EV charging is modelled explicitly in every European country, affecting both the demand shape and the price spreads that drive battery revenues.
Electricity demand from Electric Vehicles adds a price-sensitive load on top of base demand
Annual EV electricity demand is calculated in each European country covered by the model. Each country’s EV stock is projected using current battery electric vehicle (BEV) stock, population, vehicles per capita, a 17-year vehicle lifetime, and a sales-share profile that determines how quickly new sales shift to electric.
Stock is then converted to annual electricity demand using a fleet efficiency curve, a fixed annual mileage of 12,000 km per vehicle, and grid-side efficiency assumptions. The annual total is distributed across days using temperature data, so colder periods carry more driving.
For continental Europe, trajectories are cross-checked against TYNDP 2024. For GB, the ground-up trajectory is anchored to FES 2025 Holistic Transition, which puts 1.39 million BEVs on the road in 2024.
Four charging modes determine how EV demand is shaped
EV demand is modelled across four charging modes, each with a different profile and degree of price responsiveness:
- Non-managed - follows a fixed half-hourly charging profile; no price response
- User-managed - follows a user-set schedule (e.g. overnight timer); fixed profile, no optimisation
- Externally managed (smart charging) - charge is optimised by the model against wholesale prices, subject to constraints
- Vehicle-to-grid (V2G) - can both charge and export, also optimised against wholesale prices
Non-managed and user-managed EV demand adds directly to the base demand before the model runs. Cars which have externally managed smart charging and V2G are included in the market model’s optimiser as flexible assets, similar to utility scale BESS.
In 2026, non-price responsive EV demand makes up 62% of the total demand from EVs. By 2050, this is down to 56% - with 18% of vehicles doing vehicle-to-grid.
Smart charging and V2G are constrained to reflect real-world limits
Annual EV demand is removed from the base demand forecast before the 15-minute profile is shaped. It is then added back according to the mode split above. This prevents double-counting: without the adjustment, EV demand would already be embedded in the historical demand shape used to calibrate the model.
The optimiser dispatches smart and V2G EVs subject to the following constraints:
- Charger power: 4 kW per vehicle (home charger assumption)
- Battery size: 70 kWh average
- Round-trip efficiency: 88%
- Plug-in availability: a fleet scalar of 0.66 accounts for the share of vehicles connected at any time
- Minimum daily recharge: 90% of each day’s SoC depletion must be recharged within a rolling 24-hour window, with at least 70% recharged overnight between 23:00 and 06:00
- V2G export cap: exports are capped at 10% of total charge delivered, with a cycling cost applied
These constraints mean EVs cannot fully arbitrage the price spread. They must cover their driving needs first.
EVs reduce BESS revenue by around 1%, but the effect varies by region
Overnight smart charging lifts off-peak prices, shallowing the daily spread. Daytime solar troughs become slightly less deep, as a small share of the fleet charges from cheap solar throughout the day. Winter demand is higher than summer as drivers take more trips.
The net revenue impact is around -1% in our central case. However, smart charging also reduces the size of the utility-scale BESS fleet built by the capacity expansion model. In regions where the BM or ancillary services are not accessible to EVs, less BESS competition can offset or reverse the spread compression. In some areas revenues increase.