On-Site Generation: Electrical Integration

Chapter 3.5 made the strategy decision: for a gigawatt-class campus in 2026, behind-the-meter generation is the primary power strategy, not the diesel-in-the-yard afterthought, because the interconnection queue cannot deliver firm megawatts on an AI revenue clock. Chapter 4.9 handles the molecules — getting fuel to the prime movers. This chapter is the electrons in between: wiring a privately-owned power plant to a privately-owned supercomputer, where there is no utility on the far side of the bus to hide your mistakes. The grid is an infinite stiff source with seconds of spinning inertia and a fault-current capability measured in tens of kiloamps. An islanded gas plant is none of those things, and the load it must serve is the most electrically hostile in the industrial world.

Three decisions dominate everything downstream: island vs grid-parallel, where the transient absorber lives, and grid-forming vs grid-following. Each traces into its own protection, stability, and capital consequences. Underlying all of them: an AI cluster presents a high share of power-electronic load with topology- and operating-point-dependent harmonic current and workload-synchronized load steps, and a self-built plant has to be engineered around that fact from the synchronizing relay outward, not retrofitted after the first synchronized GPU pulse trips a generator.

Classical generator-and-bus design assumes a load that is large relative to no single source, statistically smooth, and forgiving of a few hundred milliseconds of frequency excursion. A training cluster violates all three. Tens of thousands of GPUs execute the same synchronous collective on the same clock, so their power transitions can align across the hall; use a named rack waveform’s idle floor, amplitude and duration instead of inferring a millisecond step from an OEM peak rating. To the plant this looks like a load that can swing hundreds of megawatts in a single step, repeatedly, with a sharp di/dt and a power-factor and harmonic signature set by thousands of switch-mode front ends. SemiAnalysis modeled exactly this — gigawatt-scale training load fluctuation as a genuine grid-blackout risk — and it is the reason xAI's Colossus first burned tens of millions of dollars a year running dummy workloads to flatten the profile before deploying over $375M of Tesla Megapacks to do the job properly (DCD / SemiAnalysis, 2025–26).

The consequence for the plant designer is that the prime movers cannot see this load raw. A reciprocating engine can ramp over 100%/minute and reach full load in roughly two minutes; an aeroderivative turbine ramps near 50%/minute with a five-to-eight-minute start (Grid Capacity Intelligence; Data Center Frontier, 2025). Both are geologically slow next to a millisecond GPU step. The online machine’s immediate electromagnetic/inertial response, subsequent governor/fuel ramp and offline starting are distinct. Place storage where the residual violates the qualified bus envelope; Chapter 4.5 sizes repetitive absorption alongside reserved outage duty. The plant's job is to present the engines a load they can actually follow, and to present the GPUs a bus stiff enough that they never know how soft the source behind it really is.

Sizing an islanded plant is not 'IT load plus a margin.' It is a stack of multipliers that compound: a stated IT-to-facility allowance for retained mechanical and electrical draw (1.4–1.5x is an illustrative planning range, not a universal island multiplier), then a redundancy overbuild (N+1 or N+2 at the generation tier), then each engine or turbine’s offered output at the site’s ambient temperature and altitude, including its hot-climate derate, then headroom for the block- and step-loading transient the machines can actually accept. Stack them and the real-world number is steep: Vantage's Shackelford campus provisions ~2.3 GW of generation for a ~1.4 GW IT load — about 64% overbuild — and a hot-climate N+1+1 aeroderivative plant needs on the order of 10–11 units to firm 200 MW (Power Engineering, 2025). Every multiplier is a fuel-supply, emissions-permit, and capital consequence, which is why the engine-vs-turbine choice in Chapter 4.9 cannot be made independently of this electrical sizing.

Block loading — the largest single load step a generator can accept without unacceptable frequency dip — is the governing transient spec, and it is where the prime-mover technology choice bites. A reciprocating engine accepts a far larger instantaneous block load (as a fraction of rating) than a turbine, which is one reason the RICE fleet dominates the speed-to-power buildout despite needing 200+ units per 500 MW. Combustion torque cannot track an arbitrary millisecond GPU edge, but the rotating machine’s electromagnetic and inertial response begins immediately; acceptable block loading depends on the offered machine and frequency/voltage limits. One 2026 design path is engines for bulk power, qualified BESS for the residual transient: the battery sources or sinks the fast di/dt in the first milliseconds-to-seconds, holding bus frequency while the engine governors catch up over the following seconds-to-minutes. NVIDIA's own production-BESS guidance frames the facility battery exactly this way — transient absorption, ride-through, and demand response as three distinct roles on one asset, with closed-loop state-of-charge control so the bridge is always charged enough to take the next pulse (NVIDIA, 2025).

Capacity and bridge energy pass; island acceptance still needs the dynamic test. Three survivors offer 3 × 9 = 27 MW against 23 MW, leaving 4 MW steady capacity. Define the initial deficit D = 23/4 MW, about 5.8 MW. The assumed bridge energy is [D × 0.020 s + ½ × D × 8 s]/3.6/0.95 = about 6.8 kWh; the 6 MW / 10 kWh reserve passes this power-and-energy screen. That result does not calculate frequency nadir, bus voltage or reactive-current headroom.

The linear-recovery crossover is [10 kWh × 3.6 × 0.95 − D × 0.020]/(½ × D), about 12 s; 12 s fails the unrounded bound. At 20 s recovery, the bridge needs about 17 kWh and fails even though three generators still have enough steady MW. Preserve the candidate only with a demonstrated response inside the allowed envelope; otherwise increase reachable reserve or shed enough load through an approved local control. The plant controller supplier owns loss-of-source detection, reference-source continuity, reactive/current limits, cooling ride-through and protection-state transition. Chapter 4.12 must witness the trip, record voltage/frequency/SoC and test resynchronization permissives before island operation is accepted. Chapter 4.9 must separately prove fuel delivery through the same event; 27 MW on a generator schedule cannot overcome a failed fuel train.

The four placements are a layered defense: a gigawatt design selects the layers needed to meet the measured residual envelope, each with a defined reachable load and power/energy duty. Assume one layer covers another's job and you pay for it — sizing the BESS for the genset's seconds-to-minutes energy while forgetting it must also hold frequency in the first cycle, or oversizing engines to chase a transient they physically cannot follow and stranding capex in machines that then idle inefficiently below their minimum-load sweet spot. Decompose by time constant instead: on-board capacitance shapes the sub-millisecond edge, the BESS owns milliseconds-to-minutes and the frequency, and the prime movers own the steady-state energy and the spinning reserve.

Firming 500 MW–2 GW from reciprocating engines means paralleling dozens to hundreds of 2–25 MW units onto a common bus — INNIO's 2.3 GW VoltaGrid order is 92 power packs at 25 MW each (INNIO, 2025). Each machine must be synchronized to the live bus before its breaker closes (matching voltage, frequency, phase angle, and phase sequence within the sync-check relay's window) and must then share the load proportionally through droop or isochronous load-sharing controls, so no single engine hunts or reverse-powers. At this unit count the synchronizing and load-sharing controller is the plant's nervous system, not a commodity panel. Its failure modes are catastrophic in a way they never are behind a stiff utility: a mis-synchronized close-in that slams a machine onto the bus out of phase, or a load-sharing loop that oscillates under the GPU pulse.

The bus-tie architecture is the reliability and maintainability fork. A single bus is cheapest and simplest but is a single point of failure and forces a full shutdown to service the switchgear. A double-ended (main-tie-main) bus splits generation across two sections with a normally-open or normally-closed tie, so a fault or maintenance event takes out half, not all — the workhorse for block-redundant AI campuses, and the natural electrical companion to the distributed/block-redundant UPS topology Chapter 4.5 favors. A ring bus gives every source two paths to the load and survives any single bus-section fault without dropping generation, at the price of more breakers and more complex protection coordination. The choice cascades directly into the relaying scheme: more bus sections and ties means more zones, more directional and differential relays, and a harder selectivity study — which is why the protection design (Chapter 4.2) cannot be finalized until the bus-tie topology is frozen.

An islanded inverter-and-engine microgrid breaks two assumptions that decades of protection practice rest on: abundant inertia and abundant fault current. A utility-fed bus has both — seconds of synchronous spinning mass damping frequency, and tens of kiloamps of fault current that make overcurrent relays trip crisply and selectively. Strip the utility away and replace half the source with inverters, and both collapse. Inverter sources are nearly inertia-less, so frequency can move fast under a step; and they are current-limited, with an illustrative ~1.2–1.5x rated range that must be replaced by the offered phase/sequence current-versus-time limits, so a fault on an inverter-dominated bus may not produce enough current to operate a conventional overcurrent relay at all. The protection philosophy must change: test minimum-fault sensitivity and selectivity in every source state, adding differential, directional or fast communication-assisted schemes where the study requires them (e.g. fast bus-differential, line-current-differential over fiber, and source-transfer logic), because you can no longer assume a fault announces itself with a 20x current surge.

For an islanded AI plant, a synchronous condenser, optionally fitted with a flywheel, is one design path: the synchronous machine adds physical inertia and fault current, reducing the burden on grid-forming controls and low-fault-current protection. ABB's orders for VoltaGrid total 27 synchronous condensers with flywheel booked through 2025 plus 35 more in a March 2026 extension — 62 machines whose entire purpose is to make a low-inertia, inverter-heavy island behave, electrically, more like the grid it replaced (ABB / DCD, 2025–26). The decision fork is study-driven: quantify minimum system strength, RoCoF/nadir, inrush, inverter overload and fault-detection selectivity, then choose a condenser-assisted or condenser-free architecture. A synchronous condenser buys mechanical inertia and higher fault current at added capital and losses; a condenser-free design buys fewer rotating assets at the cost of tighter grid-forming-control and low-fault-current-protection validation.

This is the fork that quietly determines whether a behind-the-meter plant can island at all. A grid-following (grid-feeding) inverter is a current source that needs an existing voltage and frequency reference to synchronize to — it follows the grid. Strip the grid away and a fleet of grid-following inverters has nothing to lock onto; they cannot establish a bus by themselves and cannot black-start. A grid-forming inverter is a voltage source: it imposes voltage and frequency on the bus, acts as the reference the other sources follow, may support black start when its energy source, auxiliaries, controls and energization duty are qualified, and can emulate inertia and damping within its power/energy limits. The consequence is binary — an island needs a voltage/frequency reference from a suitable synchronous generator or qualified grid-forming inverter; a BESS label alone does not establish that capability. A grid-parallel-only plant can get away with grid-following because the utility is the reference, which is exactly why the cheaper grid-following hardware proliferated first and why the migration to grid-forming is the defining 2026 inverter-control story.

The capability is now demonstrated at the scale that matters: transmission-connected grid-forming BESS runs at 200 MW and 275 kV today — Zenobē's Blackhillock Phase 1 in Scotland (300 MW/600 MWh total), the first transmission-connected battery selling inertia and short-circuit-level stability services to NESO (March 2025). True inverter-led black start is demonstrated at distribution scale (NESO's Distributed ReStart trials); a 275 kV BESS-led black start has not yet been performed. But grid-forming at gigawatt scale across multiple inverters and multiple sites introduces a failure mode the grid-following era never had — wide-area control interactions and oscillations between many fast voltage-source controllers, flagged in a 2025 preprint (arXiv 2508.16457) that models AI-workload-driven wide-area oscillation as a genuine multi-site stability risk. Grid-forming control is a tuned, coordinated control problem where the inverters, the synchronous condensers, and the engine governors must share the frequency-and-voltage job without fighting each other. Standards are converging on this — IEEE 2800 for interconnection of inverter-based resources and UL 1741/IEEE 1547 for the inverter functions — but the islanded-microgrid grid-forming case is still where vendor control firmware, not a published standard, carries the load.

Whether or not a campus runs permanently islanded, almost all of them want the option to interconnect — to import during planned maintenance, to export or provide grid services for revenue (Chapter 15.8), or simply to keep the utility as a backstop. That makes the point of interconnection (POI) a two-way electrical and regulatory boundary, and the transition across it the riskiest routine event in the plant's life. The electrical machinery is the synchronizing breaker and its check-sync relay at the POI, the anti-islanding / intertie protection the utility mandates (so the plant never energizes a de-energized utility line and endangers line workers), and the transfer scheme that decides open- vs closed- vs seamless transition. Each is a coordination problem with the utility's own relays and its dynamic model of the load — the same model submittals Chapter 4.3 governs — because to the utility, a self-generating gigawatt campus is simultaneously a large load, a potential generator, and a reliability variable it must study.

The regulatory ground under this moved decisively in late 2025. FERC's December 18, 2025 order directed PJM to create three co-location transmission services — Firm Contract Demand, Non-Firm Contract Demand, and an interim non-firm transmission service — the first national template for how a behind-the-meter plant legally relates to the grid it sits beside. PJM filed in February 2026; FERC's April 16, 2026 order accepted the filing in part (four distinct interconnection paths for co-located load) and rejected it in part (PJM's rewrite of the Co-Located Load definition), directing further compliance in May — the template is now a live, contested docket, not a future filing (the tariff mechanics live in Chapter 3.2). The electrical-integration consequence is that the transition scheme is no longer purely an engineering choice; the tariff class you take (firm vs non-firm grid backup) prescribes how seamlessly and how often you may lean on the utility, which feeds straight back into how much islanded inertia, storage, and black-start you must own outright. A plant that contracts only non-firm grid backup is, electrically, an island most of the time and must be engineered as one.

The electrical-integration design cannot be frozen against today's rack power, because the load it must serve is ramping 15–25x in four years — H100 ~40 kW, GB200 ~132 kW, GB300 ~142 kW, Vera Rubin ~190–230 kW, and a signposted ~600 kW–1 MW Kyber-class rack by 2027 (SemiAnalysis / NVIDIA). A denser rack does not merely draw more average power; it sharpens the transient — bigger steps, steeper di/dt, more energy per pulse — which pushes the absorber sizing and the grid-forming control bandwidth, not just the conductor ampacity. Controls tuned for a 132 kW-rack pulse spectrum may be marginal against a 600 kW-rack one across the BESS, the synchronous condensers, and the paralleling loop. Reserve the irreversible substrate the same way you reserve slab and water loop: BESS pad space and DC-bus headroom, synchronous-condenser foundations, switchgear lineup spares, and control-system bandwidth — none of it retrofittable into a live islanded plant without taking the cluster down. Size the firm-power tier to the contracted IT ramp; size the stability and transient tier to where the pulse spectrum is going, because that part breaks first and is hardest to add later.