Electric vehicle batteries alone could satisfy short-term grid storage demand by as early as 2030

The low-carbon energy transition hinges on electrification and the rapid deployment of renewable energy (RE) sources like wind and solar. However, the intermittent nature of RE presents challenges to grid stability and security of supply. Addressing this requires various solutions, including energy storage. Battery storage, particularly due to recent cost reductions, emerges as a key option for improving grid performance. The substantial number of batteries produced for light-duty electric vehicles (EVs) presents a potentially low-cost and efficient method for short-term grid storage. EV batteries can contribute through vehicle-to-grid (V2G) approaches during their in-use phase and as stationary storage after their end-of-vehicle-life (EoL). V2G charging allows dynamic load shifting, while retired batteries, even with reduced capacity, can provide valuable grid services. Utilizing EV batteries enhances supply flexibility and reduces the capital costs and emissions associated with new storage infrastructure. However, quantifying the total grid storage capacity from EV batteries requires careful consideration of various socio-technical factors, including business models, consumer behavior, and battery degradation. Previous studies often lack crucial details such as non-linear battery degradation, geographical temperature variations, and driving intensity, as well as the impact of different battery chemistries and consumer participation rates in V2G and second-use markets. This study addresses these gaps.

Existing literature highlights the potential of large EV fleets to strain grid stability, particularly if charging concentrates during peak times. Prior research on V2G capacity and retired battery capacity provides some insights, but often omits crucial factors like non-linear battery degradation, the effects of temperature variations, and driving intensity. Global-level studies frequently neglect the significance of battery chemistry, consumer participation in V2G markets, and the utilization rates of retired batteries in second-use applications. Therefore, a comprehensive integrated model is needed to address these shortcomings and accurately estimate the future grid storage potential of EV batteries.

This research employs an integrated model linking three sub-models: a dynamic battery stock model, an EV use model, and a battery degradation model. The dynamic battery stock model estimates future EV battery demand based on the International Energy Agency's (IEA) stated policy (STEP) and sustainable development (SD) scenarios, incorporating two battery chemistry sub-scenarios: one dominated by lithium nickel cobalt oxides (NCX) and the other by lithium iron phosphate (LFP). The EV use model incorporates data on daily driving distances, driving cycles, and charging behavior for different EV sizes and types across various regions. The battery degradation model utilizes state-of-the-art data to account for chemistry- and region-specific degradation based on calendar life and cycle life aging, considering factors like temperature, state-of-charge (SoC), depth-of-discharge (DoD), and current rate (Crate). The model explicitly covers China, India, the EU, and the US, combining other markets into a Rest of the World (RoW) region. The model first assesses the technical capacity for short-term grid storage, then analyzes factors influencing real-world capacity, such as EV owner participation in V2G markets and utilization rates of retired batteries. The technical and real-world capacities are compared with projections of future short-term storage requirements from the literature. The study focuses on short-term storage (defined as a typical 4-hour storage system), as EV batteries are less suited for longer-term, seasonal storage. A sensitivity analysis is also conducted by investigating a scenario where all battery electric vehicles (BEVs) are equipped with a smaller 33 kWh battery. The model incorporates daily driving distance distributions from Spritmonitor.de and the US Department of Energy's driving cycles. Battery state-of-charge is simulated using the FASTSim model from NREL, and battery temperature is simplified to city ambient temperature. The semi-empirical battery degradation model considers calendar and cycling life, fitting parameters to available data for NCM and LFP chemistries. The model calculates the relative battery degradation (q) by subtracting the calendar life and cycling life degradation components. The calendar life degradation is modeled as a square-root dependence on time, influenced by temperature and SoC, while cycling life degradation shows a linear dependence on energy throughput, affected by temperature, DoD, Crate, and equivalent full cycles (EFC). The model is further refined to account for path-dependent degradation through the use of virtual time, incorporating the current SoH in degradation rate calculations. The model considers battery EoL generally defined as the time when the remaining capacity is 70-80% of original capacity. The technical vehicle-to-grid capacity is defined considering capacity for driving demands, PHEV capacities, and battery degradation. Actual vehicle-to-grid capacity accounts for consumer participation rates. Technical second-use capacity is defined based on retired batteries with >70% SoH; the actual capacity considers utilization rates. The study explores the interplay of vehicle-to-grid participation rates and second-use utilization rates, providing a comprehensive estimate of the available real-world capacity under various scenarios.

The integrated model projects a substantial global technical capacity for EV battery grid storage, ranging from 32 to 62 TWh by 2050, depending on the scenario (STEP-NCX, SD-NCX, STEP-LFP, SD-LFP). The SD scenario, aligned with the Paris Agreement, exhibits nearly twice the capacity of the STEP scenario. The LFP scenario generally shows slightly higher cumulative capacity than the NCX scenario due to lower degradation rates across most regions. The growth in cumulative vehicle-to-grid and second-use capacity is projected to be substantial (a factor of 13-16 between 2030 and 2050), exceeding the growth rates anticipated in several projections of short-term grid storage demand. Technical capacity from either vehicle-to-grid or second-use alone is sufficient to meet projected demands by 2050. Analysis of vehicle-to-grid opportunities reveals that 21–26% of the global theoretical battery capacity could be available for V2G services by 2050, with driving demands being the most significant constraint (limiting available capacity by 57–63%). Battery degradation contributes to only a minor loss (5% on average). Real-world vehicle-to-grid capacity is largely dependent on participation rates; for example, reaching 10 TWh of storage capacity in 2050 requires participation rates of 38% (STEP-NCX) and 20% (SD-NCX). For second-use batteries, the model assumes 100% collection of retired batteries, with those having >70% SoH considered for repurposing. Annual technical second-use capacity is estimated to reach 2.1–4.8 TWh by 2050, with cumulative capacity reaching 14.8–31.5 TWh, assuming a 10-year lifespan. Utilization rates significantly impact actual second-use capacity. The study finds that combining vehicle-to-grid and second-use capacities, with participation and utilization rates of 50%, would yield a real-world capacity of 25–48 TWh by 2050. Meeting short-term storage demands by 2050 can be achieved with modest vehicle-to-grid participation rates (12–43%) in the absence of second-use batteries. If half of second-use batteries are utilized, the required vehicle-to-grid participation rate drops to below 10%. Even with a scenario assuming all BEVs have smaller 33 kWh batteries, EV batteries can still meet global short-term grid storage demand by 2050 with participation rates of 10–40% and half of second-use batteries utilized for stationary storage.

This study provides a contrasting perspective to previous research suggesting that large EV fleets may negatively impact grid stability. The findings demonstrate the substantial potential for EV batteries to enhance grid stability through both V2G and second-use applications. The projected technical capacity significantly exceeds projected storage demands, indicating a substantial opportunity for grid support. However, realizing this potential depends critically on market participation rates for V2G and utilization rates for second-use batteries. These rates may vary regionally based on factors like incentives and infrastructure. Even under conservative scenarios, EV batteries can satisfy short-term grid storage needs, potentially as early as 2030, depending on storage requirement estimates and participation rates. Policymakers should leverage this opportunity by implementing strategies to encourage market participation, including market-based incentives and regulations. Further research is needed to understand and address barriers to EV user participation in V2G markets.

This study demonstrates that EV batteries possess significant potential to meet short-term grid storage demands, potentially as early as 2030. The projected technical capacity significantly surpasses anticipated storage needs, highlighting a substantial opportunity for grid stabilization and renewable energy integration. However, realizing this potential depends heavily on achieving sufficient market participation rates for V2G and utilization rates for second-life battery applications. Policy interventions, including financial incentives and regulatory frameworks, are crucial to encourage participation and unlock the full potential of EV batteries in grid energy storage. Future research should focus on refining battery degradation models, investigating the impacts of diverse charging behaviors and technologies, and exploring the potential for even more widespread adoption of these technologies.

Several limitations should be considered when interpreting the results. Future battery degradation remains uncertain, and the model relies on existing data and projections that may not fully capture technological advancements or shifts in driving habits. The model simplifies battery temperature to city ambient temperature, potentially overlooking the influence of thermal management systems. The comparison with storage requirements uses a standardized 4-hour storage duration, omitting potential variations at shorter timescales. The assumption of consistent battery capacity per vehicle may not hold in the future, and radical changes to transportation systems could also affect available capacity. Furthermore, the study's assumptions regarding battery collection and repurposing rates might need adjustments as the industry evolves.