Surrogate Modeling of Computer Experiments with Different Mesh Densities

• Activity Number: 372
• Type: Invited
• Date/Time: Tuesday, August 5, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Technometrics
• Author(s): Dan Yu and Rui Tuo*+ and C. F. Jeff Wu
• Companies: Chinese Academy of Science and Chinese Academy of Science and Georgia Institute of Technology
• Keywords: Brownian motion ; finite element analysis ; kriging ; multi-fidelity data ; nonstationary Gaussian process models ; tuning parameters

Abstract

This talk considers deterministic computer experiments with real-valued tuning parameters which determine the accuracy of the numerical algorithm. A prominent example is finite element analysis with its mesh density as the tuning parameter. The aim of this work is to integrate computer outputs with different tuning parameters. Novel nonstationary Gaussian process models are proposed to establish a framework consistent with the results in numerical analysis. Numerical studies show the advantages of the proposed method over existing methods. The methodology is illustrated with a problem in casting simulation.