본문 바로가기
  • 학술대회자료

표지

DBpia에서 서비스 중인 논문에 한하여 피인용 수가 반영됩니다. 내서재에 논문을 담은 이용자 수의 총합입니다.

초록·키워드 목차

Reliability analysis is of great importance in product and process design. For this purpose, uncertainty analysis is needed, there are two types of uncertainties classification - according to amount of data, aleatory and epistemic uncertainty; - according to the subject, input variable and metamodel uncertainty. Aleatory uncertainty is irreducible and related with inherent physical randomness that is completely described by a suitable probability model. Epistemic uncertainty, on the other hand, results from the lack of knowledge such as insufficient data, and can be reduced by collecting more information. Input variable uncertainty is due to the uncertainty of needed variable for design, such as dimension tolerance, material propertyㆍload uncertainty. And lastly, the model uncertainty is due to the simplifying assumptions of response function. For practical reliability based design optimization, integration of input variable and metamodel uncertainty is required. This paper addresses Bayesian framework for the reliability analysis which can take account of both the input variable and metamodel uncertainties. Markov Chain Monte Carlo (MCMC) method is employed as a means for the simulation of posterior distribution. A couple of mathematical and engineering examples are used to demonstrate the proposed method. #Reliability analysis(신뢰성 분석) #Input variable uncertainty(입력변수 불확실성) #Metamodel uncertainty(근사모델 불확실성) #Bayesian approach(베 이지안 접근법) #MCMC(마코프체인몬테카를로)

Abstract
1. 서론
2. 베이지안 접근법
3. 입력변수 및 근사모델 불확실성 통합
4. 설계 문제에의 적용
5. 결론
후기
참고문헌

저자의 논문

DBpia에서 서비스 중인 논문에 한하여 피인용 수가 반영됩니다.
Insert title here
논문의 정보가 복사되었습니다.
붙여넣기 하세요.