The Nuclear Regulatory Commission (NRC) is looking into how it could use AI and machine learning in nuclear power generation and is seeking comment from the commercial nuclear power industry to help in that inquiry, according to an NRC request for comment published April 21.
The NRC is looking to understand how the commercial nuclear power industry is currently using AI and ML in a variety of applications, including plant design, operations, maintenance or decommission. The agency also wants to know what areas of nuclear power management would benefit most from AI and ML implementation, and what AI and ML methods are currently being used or will be used in the near future.
“AI and ML are emerging, analytical tools, which, if used properly, show promise in their ability to improve reactor safety, yet offer economic savings,” the request for comment says. “The NRC requests comments on issues listed below in this solicitation to enhance the NRC’s understanding of the short- and long-term applications of AI and ML in nuclear power industry operations and management, as well as potential pitfalls and challenges associated with their application.”
Other questions NRC is posing include:
- “What are the potential benefits to commercial nuclear power operations of incorporating AI/ML in terms of (a) design or operational automation, (b) preventive maintenance trending, and (c) improved reactor operations staff productivity?;”
- Whether AI and ML can improve the efficiency of nuclear regulatory oversight, and how?;
- The general level of AI and ML expertise already present in the nuclear power industry?; and
- Current adoption of AI and ML by the industry.
The request gives the public, industry and other stakeholders 30 days to submit comment.