In a New Report DOE Tackles AI for Nuclear Energy

• In a New Report DOE Tackles AI for All Forms of Energy and Grid Management
• DOE’s Report Details Initiatives Related to AI and Nuclear Energy
• UK Office of Nuclear Regulation Issues Strategy for Using AI

In a New Report DOE Tackles AI for All Forms of Energy and Grid Management

The US Department of Energy is facing one of the most significant technological challenges in its history. It is how to assess the potential use of artificial intelligence by the nuclear energy industry and, of equal importance, how to use AI to address the industry’s grand challenges for energy and grid management across the entire organization.

To this end, DOE has issued a report prepared by a group of about 100 experts on AI/machine learning and applied energy convened at Argonne National Laboratory in December 2023 to map out future needs related to utilizing AI. The goal was to detail pressing technical challenges and propose AI-assisted solutions.

The report covers grand challenges and opportunities for short-and-long-term action related to the use of artificial intelligence in the deployment and management of all aspects of commercial nuclear energy and other forms of energy and energy management and regulation. Download the report here

Five Vital Areas

The focus of the report centers is on five areas vital to the energy future of the U.S.  DOE’s experts wrote in the summary of the report.

• Nuclear Power,
• Power Grid,
• Carbon Management,
• Energy Storage, and
• Energy Materials.

“It will be essential to integrate these together and with other efforts in AI for science and technology. Complexity, the large-scale effort involved, real-time decision making required, robustness of systems, and safety implications all pose extra challenges.”

“The Grand Challenges described in this report span multiple disciplines and have not been solved by conventional methods. The power of AI for solving such problems lies in its capacity to simultaneously handle multiple system characteristics while incorporating both data and specific domain models, and to do so on a scale and at a complexity otherwise not possible.”

Highlights of the DOE Report Related to Nuclear Energy

DOE wrote that the nuclear industry must adopt and, where required, further develop the latest AI tools and technologies.

“AI’s transformative potential is particularly relevant in methodologies which could drastically improve the economics of nuclear system design and operation.”

“These challenges span multiple scientific and engineering disciplines and require AI’s unique ability to process vast amounts of data and integrate physics models on a scale previously unattainable. AI offers unparalleled knowledge capture and the capability to discern cross-disciplinary connections.”

This advantage is critical in three specific challenge areas where AI/ML can surpass the performance of human teams:

(1) streamlining the licensing and regulatory process;
(2) accelerating deployment; and
(3) facilitating unattended operation. Embracing and extending AI capabilities could significantly enhance the nuclear industry’s efficiency and innovation, all while continuing to improve safety.

& & &

DOE Details Initiatives Related to AI and Nuclear Energy

In a long press statement spanning all forms of energy generation, management, regulation, and use, DOE spelled out some of the key areas where it is working on challenges and opportunities for using AI in the nuclear energy field. Here are some key highlights.

Data Centers and AI

DOE has established a new Working Group on Powering AI and Data Center Infrastructure. The Secretary’s Energy Advisory Board chartered a new working group to make recommendations by June on meeting energy demand for AI and data center infrastructure.

DOE announced that over the next several months, DOE will convene utilities, clean energy developers, data center owners and operators, and regulators in localities experiencing large load growth. In these areas DOE is playing catch up since the data center industry has been talking about these issues for over a year especially with regard to SMRs.

Last February this blog recommended that DOE take the kinds of actions as detailed in its April 2024 report.

“Getting power from nuclear reactors, large or small, is a relatively new idea for data center managers. It is important for the nuclear industry to emphasize that the path for data centers to tap SMRs, or other types of nuclear power plants, is through power purchase agreements thus avoiding the inherent risks of building them.”

“A forum composed of industry leaders from the data center and nuclear worlds, along with data center customers, would help facilitate the development of business arrangements whereby the evolving rapid growth in the size of data centers and their demand for electricity could be satisfied by nuclear energy.”

As far as developing priority based lists of potential applications of AI to nuclear energy, the Internet is swimming in proposals for using the technology in all kinds of ways. This blog posted its own short list of AI applications for nuclear energy  in March 2024.

Some ideas about using nuclear energy to power data centers are likely to be over reach. In an invited article, Data Center Dynamics published an assessment on April 9, 2024, by this blog, of a reported proposal that Microsoft would seek to buy 5 GW of power from nuclear energy utilities for its data centers.  Microsoft has a big job ahead of it and that is to convince power producers in the US, and globally, that it isn’t biting off more than it can chew.

Other concrete actions DOE is taking include addressing the need to quantify data center load growth and forecasted its near-term potential. To better understand how AI will affect future energy consumption, DOE’s Lawrence Berkeley National Laboratory is currently performing a detailed national analysis of regional energy and water use across the nation’s data centers.

• Site Selection for Nuclear Power Plants

The new VoltAIc Initiative to use AI to help streamline siting and permitting at the Federal, state, and local level. DOE is investing $13 million in the initiative to build AI-powered tools to improve siting and permitting of clean energy infrastructure and has partnered with Pacific Northwest National Laboratory (PNNL) to develop PolicyAI, a policy-specific Large Language Model test bed that will be used to develop software to augment National Environmental Policy Act and related reviews.

Cybersecurity for Nuclear Power Plants

DOE’s Office of Cybersecurity, Energy Security, and Emergency Response (CESER) will be convening energy stakeholders and technical experts over the coming months to collaboratively assess potential risks that the unintentional failure, intentional compromise, or malicious use of AI could pose to the grid, as well as ways in which AI could potentially strengthen grid resilience and our ability to respond to threats.

&  & &

UK Office of Nuclear Regulation Issues Strategy for AI

• ONR shares pro-innovation approach to regulating AI in the nuclear sector

The UK Office for Nuclear Regulation (ONR) has published a policy paper outlining its pro-innovation approach to regulating artificial intelligence (AI) in the nuclear sector. (download the report)

The paper shares ONR’s strategic approach to AI and how this aligns with the expectations outlined in the UK government’s 2023 AI Regulation White Paper, which establishes a framework for regulators to interpret and then apply to AI within their remits. The government’s framework is based on five principles:

• safety, security and robustness
• appropriate transparency
• fairness
• accountability and governance
• contestability and redress

ONR and other regulators were recently asked by the Secretary of State for Science, Innovation and Technology and the Secretary of State for Work and Pensions to publish updates on how their regulatory approaches align with these principles.

In its update, ONR describes how its existing goal-setting and non-prescriptive regulatory regime provides a supportive environment for regulated entities to innovate, in particular its focus on outcomes and technology neutrality to ensure the most cost-effective solutions.

ONR’s program of work aligning with the five principles in the white paper includes:

Taking part in a five-year UK Research and Innovation (UKRI)-funded  project alongside universities, nuclear site licensees and other regulators, focusing on the development of robotics for the UK nuclear industry and how to structure associated safety issues and applications.

Developing regulatory sandboxing as a tool for utilities and regulators to test innovative technologies in collaboration while exploring the suitability of the existing regulatory framework;

Updating guidance for ONR inspectors, supported by early engagement with utilities to foster open dialogue and explain our regulatory expectations;

Regularly engaging with international partners and UK regulators, sharing best practice and collaborating to build regulatory capability; and

Further work is planned during the next 12 months, including building capability within ONR’s AI-focused team to enable open conversations about new technologies and approaches within ONR.

# # #