Data-informed Emulators for Multi-Physics Simulations

TL;DR

This paper develops machine learning-based emulators for complex multi-physics systems near nuclear waste repositories, focusing on thermal, hydrological, mechanical, and chemical processes, to improve simulation efficiency and understanding.

Contribution

It compares random forests and neural networks for building emulators, demonstrating ensemble learning advantages for nonlinear, data-limited problems in multi-physics modeling.

Findings

Emulators accurately capture the temporal evolution of key parameters.

Random forests outperform neural networks in highly nonlinear scenarios.

Clustering and predictor assimilation improve emulator accuracy.

Abstract

Machine learning techniques are powerful tools for construction of emulators for complex systems. We explore different machine learning methods and conceptual methodologies, ranging from functional approximations to dynamical approximations, to build such emulators for coupled thermal, hydrological, mechanical and chemical processes that occur near an engineered barrier system in the nuclear waste repository. Two nonlinear approximators, random forests and neural networks, are deployed to capture the complexity of the physics-based model and to identify its most significant hydrological and geochemical parameters. Our emulators capture the temporal evolution of the Uranium distribution coefficient of the clay buffer, and identify its functional dependence on these key parameters. The emulators' accuracy is further enhanced by assimilating relevant simulated predictors and clustering strategy. The relative performance of random forests and neural networks shows the advantage of ensemble learning in random forests algorithm, especially for highly nonlinear problems with limited data.