A novel reliability evaluation method for large engineering systems

Reda Farag, Achintya Haldar

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

A novel reliability evaluation method for large nonlinear engineering systems excited by dynamic loading applied in time domain is presented. For this class of problems, the performance functions are expected to be function of time and implicit in nature. Available first- or second-order reliability method (FORM/SORM) will be challenging to estimate reliability of such systems. Because of its inefficiency, the classical Monte Carlo simulation (MCS) method also cannot be used for large nonlinear dynamic systems. In the proposed approach, only tens instead of hundreds or thousands of deterministic evaluations at intelligently selected points are used to extract the reliability information. A hybrid approach, consisting of the stochastic finite element method (SFEM) developed by the author and his research team using FORM, response surface method (RSM), an interpolation scheme, and advanced factorial schemes, is proposed. The method is clarified with the help of several numerical examples.

Original languageEnglish (US)
JournalAin Shams Engineering Journal
DOIs
StateAccepted/In press - Dec 7 2014

Fingerprint

Systems engineering
Interpolation
Dynamical systems
Finite element method

Keywords

  • FORM/SORM
  • Large engineering systems
  • Reliability
  • Response surface method

ASJC Scopus subject areas

  • Engineering(all)

Cite this

A novel reliability evaluation method for large engineering systems. / Farag, Reda; Haldar, Achintya.

In: Ain Shams Engineering Journal, 07.12.2014.

Research output: Contribution to journalArticle

@article{359d4df332b94e77b389e63f00306ac3,
title = "A novel reliability evaluation method for large engineering systems",
abstract = "A novel reliability evaluation method for large nonlinear engineering systems excited by dynamic loading applied in time domain is presented. For this class of problems, the performance functions are expected to be function of time and implicit in nature. Available first- or second-order reliability method (FORM/SORM) will be challenging to estimate reliability of such systems. Because of its inefficiency, the classical Monte Carlo simulation (MCS) method also cannot be used for large nonlinear dynamic systems. In the proposed approach, only tens instead of hundreds or thousands of deterministic evaluations at intelligently selected points are used to extract the reliability information. A hybrid approach, consisting of the stochastic finite element method (SFEM) developed by the author and his research team using FORM, response surface method (RSM), an interpolation scheme, and advanced factorial schemes, is proposed. The method is clarified with the help of several numerical examples.",
keywords = "FORM/SORM, Large engineering systems, Reliability, Response surface method",
author = "Reda Farag and Achintya Haldar",
year = "2014",
month = "12",
day = "7",
doi = "10.1016/j.asej.2016.01.007",
language = "English (US)",
journal = "Ain Shams Engineering Journal",
issn = "2090-4479",
publisher = "Ain Shams University",

}

TY - JOUR

T1 - A novel reliability evaluation method for large engineering systems

AU - Farag, Reda

AU - Haldar, Achintya

PY - 2014/12/7

Y1 - 2014/12/7

N2 - A novel reliability evaluation method for large nonlinear engineering systems excited by dynamic loading applied in time domain is presented. For this class of problems, the performance functions are expected to be function of time and implicit in nature. Available first- or second-order reliability method (FORM/SORM) will be challenging to estimate reliability of such systems. Because of its inefficiency, the classical Monte Carlo simulation (MCS) method also cannot be used for large nonlinear dynamic systems. In the proposed approach, only tens instead of hundreds or thousands of deterministic evaluations at intelligently selected points are used to extract the reliability information. A hybrid approach, consisting of the stochastic finite element method (SFEM) developed by the author and his research team using FORM, response surface method (RSM), an interpolation scheme, and advanced factorial schemes, is proposed. The method is clarified with the help of several numerical examples.

AB - A novel reliability evaluation method for large nonlinear engineering systems excited by dynamic loading applied in time domain is presented. For this class of problems, the performance functions are expected to be function of time and implicit in nature. Available first- or second-order reliability method (FORM/SORM) will be challenging to estimate reliability of such systems. Because of its inefficiency, the classical Monte Carlo simulation (MCS) method also cannot be used for large nonlinear dynamic systems. In the proposed approach, only tens instead of hundreds or thousands of deterministic evaluations at intelligently selected points are used to extract the reliability information. A hybrid approach, consisting of the stochastic finite element method (SFEM) developed by the author and his research team using FORM, response surface method (RSM), an interpolation scheme, and advanced factorial schemes, is proposed. The method is clarified with the help of several numerical examples.

KW - FORM/SORM

KW - Large engineering systems

KW - Reliability

KW - Response surface method

UR - http://www.scopus.com/inward/record.url?scp=84971454100&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84971454100&partnerID=8YFLogxK

U2 - 10.1016/j.asej.2016.01.007

DO - 10.1016/j.asej.2016.01.007

M3 - Article

AN - SCOPUS:84971454100

JO - Ain Shams Engineering Journal

JF - Ain Shams Engineering Journal

SN - 2090-4479

ER -