Electrification headroom vs EV headroom

One ConEd feeder shows up twice in the snapshot I pulled, with two different capacity numbers, 1.16 MW and 1.67 MW. One estimate is from EV demand, the latter is from broader electrification.

Comparing those two numbers on how much extra demand a feeder can take depends on when that demand shows up and how it lines up with what's already on the circuit.

A depot charging trucks overnight and a building adding electric heat pumps put different loads on the same wire. ConEd's electrification map update specifically includes winter numbers for uses like heat pumps, which is exactly the kind of detail that gets lost if you only look at the top-line MW.

ConEd defines EV load capacity as the difference between a feeder's rated capacity and its peak from the past year. It’s a backward-looking observation. The electrification maps, added in January 2024 specifically to cover winter capacity for uses like heat pumps, are instead made from a three-year, 8,760-hour forecast run through ConEd's GE-MARS simulation model, adjusted for projected DER impacts. It’s a forward-looking projection. (GE-MARS is a production cost and reliability simulation tool.)

That’s why we have two different numbers for capacity, albeit the naming is not very intuitive. "How much room was left after last year happened" and "how much room do we project for a specific future season" produce different numbers on the same wire, especially if a feeder's true worst season is summer but the electrification figure is keyed to winter. The “electrification” number will look higher because it’s compared with the season that isn't the binding constraint.

A developer building an EV depot cares about the already-realized peak, because that's the number closest to describing today's risk. A developer electrifying a building's heating cares about the forecast winter number specifically, since that's the season their load will show up in, and a three-year-out DER-adjusted forecast is closer to describing the environment their project will operate in. If the utility only published the more conservative EV number, a building electrification project sizing itself might reject a site that has winter headroom that the EV figure was not meant for.

PSE&G's New Jersey EV map compares a circuit's reference maximum to a recent annual peak. Nothing in that formula specifies an EV charging schedule or shape. Calling it the “EV map” tells you who it's for. It doesn't specify in the map what math is underneath it, and that calculation turns out to assume nothing EV-specific. The specification of the calculation usually sits in a (metaphorically dusty) 300-page methodology PDF or explanatory text somewhere off to the side.

RG&E's data has substation records and no feeder records. Getting feeder-level detail there means finding another source entirely. Definitions, asset levels, and coverage all shape how a result should be used, and none of it can afford to get lost when pulling data from different utilities into one system. The source tags in the database exist to carry that context forward, so a developer can pick estimates that are comparable, notice where they still need more information, and go to the utility with a more detailed question than they would have otherwise.