PJM Capacity Prices and BESS Asset Risk: What Operators Need to Know
PJM capacity prices have increased over 800% in recent auction cycles. The commercial upside for BESS operators has never been higher — and neither has the financial downside of physical asset failure during a Performance Assessment Interval. The ITC Section 48E preservation and PJM's capacity price environment have created the most attractive moment in the history of US battery storage investment. The physical risk layer beneath the commercial opportunity is not optional to understand — it's the difference between capturing the value and watching it disappear during a PAI.
What is the PJM capacity market and why does it matter for BESS?
PJM's Reliability Pricing Model (RPM) is a forward capacity market that procures commitments from generators to be available during periods of peak demand. Resources that clear the auction receive a capacity payment — a fixed monthly payment per MW of committed capacity for the delivery year. For BESS resources, participation requires demonstrating that the battery can deliver its committed MW for a sustained duration during a Performance Assessment Interval (PAI). PAIs are called during extreme grid stress events — when the value of reliable capacity is highest. The 800% increase in PJM capacity prices reflects a structural supply shortage: retirements of coal and gas generation, interconnection constraints on new renewables, and increasing peak demand. For BESS operators who cleared recent auctions, the capacity revenue is genuinely transformative — a 100 MWh BESS system at recent clearing prices can generate $10M–$12M annually in capacity revenue.
The Risk Equation: When capacity is priced at premium levels, the cost of non-performance during a PAI is calculated against that same premium price. A battery that can't deliver its committed capacity during a summer heat event isn't just failing operationally — it's triggering a penalty obligation priced at the clearing price it was paid to be available at.
How do PJM Capacity Performance non-performance charges work?
PJM's Capacity Performance (CP) construct imposes significant financial consequences for resources that fail to perform during PAIs. Non-performance charges are calculated at 1.5 times the net Cost of New Entry (CONE) for the relevant delivery year. At recent price levels, non-performance charges for a 100 MW BESS resource failing to deliver during a multi-day PAI can reach $5M–$15M — in addition to forfeiting the capacity revenue for the performance period. The financial exposure of BESS non-performance in the current PJM environment is not an operational risk. It is a balance sheet risk.
How does State-of-Health degradation create non-performance risk?
Every BESS capacity commitment in PJM is made against a nameplate capacity figure. Your battery degrades from day one. The nameplate doesn't change. State-of-Health (SoH) is the measure of a battery's actual available capacity relative to original nameplate. A battery system at 88% SoH can deliver 88 MWh of a 100 MWh nameplate commitment. If dispatched against a 100 MWh capacity commitment during a PAI, it has a 12 MWh shortfall. Under the CP framework, that shortfall is a non-performance event — priced at 1.5× CONE against every undelivered megawatt-hour.
Why is OEM SoH reporting insufficient for capacity commitments?
Most BESS operators rely on the OEM Battery Management System (BMS) for State-of-Health data. The BMS reports aggregate SoH — a single number representing the average capacity of all cells. This hides cell-level variance. A battery system with an OEM-reported aggregate SoH of 92% may have individual rack groups at 85% SoH and others at 96% SoH. The 85% rack group is the constraint — it limits actual deliverable capacity below what the aggregate suggests. In a dispatch scenario, the constraint determines performance, not the average.
Aggregate vs. cell-level SoH: what a PAI dispatch really delivers
| Reporting layer | What it tells you | What it hides | PAI dispatch outcome |
|---|---|---|---|
| OEM BMS aggregate SoH (92%) | 92 MWh notionally available | Worst rack at 85% SoH | Looks compliant on paper |
| Cell-level SoH (constraint = 85%) | True deliverable ~85 MWh | Nothing — it's the binding constraint | 15 MWh shortfall vs. 100 MWh commitment |
| Non-performance exposure | Charge = 1.5× CONE × shortfall × hours | — | Multi-million-dollar penalty on a single PAI |
What is thermal runaway risk and how does it interact with capacity contracts?
For BESS operators with PJM capacity contracts, a thermal incident during the delivery year creates compounding exposure across five distinct dimensions: non-performance charges for any PAI events while the asset is unavailable; replacement cost for a utility-scale system in the $10M–$50M+ range depending on size and chemistry; insurance claim complications if operating condition documentation is insufficient; OEM warranty challenges if the operator cannot demonstrate proper operating conditions were maintained; and project finance covenant review for project-financed assets. The physics of thermal runaway are well-documented. The precursor signatures — cell-level temperature asymmetry, capacity fade rate acceleration, internal resistance elevation — are detectable 48–72 hours before a thermal event becomes uncontrollable. Physics-layer diagnostics provide an intervention window that BMS threshold monitoring cannot.
800% capacity price increase. 72-hour thermal runaway precursor detection window. 1.5× CONE multiplier on every undelivered megawatt-hour. These are the three numbers that define whether a BESS portfolio captures the current PJM environment — or pays for the privilege of being committed to it.
How do you build a physical intelligence layer for a BESS portfolio?
The operators capturing the full value of PJM capacity revenue have built a physical intelligence layer beneath their commercial strategy. It is the difference between a capacity commitment that holds up under audit and one that unwinds under stress.
- 1.Cell-level SoH monitoring — real-time tracking of actual deliverable capacity at the cell and rack level, not aggregate OEM reporting. This is the foundation for accurate capacity commitment and dispatch planning.
- 2.Thermal precursor detection — physics-based modeling of cell temperature dynamics, capacity fade trajectories, and internal resistance patterns, with alerting triggered by precursor signatures, not threshold violations.
- 3.OEM warranty documentation — an immutable operating condition ledger that records cell temperatures, charge/discharge rates, and thermal conditions against OEM-specified limits, making warranty claims defensible.
- 4.Capacity delivery verification — pre-dispatch confirmation of available capacity against current SoH, so PAI performance commitments are made against real physics, not nameplate assumptions.
Frequently Asked Questions
- Why have PJM capacity prices increased over 800%?
- PJM capacity clearing prices have risen sharply because of a structural supply shortage in the PJM footprint: accelerated retirements of coal and gas generation, queue and interconnection constraints on new renewables and storage, and steadily rising peak demand. The result is that the Reliability Pricing Model (RPM) is clearing at price levels that make a 100 MWh BESS system economically transformative — and make non-performance equally consequential.
- What is a PJM Performance Assessment Interval (PAI) and how is non-performance penalized?
- A Performance Assessment Interval is a period of extreme grid stress during which PJM measures whether each capacity resource delivered against its committed MW. Under the Capacity Performance construct, non-performance is charged at 1.5× the net Cost of New Entry (CONE) for the delivery year. For a 100 MW BESS that fails to deliver during a multi-day PAI, total exposure can reach $5M–$15M, on top of forfeited capacity revenue.
- Why is OEM Battery Management System SoH reporting insufficient for PJM capacity commitments?
- OEM BMS systems typically report aggregate State-of-Health — a single average across all cells. This hides cell-level variance: a system with a 92% aggregate SoH may contain rack groups at 85% SoH that become the binding constraint during dispatch. PJM dispatches against committed MW, not averages, so the worst-performing rack determines whether the asset is performing or non-performing.
- How early can thermal runaway in a BESS actually be detected?
- The thermodynamic precursor signatures of thermal runaway — cell-level temperature asymmetry, capacity fade rate acceleration, and internal resistance elevation — are detectable 48 to 72 hours before a thermal event becomes uncontrollable, provided diagnostics operate on a physics model rather than fixed BMS thresholds. That window is enough time to derate, isolate, or schedule intervention without losing PAI performance.
- What is the Ellume Battery 360 + Vector physical intelligence layer?
- Ellume Battery 360 is the operations and warranty-grade monitoring platform for utility-scale BESS, and Ellume Vector is the physics-aware diagnostic layer that runs first-principles models of cell, rack, and system behavior against live telemetry. Together they provide cell-level SoH, thermal precursor detection, an immutable OEM warranty ledger, and pre-dispatch capacity delivery verification — the four capabilities required to defend a PJM capacity contract.
