Iterative method for tuning complex simulation code

DOI10.1080/03610918.2020.1728317MaRDI QIDQ5867441zbMATH OpenOpenAlexFDO

Authors Yun Seo Am, Youngsaeng Lee, J.-S. Park

Publication date 14 September 2022

Published in Communications in Statistics. Simulation and Computation (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2007.09865

zbMATH Keywords

best linear unbiased predictioncomputer experimentskrigingcalibrationnumerical optimizationinexact computer model

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Bayesian inference (62F15) Design of statistical experiments (62K99) Applications of statistics to physics (62P35)

Abstract: Tuning a complex simulation code refers to the process of improving the agreement of a code calculation with respect to a set of experimental data by adjusting parameters implemented in the code. This process belongs to the class of inverse problems or model calibration. For this problem, the approximated nonlinear least squares (ANLS) method based on a Gaussian process (GP) metamodel has been employed by some researchers. A potential drawback of the ANLS method is that the metamodel is built only once and not updated thereafter. To address this difficulty, we propose an iterative algorithm in this study. In the proposed algorithm, the parameters of the simulation code and GP metamodel are alternatively re-estimated and updated by maximum likelihood estimation and the ANLS method. This algorithm uses both computer and experimental data repeatedly until convergence. A study using toy-models including inexact computer code with bias terms reveals that the proposed algorithm performs better than the ANLS method and the conditional-likelihood-based approach. Finally, an application to a nuclear fusion simulation code is illustrated.

Recommendations

• Inverse solution for parameter estimation of computer simulation by an empirical Bayesian code tuning method
• A statistical method for tuning a computer code to a data base.
• Sequential tuning of complex computer models
• Surrogate-assisted tuning for computer experiments with qualitative and quantitative parameters
• Tuned iterated filtering

Cites work

• scientific article; zbMATH DE number 3874435 (Why is no real title available?)
• A Bayesian calibration approach to the thermal problem
• A frequentist approach to computer model calibration
• A statistical method for tuning a computer code to a data base.
• Adaptive numerical designs for the calibration of computer codes
• An efficient methodology for modeling complex computer codes with Gaussian processes
• Analysis methods for computer experiments: how to assess and what counts?
• Bayesian calibration of computer models. (With discussion)
• Bayesian calibration of the constants of the \(k-\varepsilon\) turbulence model for a CFD model of street canyon flow
• Bayesian emulation and calibration of a stochastic computer model of mitochondrial DNA deletions in substantia nigra neurons
• Combining Field Data and Computer Simulations for Calibration and Prediction
• Modularization in Bayesian analysis, with emphasis on analysis of computer models
• Prediction based on the Kennedy-O'Hagan calibration model: asymptotic consistency and other properties
• Sequential Design with Mutual Information for Computer Experiments (MICE): Emulation of a Tsunami Model
• Sequential tuning of complex computer models
• The design and analysis of computer experiments
• The design and analysis of computer experiments.

Cited in

(2)

• A statistical method for tuning a computer code to a data base.
• Inverse solution for parameter estimation of computer simulation by an empirical Bayesian code tuning method

Describes a project that uses

Uses Software

• laGP

This page was built for publication: Iterative method for tuning complex simulation code

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5867441)