The objective of this study is to integrate artificial intelligence (AI) and machine learning (ML) in the nuclear power industry to address environmental issues. The problem addressed stems from the nuclear sector's growing need to modernize aging infrastructure and meet stricter environmental and regulatory standards, all while facing data scarcity, cybersecurity concerns, and a shortage of skilled personnel.
Methods: This study employs a systematic literature review and analysis of case studies to assess the current applications of AI and ML in nuclear power. It evaluates various AI techniques, including neural networks, fuzzy logic, and deep learning, in tasks such as fault diagnostics, reactor control, and performance optimization. Additionally, this study examines challenges related to the deployment of these technologies, focusing on data requirements, model complexity, and compliance with nuclear safety regulations.
Results: AI and ML have demonstrated considerable success across multiple domains within nuclear operations. Neural networks and fuzzy logic systems have enhanced the accuracy of reactor monitoring and the stability of control processes. Deep learning models have enabled the real-time optimization of operational parameters and predictive maintenance, resulting in significant reductions in downtime and maintenance costs. However, the implementation of these systems is hindered by the lack of explainable AI frameworks and robust datasets necessary for training high-performance models.
Conclusion: Our findings underscore the transformative potential of AI and ML in the nuclear sector. To fully harness their capabilities, the industry must overcome existing barriers through targeted research focusing on explainable AI, improved data governance, and adaptive regulatory frameworks.