Design and analysis of forward and reverse models for predicting defect accumulation, defect energetics, and irradiation conditions

Journal Article · 06 March 2018 · Computational Materials Science

DOI: https://doi.org/10.1016/j.commatsci.2018.02.048 · OSTI ID:1426798

Stewart, James A. ^[1]; Kohnert, Aaron A. ^[2]; Capolungo, Laurent  ^[2]; Dingreville, Rémi Philippe Michel ^[1]

---

1. Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
2. Los Alamos National Lab. (LANL), Los Alamos, NM (United States)

The complexity of radiation effects in a material’s microstructure makes developing predictive models a difficult task. In principle, a complete list of all possible reactions between defect species being considered can be used to elucidate damage evolution mechanisms and its associated impact on microstructure evolution. However, a central limitation is that many models use a limited and incomplete catalog of defect energetics and associated reactions. Even for a given model, estimating its input parameters remains a challenge, especially for complex material systems. Here, we present a computational analysis to identify the extent to which defect accumulation, energetics, and irradiation conditions can be determined via forward and reverse regression models constructed and trained from large data sets produced by cluster dynamics simulations. A global sensitivity analysis, via Sobol’ indices, concisely characterizes parameter sensitivity and demonstrates how this can be connected to variability in defect evolution. Based on this analysis and depending on the definition of what constitutes the input and output spaces, forward and reverse regression models are constructed and allow for the direct calculation of defect accumulation, defect energetics, and irradiation conditions. Here, this computational analysis, exercised on a simplified cluster dynamics model, demonstrates the ability to design predictive surrogate and reduced-order models, and provides guidelines for improving model predictions within the context of forward and reverse engineering of mathematical models for radiation effects in a materials’ microstructure.

• Research Organization: Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States)
• Sponsoring Organization: USDOE National Nuclear Security Administration (NNSA); USDOE Laboratory Directed Research and Development (LDRD) Program
• Grant/Contract Number: AC04-94AL85000; NA0003525
• OSTI ID: 1426798
• Report Number(s): SAND--2018-2012J; 660923
• Journal Information: Computational Materials Science, Journal Name: Computational Materials Science Journal Issue: C Vol. 148; ISSN 0927-0256
• Publisher: ElsevierCopyright Statement
• Country of Publication: United States
• Language: English