Degradation modeling using Bayesian hierarchical piecewise linear models: A case study to predict void swelling in irradiated materials

Abstract

In this case study, we illustrate the use of a data-driven degradation model in a nuclear-specific application called void swelling. Void swelling is a complex, radiation-induced degradation mechanism that changes the dimensions of materials and damages the structural integrity. Accurate modeling and prediction of void swelling processes is crucial in nuclear power plant (NPP) management and maintenance planning by providing a guideline on the future state of the materials subject to reactor irradiation. Using a Bayesian hierarchical piecewise linear regression framework with a real-world void swelling dataset, we address the following three research questions: (1) How can we construct a data-driven degradation model such that its predictions satisfy the physical properties of void swelling? (2) How can we measure the joint effect of multiple experimental factors on the swelling process? (3) How can we accurately predict the future swelling status under limited data availability? The results on a real-world void swelling dataset not only improve our understanding of the swelling process but also provide a useful reference for nuclear practitioners and degradation researchers.

Keywords:

• Bayesian hierarchical piecewise regression
• degradation modeling
• multilevel models
• void swelling

Data availability statement

The data that support the findings of this study are openly available in Mendeley Data at http://doi.org/10.17632/g4bnkxgc26.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The authors gratefully acknowledge the support provided by the Department of Energy under award number DE-NE0008993.

Notes on contributors

Ye Kwon Huh

Ye Kwon Huh received the B.A. degree in business and statistics from Korea University, Seoul, South Korea, in 2020. He is currently pursuing a Ph.D. degree with the Department of Industrial and Systems Engineering, University of Wisconsin–Madison, Madison, WI, USA. His research interests are degradation modeling, explainable machine/deep learning, and predictive analytics with applications in manufacturing and energy systems. Mr. Huh is a member of INFORMS, SME, and IISE.

Minhee Kim

Minhee Kim received the B.S. degree in industrial and management engineering from Pohang University of Science and Technology (POSTECH), Pohang, Korea, in 2017, the M.S. degree in statistics and the Ph.D. degree in industrial and products engineering from the University of Wisconsin–Madison, Madison, WI, USA, in 2021 and 2022, respectively. Currently, she is an Assistant Professor in the Department of Industrial and Systems Engineering, the University of Florida, Gainesville, FL, USA. Her research interests are decision-oriented process modeling, monitoring, and prognostics with applications in manufacturing and energy products. Dr. Kim is a member of IEEE, INFORMS, SME and IISE.

Katie Olivas

Katie Olivas received her Bachelor of Science in Nuclear Engineering and Radiological Sciences from the University of Michigan - Ann Arbor in 2023. She worked as an undergraduate research assistant for Professor Todd Allen and focused on collecting data for the void swelling materials database.

Todd Allen

Todd Allen received his Bachelor's degree in Nuclear Engineering from Northwestern University in 1984 and his Doctoral degree in Nuclear Engineering from the University of Michigan in 1997. He is currently a Professor at the University of Michigan and a Senior Fellow at Third Way, a DC-based think tank, supporting their Climate & Energy Portfolio. He is the Founding Director of the Fastest Path to Zero Initiative, focusing on decarbonized systems through combined technologies and social acceptance, and the Co-Director of the MI Hydrogen initiative at the University of Michigan. Previously, he served as a Deputy Director for Science and Technology at the Idaho National Laboratory, a Professor at the University of Wisconsin, and the Scientific Director of the Advanced Test Reactor National Scientific User Facility at INL.

Kaibo Liu

Kaibo Liu received the B.S. degree in industrial engineering and engineering management from the Hong Kong University of Science and Technology in 2009 and the M.S. degree in statistics and the Ph.D. degree in industrial engineering from the Georgia Institute of Technology in 2011 and 2013, respectively. He is currently a Professor in the Department of Industrial and Systems Engineering, University of Wisconsin-Madison, Madison, WI, USA. His research interests are data fusion for process modeling, monitoring, diagnosis, prognostics, and decision making. Dr. Liu is a member of INFORMS, IEEE, IISE, SME, and ASQ.