Open AccessImproved response surface method based on triple weighted regression
Author(s) -
Jie Wu,
Jianguo Zhang,
Lingfei You,
Nan Yang
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1043/4/042018
Subject(s) - weighting , response surface methodology , mathematics , reliability (semiconductor) , surface (topology) , regression , statistics , function (biology) , regression analysis , probability density function , algorithm , weight function , limit (mathematics) , sampling (signal processing) , computer science , mathematical analysis , medicine , power (physics) , physics , geometry , filter (signal processing) , quantum mechanics , evolutionary biology , biology , computer vision , radiology
The response surface method (RSM) is frequently used to reduce the computational burden of structural reliability analysis. In this paper, an improved RSM is proposed. The response surface is fitted by the triple weighted regression technique, which integrates weighting systems to assign weight factors to each sampling point including (1) the absolute value of the limit state function (LSF), (2) the value of the joint probability density function and (3) the distance from the design point. Numerical examples are presented to demonstrate that the proposed method gives more accurate results than RSM for both probability of failure (PoF) and LSF evaluations.