Long-term Electrification Forecasting (EV, PV and Building Electrification)

The problem

Uniform Adoption Rates Miss Every Local Peak

A regional adoption curve spread evenly across every feeder describes no feeder in particular. Electrification arrives unevenly: by income, housing stock, building age, commute patterns, and zoning. Forecasting it well is a spatial problem.

Allocation

From Regional Outlook to Feeder Realization

Each technology's adoption scenario is allocated through GIS-informed relationships — customers, buildings, land use, demographics, and grid topology — producing a feeder-level realization of the regional outlook.

• Electric vehicles — adoption, charging level and location (residential, workplace, public), daily charging shape
• Solar PV — rooftop and behind-the-fence adoption, hourly output on the shared weather draw, net-load and reverse-flow effects
• Heating electrification — heat pump and electrified space-conditioning uptake, and the shift in weather sensitivity that moves peak size, shape, and season

No double counting

Separating What History Already Contains

Historical load already reflects the EVs, rooftop PV, and heat pumps that existed when it was measured. The platform recovers native load first by adjusting for known resources, then layers future adoption on top — so no technology is counted twice.

• Measured load — Layer 1
• Known modifier adjustment — Layer 2
• Future modifier scenario — Layer 3

Outputs

Modifier Layers You Can Inspect Separately

Each technology produces its own hourly, probabilistic modifier layer per feeder. Planners can view EV, PV, and heating impacts separately, combine them, and trace each back to the adoption scenario and allocation rule behind it.