EVs as Shared Energy Storage Cut Costs and Emissions in Smart Parks
A groundbreaking study published in Automation of Electric Power Systems introduces a novel approach to enhancing the sustainability and economic efficiency of park-level integrated energy systems (PIES) by leveraging electric vehicles (EVs) as mobile energy storage units. The research, led by Yi Wang from Zhengzhou University, demonstrates how integrating EVs into the energy infrastructure of industrial parks can significantly reduce carbon emissions while lowering operational costs for both energy providers and consumers.
The concept of PIES has gained traction in recent years as industries and municipalities seek more efficient ways to manage energy consumption. These systems integrate various energy sources—such as electricity, heat, and gas—into a unified network that serves multiple users within a defined area, such as an industrial park or campus. While PIES offer improved energy utilization and reliability, their widespread adoption has been hindered by high initial investment costs, particularly for stationary energy storage systems. Traditional battery storage, while effective, remains expensive and underutilized in many cases, prompting researchers to explore alternative solutions.
Wang and his team propose a paradigm shift: instead of relying solely on fixed storage units, the system can harness the collective battery capacity of EVs parked within the industrial park. This “shared energy storage” model transforms EVs from passive consumers into active participants in the energy ecosystem. By aggregating the available battery power of vehicles during their downtime, the system gains access to a flexible, scalable, and cost-effective energy reserve.
The core innovation lies in the development of a dispatchable potential model for EV clusters. Unlike earlier studies that treated EV charging as a simple load, this research accounts for the variability in arrival and departure times, initial state of charge (SOC), and user preferences. Using a bidirectional long short-term memory (Bi-LSTM) network, the team predicts the real-time availability of EV battery capacity with high accuracy. This predictive capability allows the microgrid operator (MGO) to plan energy dispatch more effectively, ensuring that stored energy from EVs is available when needed most—such as during peak demand periods or when renewable generation is low.
The integration of EVs into the PIES framework is not just a technical upgrade; it represents a strategic rethinking of energy economics. In traditional setups, the MGO sets energy prices based on generation costs and market conditions, while users respond by adjusting their consumption. However, this one-way pricing model often fails to optimize overall system performance or incentivize sustainable behavior. Wang’s model introduces a Stackelberg game-theoretic framework, where the MGO acts as the leader and users as followers. The MGO proposes energy prices, and users respond by optimizing their energy use—including when to charge or discharge their EVs, shift flexible loads, or use electric heating—based on cost and comfort considerations.
This dynamic interaction leads to a more balanced and efficient energy market. For instance, when electricity prices are low—typically during off-peak hours or when solar generation is high—EVs are encouraged to charge. Conversely, during high-price periods, EVs can discharge back into the grid or supply power directly to users, reducing reliance on fossil-fuel-based generation. This not only lowers costs for users but also reduces the need for the MGO to operate less efficient or more polluting backup generators.
To further enhance the environmental impact, the study incorporates a tiered carbon emission trading (CET) mechanism with penalty and reward factors. Under this system, the MGO is allocated a carbon emission quota. If actual emissions fall below the quota, the operator receives financial incentives; if they exceed it, penalties are imposed. This creates a strong economic incentive to minimize carbon output. The results are striking: in simulations, the combination of EV-based storage and CET reduced total carbon emissions by up to 15% compared to conventional systems without these features.
One of the most compelling findings is the dual benefit for all stakeholders. The MGO sees increased revenue through optimized energy trading and carbon credits, while users enjoy lower energy bills and greater control over their consumption. Moreover, EV owners benefit indirectly through reduced electricity costs and potentially direct compensation for grid services, although the current model focuses on aggregated charging station operations rather than individual vehicle-to-grid (V2G) payments.
The study also highlights the importance of flexibility in load management. In the proposed model, users can shift not only their electricity use but also their heating demand. When electricity is cheap and abundant, they can use electric heat pumps to generate warmth, reducing the need to purchase thermal energy from the MGO. This “electricity-to-heat” substitution not only cuts costs but also enhances system resilience by diversifying energy pathways.
A key advantage of using EVs as shared storage is scalability. As EV adoption grows, so does the available storage capacity, without requiring additional infrastructure investment from the MGO. This contrasts sharply with fixed battery systems, which require significant capital and have limited expansion potential. Furthermore, because EVs are already being purchased for transportation purposes, their use as energy storage represents a form of dual-purpose asset utilization, improving overall resource efficiency.
The research team conducted extensive simulations using real-world data from a municipal PIES, comparing four different scenarios: one with traditional storage and no carbon trading, one with EV storage but no CET, one with CET but no EV storage, and a final scenario combining both innovations. The results consistently showed that the integrated approach outperformed all others in terms of cost savings, emission reductions, and stakeholder satisfaction.
In the baseline scenario with conventional storage, the system emitted 40,109 kg of CO₂ over a 24-hour period. Introducing EV-based storage reduced this to 38,867 kg, while adding CET alone brought it down to 33,784 kg. However, the combination of both strategies—EV storage and CET—achieved the best result, with emissions dropping to 31,558 kg. This represents a 21.4% reduction compared to the baseline and a 6.5% improvement over CET alone.
Financially, the benefits are equally impressive. The MGO’s revenue increased from 9,893 yuan in the baseline to 11,321 yuan with EV storage, despite lower energy sales, due to more efficient operations and reduced fuel costs. User income—defined as the value derived from lower energy expenses and optimized usage—rose from 21,131 yuan to 23,048 yuan in the combined scenario. Even the EV charging stations themselves saw positive returns, generating income by selling stored energy back to users during peak hours.
Beyond the numbers, the study underscores a broader shift in how we think about energy systems. Rather than viewing EVs as isolated devices, the research positions them as integral components of a smart, responsive grid. This aligns with emerging trends such as “photovoltaics, energy storage, direct current, and flexibility” (PEDF) in building design, where distributed energy resources are seamlessly integrated to maximize efficiency and sustainability.
The implications extend beyond industrial parks. The same principles could be applied to commercial districts, university campuses, or even residential communities with high EV penetration. As urban areas strive to meet climate goals, models like this offer a practical pathway to decarbonization without sacrificing economic viability.
Another significant contribution is the robustness of the EV cluster model. By using Minkowski sum theory to aggregate individual vehicle constraints into a collective dispatchable potential, the researchers ensure that the system remains physically feasible and operationally stable. This mathematical rigor distinguishes the work from more heuristic approaches and provides a solid foundation for real-world implementation.
The use of advanced machine learning—specifically Bi-LSTM—for forecasting EV availability also sets this study apart. Unlike standard LSTM models that only consider past data, Bi-LSTM analyzes both historical and future trends, improving prediction accuracy. This is crucial in a dynamic environment where small errors in SOC or timing estimates can lead to suboptimal dispatch decisions or even grid instability.
From a policy perspective, the study supports the case for incentivizing not just EV adoption but also their integration into energy markets. Regulatory frameworks that enable fair compensation for grid services, coupled with supportive tariff structures, could accelerate the deployment of such systems. Utilities and grid operators may need to adapt their business models to accommodate distributed, mobile storage resources.
Looking ahead, the research team identifies several promising directions for future work. One is the integration of EV routing and parking behavior into the model, allowing for even more precise forecasting of available storage capacity. Another is the exploration of cooperative game theory among users, where aggregation of demand response can lead to better collective outcomes. Additionally, the model could be extended to include other distributed energy resources, such as home batteries or smart appliances, creating a fully integrated demand-side management ecosystem.
The success of this approach also depends on user engagement and trust. For the system to function optimally, users must be willing to allow some degree of control over their EV charging schedules. Transparent pricing, clear communication of benefits, and user-friendly interfaces will be essential to gaining public acceptance. Pilot programs in real-world settings could help validate the model and refine its implementation.
In conclusion, the research by Yi Wang and colleagues presents a compelling vision for the future of urban energy systems. By reimagining EVs not just as vehicles but as mobile power banks, the study offers a scalable, cost-effective solution to two of the most pressing challenges in energy management: reducing carbon emissions and lowering costs. The integration of game theory, machine learning, and carbon trading mechanisms creates a sophisticated yet practical framework that could serve as a blueprint for smart energy communities worldwide.
As cities and industries continue to grapple with the twin pressures of climate change and economic efficiency, innovations like this demonstrate that sustainable solutions are not only possible but also profitable. The transformation of EVs from transportation tools to energy assets marks a significant step toward a more resilient, flexible, and low-carbon energy future.
Yi Wang, Zikang Jin, Yaoqiang Wang, Mingyang Liu, Jun Liang, Zhengzhou University; Automation of Electric Power Systems, DOI: 10.7500/AEPS20230424003