Nuclear Technology

Artificial Intelligence for Fault Detection and Diagnosis in Nuclear Systems

Artificial intelligence (AI) is increasingly transforming fault detection and diagnosis across nuclear systems, offering new opportunities to enhance safety, reliability, operational efficiency, and resilience. From nuclear power plants and fuel cycle facilities to research reactors, radioactive waste management systems, and other nuclear installations, AI-based methods can support the early identification of abnormal conditions, the localization of faults, and the interpretation of complex system behavior. Advances in machine learning, deep learning, data analytics, and hybrid physics-informed approaches enable the extraction of meaningful insights from diverse sources, including sensor measurements, process data, operational records, and simulation results. However, their application in nuclear environments presents distinctive challenges, including limited examples of abnormal conditions and rare events, changing operating conditions, high requirements for model interpretability, and the need for rigorous verification and validation (V&V) and uncertainty quantification.

Reliable fault detection and diagnosis support the safe, secure, efficient, and economic operation of nuclear systems. Artificial intelligence and machine learning approaches may identify subtle fault signatures, detect anomalies at early stages, support timely decision-making, and improve understanding of system behavior across nuclear power generation, fuel cycle operations, research facilities, and other nuclear applications. This collection provides a platform to present novel methodologies, demonstrate practical applications, address implementation challenges, and advance the scientific foundations of AI-enabled nuclear diagnostics. By cataloguing relevant works together, this collection can foster interdisciplinary thinking and accelerate the responsible development and deployment of trustworthy AI solutions throughout the nuclear sector.

This collection brings together research on emerging AI methodologies, innovative applications, and practical implementation strategies, highlighting their potential and limitations in advancing intelligent monitoring, diagnostics, and decision support across the nuclear industry. Relevant areas of interest include: operations and maintenance of nuclear facilities; online monitoring approaches to support safety, safeguards, and security; trustworthiness and transparency of AI-assisted detection and diagnostics; and applications in light water reactors, advanced reactors, fuel cycle facilities, or isotope production. We particularly encourage submissions that explore novel architectures, address regulatory concerns, or present standardized benchmark datasets.

Keywords:

1. Artificial Intelligence
2. Anomaly Detection
3. Diagnostics
4. Machine Learning
5. Decision Support

Guest Advisors:

Dr. Jamie Baalis Coble is a Professor in the Department of Nuclear Engineering & Engineering Physics at the University of Wisconsin-Madison. Her research interests expand on past work in nuclear system monitoring and prognostics to incorporate system monitoring, equipment condition assessment, and remaining useful life estimates into risk assessment, operations and maintenance planning, and optimal control algorithms. Dr. Coble is currently pursuing research in applications of data analytics, machine learning, and artificial intelligence to support risk-aware, economic operations and maintenance of nuclear power facilities, as well as applications of these techniques in other domains such as nuclear security and advanced manufacturing.

Dr. Linyu Lin is a Nuclear Engineer in the Plant Optimization department at Idaho National Laboratory, where he leads research on uncertainty-aware monitoring, diagnostics, and control for nuclear energy systems. His work spans explainable discrepancy detection and diagnosis for digital twins, physics-informed and hybrid machine learning for predictive maintenance and thermal-performance loss diagnosis at operating light water reactors, agentic AI workflows for multimodal anomaly detection, and autonomous control of research and advanced reactors validated through hardware-in-the-loop testing. He has led and contributed to DOE LDRD, Technology Commercialization Fund, and Light Water Reactor Sustainability projects in collaboration with utilities, vendors, and universities. Dr. Lin received his Ph.D. in nuclear engineering from North Carolina State University.

Disclosure Statement: Dr. Jamie Baalis Coble is an Associate Editor of Nuclear Technology. Dr. Linyu Lin has no conflicts of interest to disclose.

All manuscripts submitted to this Article Collection will undergo a full peer-review; the Guest Advisor for this Collection will not be handling the manuscripts (unless they are an Editorial Board member).

Please review the journal scope and author submission instructions prior to submitting a manuscript.

The deadline for submitting manuscripts is 31 August 2027.

To submit your papers to this Article Collection, please:

1. Check "yes" for the question, "Are you submitting your paper for a specific special issue or article collection?"
2. Select the Article Collection "Artificial Intelligence for Fault Detection and Diagnosis in Nuclear Systems" from the drop-down menu under the question, "Special Issue or Article Collection Name."

Please contact Jieyu Zheng at Jieyu.Zheng@tandf.co.uk  with any questions relating to this Article Collection.

Open Select Journals that publish open access require payment of an article publishing charge (APC); this includes manuscripts submitted to an Article Collection. APCs vary between journals and may depend on the article type, please refer to our Open Access Cost Finder tool. APCs may also differ from the standard APC if your institution or research funder has an open access publishing agreement or an open access membership with Taylor & Francis.​