Improvement of code behaviour in a design of experiments by metamodeling

TL;DR

This paper compares statistical metamodeling techniques like Kriging, kernel regression, and neural networks to enhance the reliability and efficiency of extensive numerical simulations in nuclear engineering.

Contribution

It demonstrates how different metamodels can improve code behavior analysis and failure detection in large-scale computational studies.

Findings

Kriging offers the most accurate predictions.

Neural networks provide the fastest metamodels.

All three methods detect computation failures effectively.

Abstract

It is now common practice in nuclear engineering to base extensive studies on numerical computer models. These studies require to run computer codes in potentially thousands of numerical configurations and without expert individual controls on the computational and physical aspects of each simulations.In this paper, we compare different statistical metamodeling techniques and show how metamodels can help to improve the global behaviour of codes in these extensive studies. We consider the metamodeling of the Germinal thermalmechanical code by Kriging, kernel regression and neural networks. Kriging provides the most accurate predictions while neural networks yield the fastest metamodel functions. All three metamodels can conveniently detect strong computation failures. It is however significantly more challenging to detect code instabilities, that is groups of computations that are all valid, but numerically inconsistent with one another. For code instability detection, we find that Kriging provides the most useful tools.