Enabling Safe, Cost‑Effective Long‑Term Operation of the Global Nuclear Fleet

Challenge

The pursuit of carbon reduction goals and energy security while maintaining grid reliability and affordability has become a strategic imperative for countries around the world. More than two‑thirds of the world’s 400+ operating nuclear power reactors are now over 30 years old, and nearly 200 reactors worldwide have already been authorized for long-term operation. Extending the operating lives of existing nuclear plants—known as Long‑Term Operation (LTO)— is one of the fastest, most cost‑effective ways to deliver large‑scale, carbon‑free electricity.

Achieving LTO requires addressing complex technical challenges related to materials degradation, aging management, regulatory compliance, and asset investment decisions, all while maintaining the highest safety standards. The International Atomic Energy Agency (IAEA) TECDOC‑2117, Good Practices and Lessons Learned from the Long‑Term Operation of Nuclear Power Plants (2026), documents how utilities worldwide have successfully extended nuclear plant lifetimes by combining rigorous safety reviews with proactive modernization and aging management.

EPRI’s Role

EPRI has played a foundational role in advancing the technical basis for LTO through practical science‑based aging management, materials research, and long‑term asset management methodologies.

EPRI-developed frameworks—including Materials Degradation Matrices (MDMs), Issue Management Tables (IMTs), and long‑term asset management approaches—are referenced throughout IAEA TECDOC-2117 in the context of global best practices for identifying aging mechanisms, prioritizing mitigation strategies, and supporting risk‑informed decision‑making for extended operation.

What Was Achieved

IAEA TECDOC‑2117 documents common success factors across regions and reactor technologies, that have emerged which closely align with EPRI research and guidance:

• Predictive understanding of materials degradation, enabling early identification and mitigation of life‑limiting issues in reactor vessels, internals, steam generators, cables, and concrete structures.

• Systematic aging management programs (AMPs) integrated into plant operations, inspections, and maintenance planning.

• Risk‑informed modernization strategies, including selective component replacement, digital upgrades, and targeted refurbishment.

• Long‑term capital planning tools that support economically optimized investment decisions over 60‑ to 80‑year operating horizons.

The report highlights that plants applying these approaches have demonstrated safe operation beyond original design life while maintaining high capacity factors and strong safety performance.

Value to the Industry

The practices summarized in TECDOC‑2117 reinforce the value of EPRI’s collaborative research model for nuclear plant owners, operators, and regulators:

• Safety Assurance – Science‑based aging evaluations support continued compliance with evolving regulatory expectations.

• Cost Control – Risk‑informed asset management enables smarter capital allocation and avoids unnecessary replacements.

• Operational Reliability – Proactive degradation management reduces unplanned outages and long‑term operational risk.

• Global Alignment – Shared methodologies help harmonize LTO practices across countries, reactor types, and regulatory regimes.

Global Impact

The TECDOC draws on case studies from North and South America, Europe, Asia, and beyond, covering pressurized water reactors, boiling water reactors, CANDU, and VVER designs. EPRI research is cited in connection with international initiatives such as aging lessons‑learned programs, materials research collaborations, and long‑term operation peer reviews. These efforts help ensure that insights gained in one fleet benefit the entire global nuclear community.

Looking Ahead

As the industry moves toward LTO beyond 60 years—and in some cases toward 80 years and beyond—the challenges of material degradation, knowledge retention, and modernization will intensify. TECDOC‑2117 concludes that continued international collaboration and applied research will be essential.

• EPRI remains uniquely positioned to support this next phase of long‑term operation through:

• Advanced materials research for high‑fluence and late‑life conditions

• Enhanced inspection, monitoring, and data integration technologies

• Risk‑informed, fleet‑wide asset management tools

• Knowledge transfer to support workforce continuity

Why It Matters

By translating decades of operating experience into practical, science‑based guidance, EPRI is helping utilities worldwide operate safely, reliably, and economically—today and for decades to come.

Related EPRI Products:

• ELECTRIC POWER RESEARCH INSTITUTE, EPRI Materials Degradation Matrix (Rev. 5), Technical Report 3002030559, EPRI, Palo Alto, CA (2024).

• ELECTRIC POWER RESEARCH INSTITUTE, Advanced Nuclear Technology: Advanced Light Water Reactor Utility Requirements Document, Revision 13, Technical Report 3002003129, EPRI, Palo Alto, CA (2014).

• ELECTRIC POWER RESEARCH INSTITUTE, Plant Support Engineering: Proactive Obsolescence Management, Technical Report 1019161, EPRI, Palo Alto, CA (2009).

• ELECTRIC POWER RESEARCH INSTITUTE, Plant Engineering: Guideline for the Acceptance of Commercial‑Grade Items in Nuclear Safety‑Related Applications: Revision 1 to EPRI NP‑5652 and TR‑102260, Technical Report 3002002982, EPRI, Palo Alto, CA (2014).

• ELECTRIC POWER RESEARCH INSTITUTE, Materials Reliability Program: Pressurized Water Reactor Issue Management Tables (Rev. 4) (MRP‑205), Technical Report 3002018255, EPRI, Palo Alto, CA (2020).

• ELECTRIC POWER RESEARCH INSTITUTE, BWR Vessel and Internals Project, Boiling Water Reactor Issue Management Tables (BWRVIP‑167, Rev. 5), Technical Report 3002032174, EPRI, Palo Alto, CA (2025).

• ELECTRIC POWER RESEARCH INSTITUTE, Materials Reliability Program: VVER Issue Management Tables (MRP‑471), Technical Report 3002021033, EPRI, Palo Alto, CA (2021).

• ELECTRIC POWER RESEARCH INSTITUTE, Seismic Probabilistic Risk Assessment Implementation Guide, EPRI‑103959, EPRI, Palo Alto, CA (1994).

• ELECTRIC POWER RESEARCH INSTITUTE, Seismic Fragility and Seismic Margin Guidance for Seismic Probabilistic Risk Assessments, EPRI 3002012994, EPRI, Palo Alto, CA (2018).

EPRI’s 2025 Quantum Challenge brought out the best‑of‑the‑best in the spirit of quantum innovation. With submissions from individuals and teams representing 12 different countries, last year’s Challenge highlighted creative and technically diverse applications of quantum technologies to help advance fusion energy.

That momentum is continuing with the newly opened 2026 EPRI Cyber Quantum Challenge, which invites innovators to develop quantum‑enabled and quantum‑resilient cybersecurity solutions tailored to the real‑world needs of the electric power sector. As interest and participation continue to grow, the results of the 2025 Quantum Challenge offer valuable insight into what strong, impactful submissions look like in practice.

The three winning projects—led by Kory Burns (University of Virginia), Justus Lau (University of Heidelberg), and Ridwan Sakidja (Missouri State University)—reflect the diversity of thinking and interdisciplinary collaboration needed to move quantum ideas from the laboratory to real‑world application.

Here are some key takeaways from our 2025 winners:

Tip 1: Get it on paper.

Each of the team leads for our winning projects noted that the first and most important obstacle was to get their concept on paper. This step can help cement ideas, think through nuances, and find connection points.

“One of the reasons I wanted to write this [proposal] is because I had an idea I was thinking about, and I just really wanted to have it written down on paper somewhere,” said Kory. “If I write it down it means that I kind of get to clear my thoughts and think of some cohesive timeline in which I could accomplish doing this and write to a broader audience about why this is important.”

For Justus, documenting concepts helps connect related ideas. “I’m not necessarily the best at one niche topic, but I think I am quite good at connecting things at a creative point,” said Justus.

So, when in doubt, just write it out... because you might just have a winning idea.

Tip 2: Embrace the detours. 

Exploring new quantum applications and use cases requires patience and flexibility. Even experts will find unexpected barriers, challenges or knowledge gaps that need to be resolved.

When this happens, don’t be discouraged. Instead, take the advice of our most recent winners and embrace the detours.

Ridwan and his team uncovered one of these unexpected challenges while attempting the EPRI Quantum Challenge. “We realized there was a blind spot here that not many people realized,” he said. “The challenge was bigger than we initially thought.”

While finding obstacles isn’t always easy, the winners agree that knowing about an issue sooner, rather than later, is always better – and, sometimes, an obstacle can turn into an opportunity.

For Ridwan and his team, uncovering this expanded challenge opened a completely new area of research, one with exciting and wide-reaching applications. Since discovery during the EPRI Quantum Challenge, the team has been actively pursuing research in this new area and expects to publish findings soon.

Tip 3: Seek diverse perspectives.

Over and over again, our winners emphasized the importance of cross-disciplinary and interdisciplinary collaboration. They agreed that these types of approaches held the greatest potential for quantum advantage and some of the most interesting and exciting quantum use cases.

By seeking diverse perspectives, others may be able to fast-track interdisciplinary collaborations and quantum applications for broader uses.

As the 2026 EPRI Cyber Quantum Challenge gets underway, now is the time to turn bold ideas into actionable proposals. Whether you are building on an existing concept or developing a new approach, these lessons from past winners can help strengthen your submission and accelerate progress from concept to impact.

AI’s surge is no longer a future scenario—it’s already beginning to reshape the U.S. power system. EPRI’s 2026 Powering Intelligence study, an update from its 2024 study on the same topic, shows just how consequential that shift could be: data centers, particularly AI‑focused facilities, are emerging as the largest driver of electricity demand growth this decade. By 2030, they could account for 9% to 17% of total U.S. electricity demand, more than double today’s share, with eight states facing far higher concentrations.

Why AI Is Reshaping the Energy Landscape

Today, data centers consume roughly 4–5% of all U.S. electricity. The projected range of 9% to 17% by 2030 is 60% higher than the range estimated in EPRI’s 2024 report, driven by a surge of newly announced and under construction facilities. Even at the low end of the projection, the trend represents substantial upward pressure on power systems nationwide.

AI is already responsible for 15–25% of data center power use today, and its share is rising fast. Going forward, a single modern data center may require 100 to 1,000 megawatts of capacity, equivalent to the electricity needs of hundreds of thousands of homes or a small city. While efficiencies in hardware and algorithms continue, they are increasingly outpaced by the rapid expansion of AI workloads, ranging from image and video analysis to emerging agent‑based systems. In short, the growth curve is steep, with significant uncertainty through 2030 and even broader uncertainty over longer time frames.

State-by-State Implications: A Closer Look

Beyond national trends, the study highlights critical state‑level insights, underscoring a resurgence in regional U.S. electric load growth that is shaping how utilities, policymakers, businesses, and communities prepare for the next wave of digital infrastructure buildout.

Developers are looking for a combination of factors when choosing project location: quick access to large amounts of power is often paramount, in addition to available land and water, proximity to digital infrastructure and labor markets, fast permitting, attractive incentive structures, and clean energy resources

Virginia is the largest and most concentrated U.S. data center market. Data centers already consume over 20% of the state’s electricity, the highest share in the nation. By 2030, that share is projected to increase to 39%–57%, driven by many projects already under construction or in advanced planning. Virginia’s scale makes data centers a dominant driver of grid planning, peak load growth, and infrastructure investment.

At the same time, the geographic diversity of data center activity is increasing. Under the medium growth scenario, seven additional states could see data centers exceed 20% of total electricity demand by 2030: Arizona, Indiana, Iowa, Nebraska, Nevada, Oregon, and Wyoming.

Texas is projected to remain one of the largest data center markets in absolute terms. Its total data center demand is comparable to Virginia when cryptocurrency mining is included. Georgia and North Carolina also show strong growth, reflecting continued concentration in major regional markets.

Developers of large AI training centers are prioritizing power availability and large tracts of land. Louisiana, Mississippi, New Mexico, Ohio, and Pennsylvania are favorable for new areas of concentrated development, where data center shares are projected to exceed 10% of total electricity demand by 2030. These dynamics underscore why state specific forecasting, flexible load strategies, and coordinated planning are critical as data centers scale.

A Defining Moment for the U.S. Power System

The study emphasizes that the grid can support this level of growth, but it may require accelerated deployment. An increase in transmission and generation capacity typically requires up to 10 years, whereas new data centers can be developed in a fraction of that time. This timing mismatch emphasizes the need for coordinated planning among utilities, developers, and regulators.

Through its DCFlex initiative, EPRI is collaborating with the power and digital industries to explore data center flexibility and alleviate grid bottlenecks. Its field testing on operating data centers is demonstrating a range of flexibility options – from varying computational demands to cleaner drop-in replacements for diesel generations - that can accelerate the integration of large loads into power systems, while sustaining both grid reliability and data center operations.

Big Picture

Total U.S. electric load is projected to grow at a rate of 2% to 3.6% per year through 2030, a significant increase from 0.8% annually from 2019 to 2024. And data centers aren’t the only source of growth. Electrification of vehicles and other sectors, industrial re-shoring, and specialized manufacturing growth are all poised to contribute to an expanding role for electricity in the modern economy.

This study underscores the importance of enhanced forecasting, better use of existing energy assets, and deeper collaboration to ensure a reliable and affordable power supply to support these critical trends.

Transportation electrification is reshaping how utilities approach load growth, infrastructure investment, and long‑term planning. What was once a future consideration is now unfolding in real time, as electric vehicle adoption expands into every region of the country. EPRI EV sales data for 2025 shows year‑over‑year growth of approximately 16% in the south central region, 15% in the southeast, and 12% in the Midwest. As the total cost of ownership improves, fleets are increasingly electrifying as a near‑term business decision. The adoption of EVs, particularly in fleets with large charging loads, is outpacing traditional grid planning timelines, particularly as EV demand coincides with broader building electrification, requiring utilities to engage earlier and plan differently than before.

For decades, utilities planned around large, predictable loads with long lead times, catered to industrial facilities, commercial developments, or new substations that would take years to materialize. Electric transportation has flipped that model. Medium- and heavy-duty EV fleets can now be procured and deployed in months, and create high-impact loads that demand earlier coordination and more information.

On a recent episode of EPRI’s podcast, The EPRI Current, EPRI transportation experts Mike Rowand and Watson Collins discussed how EV load growth, particularly from fleets, may ultimately rival or surpass other high-profile sources of demand, such as data centers. The key question for utilities is no longer just whether new load is coming, but where, how quickly, and how much visibility they will have in advance. Without early insight, utilities risk reactive upgrades, higher costs, and strained customer relationships. With over 3,200 electric utilities in the U.S., each with its own processes and points of contact, fleet operators and charging providers often struggle to determine whom to engage and when.

GridFAST™ was developed to address this early-stage coordination gap. The platform provides a secure, centralized portal that standardizes early communication between utilities and EV commercial customers. Using a standardized intake form, customers can submit preliminary project information and match a site to the correct utility years before construction begins across any U.S. service territory.

For utilities, this early signal enables proactive planning – not just for individual projects, but for aggregated demand across territories. Importantly, GridFAST does not replace a utility’s interconnection or service request process; instead, it complements them by moving the initial conversation upstream, so that changes remain affordable and timelines remain flexible. Fleet operators using GridFAST have highlighted how the platform simplifies multi-site planning for them by identifying a single point of contact for them, and through the standardized project information that is captured for the utility in the tool.  This translates into earlier visibility into future demand that might otherwise arrive with little notice, allowing investments to align more closely with actual customer needs.

As Rowand and Collins note in The EPRI Current, uncertainty does not equal inaction. Utilities don’t need perfect forecasts to move forward, but they do need better information earlier in the process. GridFAST provides those signals, enabling utilities to plan with greater confidence even as the pace of electrification evolves. For EV customers, early engagement can prevent overbuilding, reduce delays, and align charging strategies with the grid's realistic capacity.

With uncertainty around gas and diesel prices on the horizon, there is even greater renewed interest in EVs. Tools enabling earlier, clearer communication between customers and utilities show that utilities and customers can move faster together, without sacrificing reliability or affordability. Developed in collaboration with utilities, fleets, and charging providers, GridFAST is a cornerstone of EPRI’s electric transportation program and EVs2Scale2030™ initiative, which focuses on addressing barriers to large-scale transportation electrification.

To learn more or request a GridFAST demo, contact [email protected].

Europe’s power system is entering a period of accelerated change, with new technologies, operating models, and policy pressures reshaping how grids are planned and run. Against that backdrop, discussions held during EPRI’s 10th European Workshop Week reflected a broader reality facing the sector: existing frameworks are being tested by the pace and complexity of transformation.

The value of those conversations lay less in the gathering itself and more in the space they created for practical exchange. System operators, utilities, researchers, and technology providers compared notes on how research and innovation are translating into day‑to‑day operations—and where gaps between ambition and implementation still persist.

The ten insights that follow capture recurring themes from those discussions, highlighting practical priorities now being actively debated and refined by those closest to real‑world delivery.

1) Reliability is evolving due to electrification, complexity, and uncertainty

Europe's reliability challenge is driven by fast electrification, increased grid complexity, extreme events, and strategic uncertainty. Maintaining reliable service requires coordinated efforts, better risk detection, and flexible operational strategies.

As electrification accelerates, more sectors and services are becoming tightly linked to the power system, increasing the impact of disruption and raising expectations for resilience. System operators now face increased pressure to predict challenges sooner, adapt faster, and handle reliability amid greater uncertainty.

2) Innovation isn’t the bottleneck—implementation is

Innovation programming focused not just on generating new ideas, but on applying innovation within utilities through internal collaboration, governance, and execution. While many pilot projects exist, scaling them is challenging due to the need for operational, digital, and regulatory alignment.

As Durgesh Manjure, EPRI’s Vice President of Integrated Grid & Energy Systems, said, “There's broad alignment on the drivers of risk. What is changing is really the recognition that progress depends not only on developing new solutions and technologies and capabilities, but also ensuring a collaborative, coordinated path forward to make sure that we consider a holistic system needs to make sure that we build a safe, reliable and affordable electric system in the future.”

3) Modeling and validation must keep pace with a more digital grid and large loads from data centers

As AI and data centers expand, better modeling, validation, and practical implementation are increasingly essential. Evolving technologies require updated practices to ensure effectiveness in real-world power systems.

Participants emphasized the importance of using the right level of model reliability, supported by stronger screening methods, clearer governance, and better use of generic and performance-based models. Validation is also becoming more structured, blending benchmark tests with operational data to build confidence in real world behavior.

Taken together, these shifts reinforce a clear message: modern modelling is not an academic exercise. It is a practical, system level capability that underpins planning decisions, operational readiness, and the reliable integration of new technologies at scale.

4) Transmission operations are adapting to a faster, more dynamic system

Transmission operators are already working in a system that moves faster, behaves less predictably, and leaves less margin for error. As inverter-based resources, large l loads from AI, and tighter operating conditions become more common, traditional assumptions about grid behaviour are being tested.

This ongoing transformation is reshaping daily operations, leading to a stronger dependence on real-time awareness, enhanced monitoring of oscillations and system stability, and more frequent use of advanced decision-support tools within control rooms. Operators are now prioritizing coordination across regions, control centers, and disciplines to ensure rapid responses when conditions shift.

Keeping the system reliable now depends as much on modern operational tools, training, and coordination frameworks as it does on physical infrastructure. Transmission operations are evolving from managing a largely predictable system to actively navigating a more dynamic grid.

5) Distribution & DER: digitalisation and customer tech are rewriting the rules

A key takeaway is that the rapid expansion of customer-owned technologies—such as EVs, heat pumps, rooftop PV, and storage —is redefining the planning and operation of distribution networks. Digitalization is no longer just a useful tool; it has become essential infrastructure for distribution system operators (DSOs).

Another important insight is the increasing need for DSOs to have much greater visibility and control at medium- and low-voltage levels. Achieving this depends on interoperable data platforms, modern SCADA systems, and advanced tools in control rooms. Moreover, consistency is crucial to progress from pilot projects to solutions that can be scaled and reliably operated.

Innovation only delivers value when it can be translated into processes DSOs can reliably run every day. Digital tools, flexibility markets, and customer participation must be designed with execution in mind to allow DSOs can safely integrate new technologies while keeping networks reliable, secure, and ready to grow.

6) Renewables are entering a new phase: performance, reliability, and value

As wind and solar fleets mature, the focus is shifting from capacity growth to operational excellence. Even small performance losses now have material financial impacts, putting availability, reliability, and O&M discipline front and centre.

Solar operations are increasingly using inverter-level benchmarking and predictive diagnostics to improve reliability and maximize lifetime value. In wind, operators rely on SCADA-based analytics and health monitoring to detect performance issues early, minimize downtime, and control operating expenses.

Across both technologies, data driven performance and reliability are becoming essential to protect revenue and deliver long-term value from renewable assets.

7)  Data centers: flexibility becomes essential to speed‑to‑power

Rapid growth in data centers—driven in large part by AI—has made large loads a central grid challenge. At European Workshop Week, discussions focused on how flexibility can unlock capacity and accelerate connections without undermining reliability or affordability.

Through EPRI’s DCFlex initiative, participants explored a common framework, Flex MOSAIC™, for classifying data center flexibility, giving system operators clearer signals on response speed, duration, and certainty. This can help reduce planning uncertainty, support faster interconnection, and link large load flexibility directly to the grid.

Demonstrations reinforced a clear takeaway: data centers are moving from passive consumers to active, grid responsive participants—and flexibility will be key to integrating them at scale.

8) Energy storage is moving fast—but deployment is still constrained by bankability, grid integration, and safety

Energy storage discussions focused on emerging technologies, deployment lessons, and the planning and safety considerations required for successful integration.

Where the conversation got especially practical was in what’s holding deployment back: conservative financing environments, grid connection challenges, and the need to translate technology promise into compelling business cases and implementation readiness. Safety and reliability were treated as central—not optional add-ons.

9) AI is moving from hype to operational reality—and success depends on foundations, not flash

The most revealing insight from the AI discussions was not about new capabilities, but about restraint. Utilities are beginning to recognise that value from AI does not come from deploying the latest model, but from embedding intelligence into the way engineering and operational decisions are already made. The challenge is less about what AI can do, and more about what organisations are prepared to trust, govern, and sustain.

This is where domain‑specific approaches matter. AI that understands power‑system context—rather than generic automation—emerged as the differentiator, reinforcing why EPRI’s work on tools like Raphson and initiatives such as Open Power AI (OPAI) is focused on orchestration, transparency, and repeatability. The emphasis is on AI that augments expert judgement, leaves an auditable trail, and fits within existing planning and operational disciplines.

Underlying the discussion was a clear hierarchy of priorities: affordability over novelty, safety over speed, and integration over experimentation. Cybersecurity, data governance, and decision auditability are shaping what is considered acceptable AI, not as constraints but as enablers of trust. The insight was clear—AI’s next phase in the power sector will be defined not by ambition, but by discipline.

10) Collaboration is becoming essential infrastructure for reliability
One of the clearest signals from the discussions was that collaboration itself is now a reliability enabler. EPRI’s model reflects a growing recognition that no single organisation has full visibility of today’s power system challenges.

As grids become more interconnected, many of the toughest issues now sit at the interfaces: between TSOs and DSOs, planners and operators, utilities and large customers, and technology providers and regulators. In this environment, shared learning is not a “nice to have”—it is a practical way to surface operational reality, reduce duplicated effort, and shorten the path from insight to implementation.

Taken together, these ten insights reflect a sector that is aligned on the challenges ahead, but increasingly focused on how to deliver solutions in a system growing more complex by the day. European Workshop Week made clear that progress will depend on stronger coordination, clearer decision‑making, and closer collaboration across transmission and distribution. By translating shared understanding into practical, scalable action, the industry can continue to build an electricity system that remains safe, reliable, and affordable as Europe’s energy transition accelerates.