V2G is often discussed as an important feature - but when is it valuable and how much could it be worth?
V2G (Vehicle-2-Grid) is often mentioned as an upcoming feature for charging infrastructure, but what is it and when is it really valuable? There is potential for savings or even profit with the right approach, but the technical complexity and battery life costs need to be factored in.
Vehicle-2-Grid (V2G) allows a parked EV to return energy to the utility grid when it is connected to the charger. The EV owner can then operate as an energy trader, buying cheap energy when there is excess (for instance on a windy or sunny day when renewables are producing more than needed), storing it, and selling it back to the electric utility when it is more expensive.
V2G is also beneficial to the utility as generating and distributing electricity at peak times can be very expensive. V2G can happen in a number of ways including by providing ‘grid-services’ which work over short time spans to help stabilize the electricity grid. V2G lets EV owners take advantage of underused battery and charger assets to earn additional revenue.
Making V2G work requires cooperation from many parties. A bi-directional charger is just the first step. The EV owner, the charger owner, the distribution utility, and an energy trader all need to communicate and coordinate automatically in real-time. The bi-directional charger needs to communicate with the EV to ask it to return power, usually via ISO15118. The charger needs to connect with central control, often via OCPP, to find out when to return power. The utility and charger central control need to communicate, with protocols like OpenADR.
The EV user also has a schedule to keep and needs to ensure vehicles are charged when they are needed, which requires integration with the scheduling or planning tools. This is actually easier for heavy vehicle fleets as the schedules are usually planned and regular, compared to a passenger car driver who may decide to go on a road trip at short notice. The energy trader needs to see the price is attractive, then send that message to the chargers, vehicles and utility. All of these standards are becoming more common, but V2G only works if all stakeholders are involved and with financial incentives for everyone.
A simple version of smart charging is planned charging, charging when the energy is cheap (like after 10pm when the rates get cheaper) and never returning energy to the grid. Strictly speaking, it isn't part of V2G because energy is never returned from the vehicle to the grid although it is sometimes included in V2G discussions. Depending on the tariff, planning charging when energy is cheap can generate considerable savings, requires less systems integration and has minimal to no pack life consequences as there are no more charge cycles.
A simple hypothetical base with 100 vehicles, each charging 300kWh per day, would consumes 380k€ worth of energy (1) in 2019. Simple planned charging, just charging between 10pm and 6am, would have saved around 50k€ in 2019 since overnight prices are cheaper than daytime. Charging at only the cheapest overnight hours (assuming they are predictable) could potentially save around a further 25k€ but would likely increase utility demand charges. For many fleets, especially early in electrification, the relative simplicity of planned charging, especially on a utility tariff with fixed rates and peak times, makes it a preferred option.
The basic idea of selling energy back to the grid at peak times sounds attractive, but how profitable is it really? The simplest way is by looking at hourly prices then buying low and selling high. With hypothetical perfect trading on Dutch wholesale energy prices, buying on every low and selling on every high could have earned ~14€/kWh/yr of storage over 2019. That would require ~1300 cycles, using up most of a typical pack cycle life, without a strong system of minimising the number of trades and care for pack life.
A simple improvement would be to limit to one trade a day between 6pm and 6am, since the rest of the time the fleet is busy working, resulting in only ~8€/kWh/yr . That means, for the same base of 100 buses with 300kWh batteries, a possible 240k€ annual revenue. Operational constraints will mean that not all of those trades can be done, not all vehicles will be able to cycle fully every night, and traders won’t have the prescient knowledge that comes with analyzing historical data.
The above example uses very simple trading, buy low, sell high, but usually ‘grid-services’ are a more profitable way to apply V2G. The utility grid needs to be constantly kept in balance between production and consumption. Energy storage helps stabilize the system over short periods, especially if there are swings like a gust of wind increasing power at a windmill.
The grid operator pays for services which can act as storage, and quickly (<30sec) provide or consume energy, or equivalently increase or decrease power consumption. Most charger and batteries could likely handle the fast response times at the hardware level, provided the software is properly designed for it. This requires not only communication between all the parties but quick updates. Grid-Services usually encompasses a range of grid stabilization products with defined technical and legal requirements, from Frequency Containment Reserve (FCR) which measures the operating frequency of the grid and responds within seconds to manual Frequency Restoration Reserve (mFRR) (2) , responding in minutes to a signal from the utility.
The prices for a specific grid-services product, (automatic Frequency Restoration Reserve, aFRR in this example) often spike much higher for brief periods when grid balancing is needed and therefor can earn more than simple energy trading. aFRR capable storage capacity used overnight (10pm - 6am) in 2019 could have generated as much as 77€/kWh/yr, or ~€2.3M/yr for the base described above. (revenue, not profit) There are several simplifications in this calculation, including assuming the site always has the utility capacity to meet the grid services demand and charge for the next day. In practice, due to power limitations and restricting the battery depth of discharge to preserve cycle life this would likely be much less.
The grid services markets are volatile, so there will also be year to year variation in revenue. Like other energy prices, the price is usually highest in peak periods and lowest at night, which is also when buses are busiest. (For comparison an analysis looking at cars in Denmark found grid services could earn 700€/car/year with industrial tariffs, but not residential.)
However, there are complexities to trading in the grid services markets. These services are not sold to retail by the kWh, they usually have minimum bid sizes of 1MW (varies by product and market) and minimum availability times (e.g. 4h). This means a minimum capacity of 4MWh may be required to bid. There is a conceptually simple (though complex to implement) solution in aggregation. A trading company bundles together many small producers, including V2G cars or other sources, and sells them in aggregate. If one asset, for example a bus, isn’t available or fails, then another can take its place. Provided there are enough assets and the statistics are well understood this can provide robust service.
Batteries have a limited lifetime, they can’t be cycled endlessly and preserve their capacity. This is a big risk area for V2G. Using an EV battery for V2G reduces cycle life from the main application, driving. The right control software can help take advantage of this. Beyond all the initial costs for communication and bi-directional chargers, battery lifetime is compromised every time it is cycled. At a first approximation, $200/kWh for batteries (a low estimate, especially for packaged, lower production volume heavy duty vehicles) and a typical 2000 cycles of usable life, each 1kWh cycle costs $0.10.
Fleet vehicles work far harder than personal cars in annual mileage and hours and so are more likely to need batteries replaced The cycle life degradation of batteries is a complex subject, and slow, multiple partial cycles in the middle of the charge range will result in much less lifetime degradation than the same energy in fast full cycles. For V2G to be financially interesting, the savings need to outweigh the battery lifetime costs.In addition, charging isn’t perfectly efficient, selling 1kWh of energy back to the utility may need 1.1kWh to recharge afterwards. There are very few times in a year when the price difference is enough to cover the cost of the battery life degradation.
In some ways, fleets are a good choice for V2G as there are many vehicles on a single operator with quite predictable schedules and large batteries. The downside is most fleets work hard, and therefor spend much less time connected to the charger and the grid compared to personal cars which are parked a substantial part of the time. With less time connected to chargers, less time is available for energy trading. In many cases the trades require both times to sell the energy and times to buy it. But the trade won't work if the vehicle is out doing its primary function at either time.
Think of the hours the fleet is parked like the hours your stockbroker is open; it is possible to make money with a broker only open on Tuesdays and Thursdays, but the broker that is open every day will have the bigger advantage. In any calculation for V2G, remember that much of the time the vehicle is driving, or is doing charging it requires for driving, leaving fewer hours for flexible charging or energy trading.
Like a stock price, it is hard to predict. There is downward price pressure as more V2G, demand-response, and energy storage projects start entering markets. Some can be very simple, such as keeping commercial freezers cooler so their cooling can be shut off for brief periods and the energy sold for frequency control. These could cost less to operate than V2G, especially considering battery life. There is upward pressure as more variable renewable energy generation like wind and solar is installed and the needs for grid services increase. Commodity markets usually stabilize around the price of the lowest marginal cost producer, so if easier to implement and cheaper to operate grid services solutions than V2G become common it could undercut the business case.
Ultimately, the energy markets are a trading game. The profit that can be made depends on specific technical capabilities, the markets they are connected to and the skill of the traders. For most fleet operators like transit agencies, energy trading isn’t a core competence and may be better contracted out to a specialist company. Setting the correct rate, and contractual rules around operational flexibility and battery lifetime however will be quite complex.
This is not intended as financial advice and ChargeSim accepts no liability for actions taken based on it. The energy markets and technical constraints for V2G implementations are complex subjects and specific analysis by competent experts for a specific location should be done before committing any project.
(1) Average NL 2019 wholesale prices (energy price data source: Entsoe) calculated on 6 days per week. Distribution and demand charges from distribution operators are not included.
(2) These are European terms, used by ENTSOE. Transmission operators in the US often have similar types of services, but may have different names and specification details.
This is not intended as financial or technical advice and ChargeSim accepts no liability for actions taken based on it. Always consult a professional about your specific situation.