APR Energy

05/06/2026

Many people in the data center industry are used to software that can largely run itself once it’s deployed.

Power generation is hardware.

Large-scale generation systems run continuously and require operators, technicians, and maintenance teams to keep them performing as expected.

Automation and monitoring tools help, but reliable electricity still depends on the people responsible for operating and maintaining the equipment.

Power systems don’t just run themselves. They run because skilled teams are there to operate them.

04/28/2026

Power permitting for on-site generation is one of the variables that most affects data center project timelines and varies state to state, even city to city.

A few things worth knowing early:

Permitting requirements vary significantly by state and jurisdiction. Some utilities are the sole provider of generation, transmission, and distribution services within their jurisdiction. Others allow private ownership with varying incentive structures. What's straightforward in Texas may not be in Virginia.

Air quality permitting is increasingly consequential. As on-site generation scales beyond backup use into primary power, different thresholds apply. Starting air quality modeling during site due diligence, not after, changes what's possible in the design phase.

Fuel sources affect permitting timelines and complexity. Natural gas is generally faster to permit than diesel and carries lower emissions exposure. But gas pipeline access is its own feasibility question that needs to be assessed early.

Permitting is one of the earlier variables that shapes what's realistically buildable, and one of the harder ones to recover from if it surfaces late.

04/23/2026

Redundancy in a data center power system is often treated as a design checkbox: N+1 or even N+2 at the generator level, dual feeds, and UPS coverage.

But redundancy and availability are not the same thing. A system can be fully redundant on paper and still fall short of the availability an AI workload requires — if maintenance windows aren't coordinated, if load ramp hasn't been modeled across the full campus, if generation sources aren't synchronized under real operating conditions.

The gap between the two shows up during operations, not during commissioning.

04/22/2026

At Data Center World in Washington, D.C. this week, one theme keeps coming up: as grid connection timelines stretch further out, more operators are looking at on-prem power to keep builds on schedule.

It’s a challenge APR knows well.

Thanks to the team at Data Center World & AFCOM for a productive conference. We’re already looking forward to next year.

04/16/2026

S&P Global’s recent analysis on U.S. data center energy demand makes something clear: on-site power generation decisions used to be driven primarily by what a data center required. Now availability is shaping them just as much.

Large frame turbines, aeroderivative turbines, reciprocating engines — each has different lead times and different implications for how a power system gets designed. On-site power is increasingly favoring smaller, modular generation for near-term projects, while longer-term campus buildouts typically require a different configuration entirely.

Link for more details:

Discover how surging data center energy demand is outpacing US grid capacity. Read our expert analysis on navigating the data center power crunch through 2030.

04/14/2026

The power requirements for AI data centers are pushing the industry into generation requirements that exceed that of typical utility projects.

As campuses move toward hundreds of megawatts, and in some cases gigawatt-scale planning, the conversation is shifting from simply securing megawatts, to how power systems operate together once the campus is running.

Multiple generation sources, redundancy layers, electrical distribution systems, and controls all have to coordinate continuously as load ramps across the site.

At that point, the challenge isn't just individual components. It's how the entire system performs as demand evolves.

04/09/2026

Data Center World is less than two weeks away. APR Energy will be in Washington, D.C., April 20-23. David Dorman, Libby Owen, and David Wong will be there discussing how data centers are navigating power constraints, phased buildouts, and on-site generation.

Send us a message, or request a meeting: https://www.imillerpr.com/landing-page-apr-energy-request-to-meet-with-apr-energy-at-these-upcoming-events/

Data Center World & AFCOM

04/07/2026

Most data center power conversations start with the same question: "Where will the electricity come from?"

But as projects move from concept to ex*****on, the conversation shifts. Who installs the infrastructure? Who operates it over time? How does it evolve as the campus grows?

Large campuses rarely stay static. Capacity expands, load profiles change, utility timelines shift. Some projects rely primarily on grid supply. Others bring in on-site generation to supplement or accelerate availability. In some cases, the developer owns and operates the assets. In others, a long-term partner carries that responsibility.

The question isn't only where electricity comes from. It's how the entire power system is delivered, operated, and adapted over the life of the campus.

04/02/2026

As demand for on-site data center power grows, the pool of providers is expanding — and while many may appear similar on the surface, the depth of experience behind each one varies significantly. Established companies from adjacent industries and newer players with limited track records are competing for the same projects.

Standing up and operating a complete power plant to utility-grade standards is a different discipline than powering a drilling operation. Both are operationally demanding, but the requirements for continuous performance, safety, and long-term accountability don't look the same.

That distinction matters when choosing who operates your power.

03/31/2026

A recent Reuters article explores a shift underway in how data centers relate to the grid. For most of the industry's history, that relationship was straightforward: customer and supplier. It’s starting to change.

PJM Interconnection, the largest regional electric grid in the U.S., is projecting supply shortages as early as next year. In turn, utilities and regulators are increasingly urging data centers to take part in demand response, which is to reduce consumption during peak periods or switch to on-site generation to free up grid capacity.

Historically, data centers have not participated in demand response. The flexibility wasn't there. AI data centers may change that — energy-intensive LLM training workloads may be shiftable across locations, opening the door to demand response in ways traditional cloud facilities couldn't support.

Technology companies are beginning to commit to this shift. Google has already signed contracts with utilities to lower consumption when called upon.

For facilities with on-site generation, how this relationship evolves, operationally and contractually, is still being worked out.

Link:

The U.S. technology industry is being pushed to shrink its power use in times of high demand, amid growing public concern ​that Big Tech's massive electricity needs for its expansion of data centers are maxing out the country's grid.