Stranded power assets: Identifying and evaluating opportunities for data center development

The rapid growth of AI and hyperscale data centers is creating unprecedented demand for reliable, scalable electricity.

Across many markets, one of the biggest constraints is not total generation capacity, but the mismatch between where power infrastructure already exists and where new large loads are trying to connect. That mismatch is creating a new opportunity: stranded power assets.

A stranded power asset is a generation resource or infrastructure position that retains useful physical capability but cannot fully capture its economic value under current market conditions. These assets may include underutilized natural gas plants, nuclear plants under economic pressure, renewable projects facing curtailment from transmission congestion, generation located behind bottlenecks, or partially developed projects with valuable interconnection positions. In the past, these sites may have been viewed primarily as retirement, restructuring, or redevelopment candidates. Today, they may represent strategic platforms for data center growth. 

Why stranded assets matter for data centers

Large data centers are changing the way the market thinks about power and location. Historically, generation was built to serve load through the grid. Increasingly, large-load developers are reversing that model by locating data center campuses where power and infrastructure already exist. This shift is important because transmission expansion is slow, interconnection queues are congested, and permitting for new infrastructure can add significant time, cost, and uncertainty.

An existing power site may already include a high-voltage interconnection, substation equipment, transmission access, land, water, fuel supply, environmental permits, and an established industrial footprint. In many cases, the generation asset itself is only part of the value. The larger opportunity lies in the combination of infrastructure readiness, speed-to-power, and the potential to support phased data center development. 

Why identification requires more than a plant list

Not every underutilized or retired asset is a viable data center opportunity. Nameplate capacity alone does not determine whether a site can support a large new load. The key question is not simply how many megawatts are installed, but how many megawatts are dependable, deliverable, expandable, and economically competitive.

To answer that question, stranded asset analysis must consider:

• Generation capability and equipment condition,
• Transmission deliverability under normal and contingency conditions,
• Existing interconnection rights,
• Fuel availability and infrastructure,
• Environmental and permitting restrictions,
• Ability to meet hyperscale reliability requirements,
• Market economics,
• And potential for expansion through renewables, storage, or additional generation.

This requires an integrated technical and economic approach.

How we use modeling to identify stranded assets

Consulting Services uses proprietary nodal modeling and grid analytics to identify stranded asset opportunities systematically rather than anecdotally. Our methodology is based on high-fidelity “Country Models,” which act as digital twins of the power system by integrating transmission topology, generation fleets, fuel infrastructure, and market structure into a single analytical framework. These models operate at the nodal level, allowing us to isolate highly localized congestion, deliverability constraints, and infrastructure advantages that may not be visible in zonal or hub-level analysis.

This nodal granularity is critical because stranded asset opportunities are inherently local. A site may appear weak from a broad market perspective but prove highly strategic once analyzed at the node, where interconnection position, transmission headroom, basis differentials, and surrounding infrastructure can materially change its value.

1. Nodal market screening

We begin by using production cost modeling to forecast long-term nodal Locational Marginal Pricing (LMP) and compare nodal outcomes against regional hubs. This helps identify sites with favorable price exposure, lower congestion, and stronger long-term operating economics. By screening at the nodal level, we can distinguish between assets that are structurally disadvantaged and assets that are simply underutilized under current market conditions but may become highly valuable if paired with large new load.

2. Transmission headroom and deliverability analysis

Once candidate nodes are identified, we evaluate transmission headroom and incremental loadability using methods aligned with interconnection practice, including first-contingency transfer capability concepts. This enables us to determine how much new load could realistically be supported at or near a site without violating thermal limits, voltage criteria, or N-1 reliability requirements.

This step is essential for identifying the difference between installed capacity and truly deliverable capacity. It is also one of the most important filters in determining whether an asset can offer meaningful speed-to-power for a large data center campus.

3. Fuel infrastructure and thermal asset screening

For thermal assets, we overlay pipeline infrastructure, supply availability, utilization rates, import dependencies, and basis differentials relative to benchmark hubs. This allows us to identify which assets have the fuel security required for long-term dependable operation. It also helps distinguish between stranded thermal assets that are commercially attractive and those whose value is limited by gas deliverability or infrastructure congestion.

4. Renewable and hybrid resource potential

Our methodology also screens for nearby renewable resource quality and integration potential. This is particularly important because many stranded asset opportunities are most attractive when evaluated as hybrid configurations that combine existing dispatchable generation with renewables, storage, or phased additions. This enables a more flexible strategy for meeting both uptime and sustainability goals.

5. Reliability and grid impact assessment

After screening, we apply contingency-based reliability analysis and high-fidelity power system studies to assess whether a candidate site can support new hyperscale load under realistic system conditions. This includes evaluating thermal constraints, voltage performance, system strength, short-circuit behavior, and dynamic or transient stability where necessary. Because a hyperscale load is a system-wide event, this work is essential to determine whether the asset can be repurposed without creating unacceptable reliability concerns.

6. Economic and strategic down-selection

Finally, we integrate all analytical layers into a prioritized shortlist of stranded assets based on criteria such as power price outlook, transmission headroom, fuel security, infrastructure readiness, speed-to-power, and expansion potential. This allows clients to focus on the assets most likely to support cost-effective, reliable, and scalable data center development. 

How Consulting Services supports evaluation and development

Once opportunities are identified, we support the next stage of decision-making by evaluating how each site could be developed. This includes designing grid-connected, hybrid, or islanded architectures; optimizing the mix of thermal generation, renewables, and storage; and applying capacity expansion and stochastic resource adequacy analysis to balance uptime, cost, and carbon goals.

We also support long-term market analysis by forecasting nodal prices, modeling congestion and volatility, stress-testing future scenarios, and evaluating the optionality of market participation. In parallel, we conduct detailed grid studies using industry-standard tools to assess transmission impacts, voltage recovery, oscillatory behavior, transient stability, and system strength. This provides a rigorous basis for investment decisions and interconnection strategy.

Beyond technical analysis, Consulting Services supports regulatory and stakeholder engagement by providing the analytical evidence needed to show how a project affects the grid, local ratepayers, and broader community outcomes. This helps position stranded asset redevelopment not only as a private infrastructure solution, but as a project that can improve overall system value.

Final thoughts

Stranded assets represent more than underused generation. They represent a new category of infrastructure opportunity at the intersection of power markets, grid physics, and digital infrastructure development. But identifying the right opportunities requires more than a database of existing plants. It requires a nodal, system-level understanding of where physical infrastructure, market economics, fuel supply, and transmission capability align.

By combining proprietary nodal modeling, grid impact studies, market forecasting, and asset optimization, Consulting Services can help clients identify which stranded assets are truly actionable and how those assets can be transformed into reliable, scalable platforms for data center growth.