KNN Classification Based MCS Method of Structural Reliability Analysis

A KNN Classification Based MCS (Monte Carlo Simulation Method) is proposed for the reliability analysis which hindered by the implicit nature of the performance function. In the method, Markov chain is adopted to simulate a small amount of training samples, KNN classification is used to generate sur...

Full description

Saved in:
Bibliographic Details
Published inApplied Mechanics and Materials Vol. 638-640; pp. 136 - 139
Main Authors Su, Guo Shao, Zhao, Ying, Yan, Liu Bin
Format Journal Article
LanguageEnglish
Published Zurich Trans Tech Publications Ltd 01.01.2014
Subjects
Response Surface
KNN
Monte Carlo
Structural Reliability
Markov Chain
Online AccessGet full text

Cover

More Information
Summary:A KNN Classification Based MCS (Monte Carlo Simulation Method) is proposed for the reliability analysis which hindered by the implicit nature of the performance function. In the method, Markov chain is adopted to simulate a small amount of training samples, KNN classification is used to generate surrogate model of performance function, MCS is used to estimate the failure probability. An iterative algorithm is presented to improve surrogate precision dynamically in the region contributing to the failure probability significantly. The study results demonstrate that the proposed method has superior performance to the traditional response surface method.
Bibliography:Selected, peer reviewed papers from the 2014 3rd International Conference on Civil, Architectural and Hydraulic Engineering (ICCAHE 2014), July 30 -31, 2014, Hangzhou, ChinaSelected, peer reviewed papers from the 2014 3rd International Conference on Civil, Architectural and Hydraulic Engineering (ICCAHE 2014), July 30 -31, 2014, Hangzhou, China
ISBN:3038352586
9783038352587
ISSN:1660-9336
1662-7482
1662-7482
DOI:10.4028/www.scientific.net/AMM.638-640.136