Three key actions to unblock EV flexibility for the grid.
Policymakers and regulators can avoid this disastrous outcome. ev.energy recognizes the importance of accurate and reliable real-time measurements from an EV Virtual Power Plant (VPP), which are necessary if this technology is to become a trusted grid asset at the gigawatt scale. However, EV flexibility participates in power markets not as individual assets but in aggregate, as part of a VPP, so it should be treated as a group. Market design and regulations for EV assets should align as closely as possible with the rules for larger assets, and data requirements should also reflect this aggregation. In collaboration with National Grid ESO, ev.energy ran live trials with EV drivers in the UK using a range of charging hardware to show how aggregated asset metering can help address the requirements of grid services, such as accuracy error, refresh intervals, and response times. This report evaluates the metering capabilities of various EV charging technologies relative to energy markets' requirements, highlighting the urgent need for change and convergence. Today fewer than 4% of EVs in the US and 12% in the UK are enrolled in active managed charging programs. By 2030 there will be 13 million battery EVs in Great Britain, 34 million in the United States, and 37 million across Europe. If all EVs on British roads in 2030 were charged at one time, the 90GW surge in demand would double the peak demand today. But with the right reforms, deployed without delay, we can ensure a safer, more reliable, cleaner, and more affordable power system.
EV flexibility is essential alongside grid-scale batteries.
The shift to electrified transport is happening at the same time as the decarbonization of our electricity system. These two trends can be complementary thanks to the flexibility of EV charging. Managed EV charging, which shifts EV charging to optimal times, can help power system operators ensure a reliable power supply, integrate more renewable energy, reduce EV drivers' costs, and cut total system costs for all grid users. Additionally, the grid has different flexibility needs, which are exacerbated by variable renewable generation such as wind and solar power. Already we have seen a change in our power systems, with the proportion of wind and solar generation growing from limited volumes to 38% of electricity generation in Great Britain, 22% in the EU, 14% in the US, and 11% globally. These values must continue to grow to meet decarbonization targets. Over time, it is becoming increasingly apparent that exploiting EV charging flexibility is not a nice to have, but an essential part of the grid operations toolkit alongside grid-scale batteries.
The grid's flexibility needs and how EVs can meet them.
Table 1 below shows the grid's various flexibility needs, which are growing primarily due to more wind and solar generation coming online, and how EV flexibility can help meet this need.
Aligning EV infrastructure with energy market requirements.
Unlocking EV charging's flexible potential depends on aligning EV infrastructure capabilities with energy market requirements and reaching wide-scale managed charging adoption. This is especially important for home charging, which accounts for about 80% of all energy delivered and will be the main source of flexible charging. Public charging offers less flexible capacity, as drivers generally want as much charge as quickly as possible from these charge points. Today, EVs are already responding to grid requirements through managed charging programs offered by a growing number of utilities and software providers. The five levels of managed charging program shown below set the maturity ladder.
Figure 2 · The five levels of managed EV charging
From unmanaged charging to bidirectional exporting.
Active managed charging signals can be implicit (saving money by shifting load to cheaper times of a time-varying tariff, e.g. a time-of-use tariff) or explicit (earning money by participating in demand response programs such as the Emergency Load Reduction Program in California or the Demand Flexibility Service in Great Britain). Source: ev.energy, The 5 Stages of Managed Charging.
Avoided cost
Stage 2 · Smart Active (V1G) · per enrolled EV per year
Up to $575
Avoided across generation, transmission, distribution, energy, ancillaries and customer ops · ev.energy's Cost-Avoidance StackEvery session, dispatched against your grid signals.
This is where most utilities run their flagship managed charging program. Real-time charging control across enrolled vehicles, recalculated every 30 minutes against grid, price, and carbon signals. Dispatchable load when you need it, with ~80% compliance, no snap-back rebound, settlement-grade off-peak compliance per session. The platform answer the regulator wants to see.
Powered by Eve Programs + Eve Sync core + Eve Ops core
V2X holds additional potential for grid operators.
To date, flexibility services have focused on turning charging imports up or down; this is also known as V1G. However, bidirectional charging, import and export from the vehicle's battery, holds additional potential for grid operators. For more information, see the ev.energy whitepaper 'Unlocking barriers to vehicle-to-everything (V2X) energy flexibility'.
Boundary meters alone won't cut it. Asset meters complete the picture.
When pooling EV assets together, we must think about how the charging is controlled and measured. The boundary meter is the main electricity meter in the household and is a good start to influence flexible charging patterns, but using only the boundary meter has some drawbacks. Asset meters, also known as sub-meters or dedicated metering devices, allow for more accurate reporting and forecasting directly from the EV at more frequent intervals. As they work alongside smart and traditional boundary meters, a lack of advanced metering infrastructure is no longer a problem, and the boundary meter can remain as the primary meter for billing household energy usage. Ultimately, asset metering can enable grid services, rebate programs, fleet reimbursement, and other services, without having to install additional infrastructure or change any billing processes.
The EV asset can be metered at the vehicle or the charge point.
Ideally, a VPP operator would integrate with both, though only one integration is needed for most grid services. The fragmented nature of the energy and automotive industries, and manufacturers' limited exposure to the power sector, means that meters built into EV assets have often not been designed to meet power system metering standards. That doesn't mean current meters can't meet these standards, it's just not explicit. Even if the manufacturer wanted to design to a standard, it would be difficult because of the variety of standards across power markets and regions, and the lack of clear value. Manufacturers sometimes limit or charge fees for access to data from the EV asset. This is an understandable commercial practice but may reduce service provision to the power system and value for consumers. In recent years, governments have taken steps to improve interoperability and dismantle some closed systems, for example by banning mobile phone providers from selling phones that are locked to their proprietary network. Governments could consider a similar approach for EV assets. Interoperability is essential to unlocking the full potential of EV flexibility. Again, this is where standards become important. For example, the UK government supported the development of PAS 1878: 2021 Energy smart appliances. ev.energy is working on the policy and technical details of how to implement this specification.
Markets were designed for thermal plants, not distributed resources.
EV VPPs must access the full range of grid services to fully exploit EV charging's flexibility potential. However, energy markets were designed for the incumbent power system of large thermal and hydropower plants, not for distributed resources like EVs. This misalignment causes missed opportunities to harness EV flexibility. Distributed, small-scale assets face other barriers to entering power markets and grid services, such as minimum bid thresholds, access for aggregated assets, and imbalance impacts of flexibility actions on other stakeholders, such as energy suppliers. When considering the requirements to enter flexibility markets, the main metrics are.
Coordination between utility, system operator, vehicle, and EVSE.
Alignment between EV charging infrastructure and the demands of the grid is paramount to fully exploit EV flexibility in a decarbonized energy ecosystem. However, the historical lack of engagement between two crucial industries, utility and power system operators on one side, and vehicle and charge point manufacturers on the other, poses a significant barrier to achieving this alignment. While utility and grid operators rightfully prioritize safety and reliability for consumers, their stringent requirements often hinder the optimal utilization of EV flexibility, impeding the acceleration of power system decarbonization efforts. Simultaneously, without guidance or incentives, manufacturers have yet to ensure vehicle and charge point meters meet the standards expected by grid operators. Asset metering is still a nascent concept in power system operations, yet the case for EV flexibility underscores the urgent need to expedite widespread adoption. While recent efforts have begun to bridge this gap, there remains a substantial need for further collaboration and coordination to fully realize the potential of integrating EVs with the power grid.
Investigating EV asset participation in real-time power markets.
In response to these metering challenges, ev.energy partnered with National Grid ESO (NGESO), the system operator in Great Britain, to investigate the issues of participating EV assets in real-time power markets, in particular the Balancing Mechanism (BM). The BM is NGESO's primary tool to balance supply and demand in the British power system. This work included ev.energy, Flexitricity, and the Power Responsive team at NGESO. Several charge point operators also contributed to the research. The trial focused on the metering accuracy and reporting intervals of aggregated meters.
Aggregate accuracy meets ±1% with just 10 assets.
As part of the Operational Metering Trial, the first case study investigated the aggregate accuracy error of a pool of EV assets. Our analysis showed that the aggregate accuracy error can meet the requirements of a particular service, even though the individual assets in the aggregated group do not. NGESO requires an accuracy error of ±1% for operational metering. Our analysis showed that aggregation can quickly improve the overall accuracy of individual assets with an accuracy error of ±2% (manufacturers say that most chargers and vehicles achieve this). At 10 assets reporting at one moment, the aggregate error reaches ±1% required for the BM. With more than 100 assets, the error falls below ±0.5%. This happens because the random error of the assets cancels out one another when pooled together. For more information on the methodology, refer to Appendix B.
A 1-second aggregated feed from 60-second asset reports.
System operators require near real-time metering of assets providing balancing services for operational metering. Many EV assets would struggle to meet this reporting requirement as they were not designed with the specifications in mind. The 1-second reporting would also create an immense amount of data and transfers across thousands of EV assets. ev.energy's trials showed that a pool of 150 assets can quickly respond to instructions and send an aggregated 1-second operational metering data stream to the system operator. The live tests showed the changes to aggregated metering when the individual asset reports metering less frequently. Each charge point should have the capability to meter once per second, as per the UK's Smart Charge Points Regulations, but the hindrance is the communications to report that metering. The longer interval refresh rates created a lag in the aggregate metering against the 'real' power flow. The average lag increased with longer refresh intervals: a 4-second lag for a 10-second interval, a 14-second lag for a 30-second interval, a 29-second lag for a 60-second interval, and a 57-second lag for a 120-second interval. The majority of home charge points in the UK readily report with a 60-second metering interval. Following these trials, the Power Responsive team at NGESO commissioned an independent expert study to investigate the implications of the longer asset metering intervals on cost-effectively maintaining system security.
95% load shed in 10 seconds. Full ramp-down in 20.
In the live tests, we saw a large difference in response times between ramp-up and ramp-down phases. When ramping down, the pool shed 95% of its load within 10 seconds and fully reached zero after 20 seconds. In comparison, when ramping up EV chargers, the asset pool took approximately 75 seconds to reach 1MW and 60 seconds to reach 95% of the target charge rate. The difference in the ramp rates is due to the EV asset itself increasing the power delivery. While ramping down, the equipment simply switches off the charging. These response times are comparable to or quicker than fast-acting conventional generators, which shows that EV charging can be an effective resource for grid management. All the different interval data streams showed an initial response within a few seconds and although they reached the final dispatch position at different times due to the systemic time lag in operational metering, the true response time remains the same. Our live test results show the response time of the asset pool from when the instruction was received at the asset to the full dispatch.
EV Flex can forecast and measure flexibility capacity required for grid stability.
Our research showed that EV Flex, ev.energy's VPP, can forecast and measure the flexibility capacity of the EV charging fleet using asset metering data, as required for grid stability. Figure 11 in the report visualizes the real-time flexible potential of 5,000 EV charging assets over five days, displaying both turn-up and turn-down potential modulating in line with daily plug-in and grid-need patterns.
A 5x increase in EV assets eligible to participate.
For over five years, ev.energy has successfully scaled a commercially operational VPP. Our data highlights a crucial point: the path to harnessing EV flexibility is blocked without a collaborative effort to reform flexibility infrastructure and align market rules. This report lays out the metering capabilities of different EV charging technologies and compares them to the current metering requirements of flexibility services. The disparities highlight the need for change and convergence. The findings from the Operational Metering Trial show that when aggregated, EV assets can provide highly accurate real-time metering, where the error reduces to less than 1% as the VPP scales. EV assets are capable of impressive response times and when effectively controlled, an EV VPP integrated with EVSE can deliver a full turn-down response within 30 seconds. A turn-up response takes longer, around 60-90 seconds, as this requires power electronics to ramp up power delivery to the battery. Sampling millions of distributed meters every second is not practical or necessary for the majority of flexibility services. By using asynchronous polling, data from thousands of assets flows to the aggregator's platform, where it is combined and sent to the system operator as a single power feed that updates every second with fresh data. This new ev.energy-led approach to aggregated asset metering has been adopted by the National Grid Electricity System Operator in Great Britain for its real-time balancing market. This change has resulted in a five-fold increase in the number of EV assets capable of participating in the market by removing operational metering barriers for the majority of smart charge points. This is a huge step forward for demand response assets participating in power markets. However, more work is still needed in Great Britain and other markets around the world to align EV asset metering capabilities with market needs.
Six recommendations for utilities, regulators, and manufacturers.
To unlock value from EV aggregators, system operators need to appreciate the capabilities of aggregated EV assets, and market requirements should reflect the size of the VPP as a total asset. This will give the operators a new tool to effectively manage the grid. EV and charge point manufacturers can work in partnership with aggregators to ensure their products meet the requirements for grid services before they land on the showroom floor. This will reduce EV charging costs for drivers and open new revenue streams for the manufacturers. To achieve more flexible EV charging, ev.energy recommends.
Ancillary service metering and response requirements by market.
Grid services, which address the variety of the grid's needs, vary across regions. Ancillary services are one form of grid services, which are common across different regions. In Europe, ENTSO.E has published general service categories. The tables below are an overview of the ancillary services' metering accuracy, metering interval, and response times for select markets: Great Britain, Netherlands, and CAISO (California). Where appropriate, the metering accuracy value corresponds to the required accuracy at a 7kW charger.
Three case studies behind the headline findings.
Case Study 1 · Aggregate Metering Accuracy. System operators want to know what is happening on the network. While a highly accurate meter is an acceptable cost when there is only one or several meters at a large power plant, the cost is excessive for individual charge points. We investigated whether the aggregate accuracy error of a pool of chargers is less than the individual chargers' accuracy error. The BM requires a meter to have a 1% accuracy error. The minimum standard for EVSE is 10% (Electric Vehicles Smart Charge Points Regulations 2021); in reality most EVSE models have an accuracy error of 2%. Through modeling we showed that while individual assets do not meet the BM's accuracy requirement, pooling the assets into a VPP achieves the specified accuracy. We applied the Central Limit Theorem: for a meter with ±2%, at least 10 chargers with synchronous metering bring the aggregate error below 1%. For a ±10% meter we need at least 100 vehicles. Case Study 2 · Aggregate metering intervals. We explored the operational metering of individual charge points and how it compared to a fleet reporting every second. The aggregated 1-second feed showed asynchronous reporting at 60-second intervals produced a consistent number of meter readings every second. Ramp-up: ~75 seconds to 1MW, 60 seconds to 95% of target. Ramp-down: 95% load shed in 10 seconds, full reach to zero in 18 seconds. Average time lag: 4s (10s interval), 14s (30s), 29s (60s), 57s (120s). Case Study 3 · Controlling ramp rates. Another part of the trial investigated how controlling EV charging ramp rates influences the time lag. We lengthened the ramp rate by sequencing dispatch signals to charge points, increasing the response time from 60 seconds to approximately five minutes. The staggered ramp rate reduced the average time lags by 10% against the non-staggered ramp rate: 4s (10s interval), 13s (30s), 26s (60s, was 51s), 56s (120s).
Power Responsive, Flexitricity, and ev.energy.
Power Responsive is a stakeholder-led program, facilitated by National Grid ESO, to stimulate increased participation in the different forms of flexible technology such as Demand Side Response (DSR) and storage. It brings together industry and energy users, to work together in a coordinated way. A key priority is to grow participation in DSR, making it easier for industrial and commercial businesses to get involved and realize the financial and carbon-cutting benefits. Flexitricity created and now operates the first, largest, and most advanced flexible energy portfolio in Great Britain and has unsurpassed knowledge of the market and its requirements. Flexitricity's virtual power plant brings together a wide range of asset types across many industries, from large-scale energy storage projects and gas peakers, hospitals, universities, and produce growers to smart homes and EV flexibility. ev.energy is a proven specialist in managed EV charging with activity across Europe, North America, and Australia. Powering over 40 managed charging programs across the globe, we work with utilities, manufacturers, and others to design and deliver customer-facing EV charging management that delights users while delivering grid and network benefits. With over 150,000 EVs under management on our EV Flex VPP worldwide, ev.energy has the depth of knowledge and experience to deliver EV charging management for a clean energy future. To reach out to the team, please contact innovation@ev.energy.
Aggregated asset metering already unlocked a 5x increase. Let's keep going.
The path to harnessing EV flexibility is blocked without a collaborative effort to reform flexibility infrastructure and align market rules. National Grid ESO has shown the way. Other system operators, regulators, manufacturers, and aggregators can follow.