• support vector machine
• response surface method
• noisy samples
• implicit limit state function

Disclosure statement

NThe current authors have declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.