Response Surface Methodology: Process and Product in Optimization Using Designed Experiments
Response Surface Methodology: Process and Product in Optimization Using Designed Experiments
Design and Modeling for Computer Experiments (Computer Science & Data Analysis)
Design and Modeling for Computer Experiments (Computer Science & Data Analysis)
Grounding studies in a medium voltage DC shipboard power system with uncertain parameters
Proceedings of the 2010 Conference on Grand Challenges in Modeling & Simulation
Hi-index | 0.00 |
Practical issues in conducting parametric studies for computationally expensive simulations using response surface approaches are discussed. For cases in which prediction variance of response surface models is significant, an approach utilizing Gaussian process models is considered. Cross-validation techniques for the purposes of validating predictive distributions are discussed, and Sobol' sensitivity indices are examined as a metric for quantifying the contribution of response surface model uncertainty in parametric studies. The approach and associated issues are illustrated through an application to an uncertainty analysis involving an electromagnetic transient simulation of a notional all-electric warship for a pulse-load charging scenario.