Use of Importance Sampling Constraints in System Optimization

See related content

by Yingwei Liu, Nanchang Aircraft Manufacturing Co, Nanchang, China,
Fred Moses, Nanchang Aircraft Manufacturing Co, Nanchang, China,

Document Type: Proceeding Paper

Part of: Probabilistic Mechanics and Structural and Geotechnical Reliability:

Abstract: An important application of system reliability analysis is to incorporate overall system risk as a constraint in a structural optimization. The paper describes the use of Monte Carlo Importance Sampling (MCIS) estimates of structural system risk in several optimization examples. It is seen that such MCIS estimates do not lead to stable derivatives needed in minimum weight search algorithms for finding an optimum design. Comparisons of MCIS estimates with conventional system reliability bounds show the latter to provide more stable reliability estimates needed for optimization.

Subject Headings: System reliability | Structural systems | Risk management | Failure analysis | Optimization models | Sensitivity analysis | Monte Carlo method | System analysis | Europe | Monaco | Monte Carlo

Services: Buy this book/Buy this article


Return to search