A new technique for performing probabilistic analysis and optimization design using data classification methods is investigated. The approach is based on nonlinear decision boundaries constructed from data classification methods. A statistical learning tool known as support vector machine (SVM) is used to construct the boundaries. An adaptive sampling technique is used to generate samples and update the approximated decision function. The proposed approach is demonstrated with several benchmark and engineering problems.
Efficient Probabilistic Analysis and Design Optimization Using Data Classification Decision Boundaries
- Views Icon Views
- Share Icon Share
- Search Site
Wang, L, Subramaniyan, AK, & Beeson, D. "Efficient Probabilistic Analysis and Design Optimization Using Data Classification Decision Boundaries." Proceedings of the ASME 2010 International Mechanical Engineering Congress and Exposition. Volume 11: New Developments in Simulation Methods and Software for Engineering Applications; Safety Engineering, Risk Analysis and Reliability Methods; Transportation Systems. Vancouver, British Columbia, Canada. November 12–18, 2010. pp. 161-171. ASME. https://doi.org/10.1115/IMECE2010-39921
Download citation file: