Artificial Intelligence in Reactor Physics: Current Status and Future Prospects

2025-03-04

AI Summary

This paper presents a systematic review of machine learning (ML) applications in reactor physics, addressing core challenges including steady-state/transient analysis, burnup calculations, safety design, and real-time monitoring. Current research suffers from fragmentation, poor model generalizability, and insufficient theoretical foundations. To overcome these bottlenecks, we propose the first industry-adaptive ML integration framework for reactor physics, identifying surrogate modeling and digital twin implementation as critical enablers. The framework synergistically combines supervised learning, neural networks, Bayesian calibration, and uncertainty quantification to either accelerate deterministic solvers or correct biases in stochastic methods. Validated across over one hundred industrial case studies, the approach demonstrates substantial improvements: average computational speedups of 10–100× and enhanced prediction accuracy. Furthermore, the proposed transferable model architecture enables scalable deployment of high-fidelity simulation and intelligent operations and maintenance in nuclear engineering practice.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Stochastic Optimization
• Cognitive Modeling & Cognitive Systems: Applications

Application Category

• User Modeling, Personalization and Recommendation: ML for personalized search and recommendations
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Security and Privacy: Security and privacy of machine learning and AI applications

Abstract

Reactor physics is the study of neutron properties, focusing on using models to examine the interactions between neutrons and materials in nuclear reactors. Artificial intelligence (AI) has made significant contributions to reactor physics, e.g., in operational simulations, safety design, real-time monitoring, core management and maintenance. This paper presents a comprehensive review of AI approaches in reactor physics, especially considering the category of Machine Learning (ML), with the aim of describing the application scenarios, frontier topics, unsolved challenges and future research directions. From equation solving and state parameter prediction to nuclear industry applications, this paper provides a step-by-step overview of ML methods applied to steady-state, transient and combustion problems. Most literature works achieve industry-demanded models by enhancing the efficiency of deterministic methods or correcting uncertainty methods, which leads to successful applications. However, research on ML methods in reactor physics is somewhat fragmented, and the ability to generalize models needs to be strengthened. Progress is still possible, especially in addressing theoretical challenges and enhancing industrial applications such as building surrogate models and digital twins.

Problem

• Review AI applications in reactor physics.
• Explore ML methods for nuclear industry challenges.
• Address generalization and theoretical gaps in ML models.

Innovation

• AI enhances reactor physics simulations and safety.
• Machine Learning improves nuclear reactor core management.
• Surrogate models and digital twins advance industry applications.