Multiple response surfaces method with advanced classification of samples for structural failure function fitting

Youbao Jiang, Linjie Zhao, Michael Beer, Edoardo Patelli, Matteo Broggi, Jun Luo, Yihua He, Jianren Zhang

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

The current response surface methods based on classifier usually fail to classify all samples correctly, thus neglect the effects of the misclassified samples on the fitting function. To overcome this issue, an improved multiple response surfaces method is proposed. It is mainly based on the techniques of sector division and correct classification of samples. The main steps are: (1) compute a normalized inner product coefficient between the closest sample to the origins and any other one, and sort samples by the coefficient values; (2) select a reasonable number of sorted samples (i.e. range of normalized inner product coefficient) for each sector to assure that the samples in the sector can be classified correctly; (3) divide the overall space into multiple sectors based on such ranges and execute an approximation sector by sector based on support vector machines. A main merit of this method is that it can approximate implicit failure functions well as the number of samples is large enough due to the features of the correct classification of all samples. In addition, it can be applied to both single failure functions and multiple failure functions (explicit ones and enveloped ones). Numerical examples show that the proposed method can achieve a good fitting of implicit failure functions, and the reliability results are accurate, too.
LanguageEnglish
Pages87 - 97
Number of pages11
JournalStructural Safety
Volume64
Early online date19 Oct 2016
DOIs
Publication statusPublished - 31 Jan 2017

Fingerprint

Multiple Responses
Response Surface Method
Sector
Scalar, inner or dot product
Support vector machines
Coefficient
Classifiers
Range of data
Sort
Divides
Division
Support Vector Machine
Classify
Classifier
Numerical Examples

Keywords

  • structural reliability
  • multiple response surfaces
  • support vector
  • correct classification
  • failure function
  • sector division

Cite this

Jiang, Youbao ; Zhao, Linjie ; Beer, Michael ; Patelli, Edoardo ; Broggi, Matteo ; Luo, Jun ; He, Yihua ; Zhang, Jianren. / Multiple response surfaces method with advanced classification of samples for structural failure function fitting. In: Structural Safety. 2017 ; Vol. 64. pp. 87 - 97.
@article{66b18acf04d54ce0b56914f36705147c,
title = "Multiple response surfaces method with advanced classification of samples for structural failure function fitting",
abstract = "The current response surface methods based on classifier usually fail to classify all samples correctly, thus neglect the effects of the misclassified samples on the fitting function. To overcome this issue, an improved multiple response surfaces method is proposed. It is mainly based on the techniques of sector division and correct classification of samples. The main steps are: (1) compute a normalized inner product coefficient between the closest sample to the origins and any other one, and sort samples by the coefficient values; (2) select a reasonable number of sorted samples (i.e. range of normalized inner product coefficient) for each sector to assure that the samples in the sector can be classified correctly; (3) divide the overall space into multiple sectors based on such ranges and execute an approximation sector by sector based on support vector machines. A main merit of this method is that it can approximate implicit failure functions well as the number of samples is large enough due to the features of the correct classification of all samples. In addition, it can be applied to both single failure functions and multiple failure functions (explicit ones and enveloped ones). Numerical examples show that the proposed method can achieve a good fitting of implicit failure functions, and the reliability results are accurate, too.",
keywords = "structural reliability, multiple response surfaces, support vector, correct classification, failure function, sector division",
author = "Youbao Jiang and Linjie Zhao and Michael Beer and Edoardo Patelli and Matteo Broggi and Jun Luo and Yihua He and Jianren Zhang",
year = "2017",
month = "1",
day = "31",
doi = "10.1016/j.strusafe.2016.10.002",
language = "English",
volume = "64",
pages = "87 -- 97",
journal = "Structural Safety",
issn = "0167-4730",
publisher = "Elsevier",

}

Multiple response surfaces method with advanced classification of samples for structural failure function fitting. / Jiang, Youbao; Zhao, Linjie; Beer, Michael; Patelli, Edoardo; Broggi, Matteo; Luo, Jun; He, Yihua; Zhang, Jianren.

In: Structural Safety, Vol. 64, 31.01.2017, p. 87 - 97.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Multiple response surfaces method with advanced classification of samples for structural failure function fitting

AU - Jiang, Youbao

AU - Zhao, Linjie

AU - Beer, Michael

AU - Patelli, Edoardo

AU - Broggi, Matteo

AU - Luo, Jun

AU - He, Yihua

AU - Zhang, Jianren

PY - 2017/1/31

Y1 - 2017/1/31

N2 - The current response surface methods based on classifier usually fail to classify all samples correctly, thus neglect the effects of the misclassified samples on the fitting function. To overcome this issue, an improved multiple response surfaces method is proposed. It is mainly based on the techniques of sector division and correct classification of samples. The main steps are: (1) compute a normalized inner product coefficient between the closest sample to the origins and any other one, and sort samples by the coefficient values; (2) select a reasonable number of sorted samples (i.e. range of normalized inner product coefficient) for each sector to assure that the samples in the sector can be classified correctly; (3) divide the overall space into multiple sectors based on such ranges and execute an approximation sector by sector based on support vector machines. A main merit of this method is that it can approximate implicit failure functions well as the number of samples is large enough due to the features of the correct classification of all samples. In addition, it can be applied to both single failure functions and multiple failure functions (explicit ones and enveloped ones). Numerical examples show that the proposed method can achieve a good fitting of implicit failure functions, and the reliability results are accurate, too.

AB - The current response surface methods based on classifier usually fail to classify all samples correctly, thus neglect the effects of the misclassified samples on the fitting function. To overcome this issue, an improved multiple response surfaces method is proposed. It is mainly based on the techniques of sector division and correct classification of samples. The main steps are: (1) compute a normalized inner product coefficient between the closest sample to the origins and any other one, and sort samples by the coefficient values; (2) select a reasonable number of sorted samples (i.e. range of normalized inner product coefficient) for each sector to assure that the samples in the sector can be classified correctly; (3) divide the overall space into multiple sectors based on such ranges and execute an approximation sector by sector based on support vector machines. A main merit of this method is that it can approximate implicit failure functions well as the number of samples is large enough due to the features of the correct classification of all samples. In addition, it can be applied to both single failure functions and multiple failure functions (explicit ones and enveloped ones). Numerical examples show that the proposed method can achieve a good fitting of implicit failure functions, and the reliability results are accurate, too.

KW - structural reliability

KW - multiple response surfaces

KW - support vector

KW - correct classification

KW - failure function

KW - sector division

U2 - 10.1016/j.strusafe.2016.10.002

DO - 10.1016/j.strusafe.2016.10.002

M3 - Article

VL - 64

SP - 87

EP - 97

JO - Structural Safety

T2 - Structural Safety

JF - Structural Safety

SN - 0167-4730

ER -