NRC Explores the Potential Role of Artificial Intelligence in Nuclear Power Business Operations | Morgan Lewis – Up & Atom


As artificial intelligence (AI) and machine learning tools are increasingly adopted in various products and industries, NRC has begun to study the roles that these technologies can play in commercial nuclear operations. On April 21, as part of its study, the NRC Office of Nuclear Regulatory Research public comments requested on the role of these technologies “in the different phases of the operational experience of nuclear electricity production and plant management”. NRC seeks feedback on “the state of practice, benefits and trends [these technologies’] computational tools and techniques in predictive reliability and predictive safety assessments in the commercial nuclear industry. These technologies “are emerging analytical tools which, if used correctly, show promise in their ability to improve reactor safety, while providing economic savings”. Comments are due before May 21, 2021.

NRC intends to use the feedback to improve its understanding of the benefits of AI and machine learning as well as the “potential pitfalls and challenges associated with their application.”

NRC invited comments on the following questions:

  1. What is the status of the development of the commercial nuclear industry or the use of artificial intelligence / machine learning tools to improve certain aspects of the design, operation, maintenance or the decommissioning of nuclear power plants? What tools are used or developed? When should tools under development be used?
  2. What areas of commercial nuclear reactor operation and management will benefit most, and least, from the implementation of AI / machine learning? Possible examples include, but are not limited to, inspection assistance, incident response, power generation, cybersecurity, predictive maintenance, safety / risk assessment, monitoring of plant performance. systems and components, operational / maintenance efficiency and shutdown management.
  3. What are the potential benefits for commercial nuclear operations of integrating AI / machine learning in terms of (a) operational design or automation, (b) preventive maintenance trend, and (c) improved staff productivity? operation of reactors?
  4. What artificial intelligence / machine learning methods are currently in use or will be used in the near future in the management and operation of commercial nuclear power plants? Examples of possible machine / machine learning methods include, but are not limited to, artificial neural networks, decision trees, random forests, support vector machines, clustering algorithms, reduction algorithms dimensionality, data mining and content analysis tools, Gaussian processes, Bayesian methods, natural language processing and image scanning.
  5. What are the advantages or disadvantages of a high-level top-down strategic goal to develop and implement AI / machine learning in a wide range of general applications versus a targeted ad-hoc, case-by-case approach? by case?
  6. When it comes to AI / machine learning, what stage of technology adoption is the commercial nuclear industry currently experiencing and why? The current model of technology adoption characterizes phases into categories such as innovation phase, early adoption phase, early majority phase, late majority phase, and lag phase.
  7. What are the challenges of balancing the costs associated with the development and application of AI / machine learning tools against the operational and technical advantages of the plant when integrating AI / learning automatic in operational decision-making and workflow management?
  8. What is the general level of AI / machine learning expertise in the commercial nuclear power industry (e.g. expert, well versed / proficient, or beginner)?
  9. How will AI / machine learning affect the commercial nuclear power industry in terms of efficiency, costs and competitive positioning relative to other sources of power generation?
  10. Does AI / machine learning have the potential to improve the efficiency and / or effectiveness of nuclear regulatory oversight or otherwise affect regulatory costs associated with safety oversight? If so, in what ways?
  11. AI / machine learning typically requires the creation, transfer, and evaluation of very large amounts of data. What are the concerns, if any, regarding data security in relation to the exclusive operating experience of nuclear power plants and design information that may be stored in remote networks off-site?

The NRC is in the early stages of its review and the agency does not promise to use the information gathered for formal regulatory action. Morgan Lewis will continue to monitor the regulatory initiatives of the NRC.

[View source.]

Source link


About Author

Leave A Reply