Shared Energy Storage Market Revolutionized by New Auction Mechanism
In a significant leap forward for energy infrastructure innovation, a groundbreaking study introduces a novel operational framework designed to enhance the efficiency, fairness, and economic viability of shared energy storage systems. This advancement comes at a pivotal moment as the global energy sector intensifies its efforts to integrate renewable sources and achieve carbon neutrality. The research, spearheaded by Gao Ciwei and his team at the School of Electrical Engineering, Southeast University, proposes a sophisticated mechanism based on combinatorial auction theory, specifically tailored for a broadened definition of energy storage. This new model promises to address long-standing challenges in the sector, including low user participation, inefficient resource utilization, and the complex interplay between diverse energy storage technologies and user demands.
The transition to a sustainable energy future is fundamentally dependent on the ability to manage the inherent volatility of renewable power sources like solar and wind. These sources, while clean and abundant, generate electricity intermittently, creating significant challenges for grid stability and reliability. Energy storage systems act as a crucial buffer, storing excess power during periods of high generation and releasing it during times of high demand or low generation. However, the high capital cost and limited lifespan of storage hardware have been major barriers to widespread adoption. The concept of shared energy storage, analogous to the sharing economy model popularized by companies like Airbnb and Uber, offers a solution by decoupling ownership from usage. This allows multiple users to access a common pool of storage capacity, thereby spreading the investment cost and improving the overall utilization rate of the assets.
Despite its promise, the existing shared storage market has struggled with inefficiencies. Traditional transaction models have often treated storage as a monolithic commodity, offering only simple, single-period energy or capacity leases. This one-size-fits-all approach fails to capture the nuanced needs of different users. For instance, a data center with a 5G base station may require a guaranteed power output for backup, while an electric vehicle (EV) charging station might prioritize flexible charging capacity over a longer duration. The lack of differentiated service offerings has led to suboptimal matching, reduced user satisfaction, and ultimately, lower participation rates. Furthermore, previous auction mechanisms have frequently overlooked the physical constraints of the storage systems themselves, such as state of charge (SoC), charging and discharging power limits, and energy capacity. This oversight can lead to economically unsound bidding strategies for operators and risks of over-discharging or under-utilizing the storage assets, jeopardizing their long-term health and profitability.
The research by Gao Ciwei, Lin Gujing, Song Meng, Zhang Zitao, Chen Tao, Zhang Weishi, and Zhou Jianing directly confronts these limitations. Their core innovation lies in the creation of a “generalized energy storage” (GES) model. This model is not confined to conventional battery systems but is designed to be a unifying framework that can encapsulate the operational characteristics of a wide array of resources with inherent storage capabilities. This includes not only traditional batteries but also virtual or “demand-side” resources such as thermal energy storage in air conditioning systems, the flexible charging patterns of EVs, and the backup batteries in 5G communication base stations. By aggregating these diverse resources into a single, standardized model, the researchers dramatically reduce the complexity of managing a heterogeneous fleet of storage assets. This aggregation allows a shared energy storage operator (SESO) to present a simplified, yet highly flexible, interface to the market, effectively turning a complex network of distributed resources into a single, powerful, and controllable market participant.
A pivotal element of the new mechanism is the redefinition of the transaction itself. Instead of selling raw energy or a simple capacity lease, the researchers introduce two distinct and complementary trading varieties: power usage rights and capacity usage rights. This dual-market structure is a fundamental shift that caters to the diverse operational profiles of potential users. Power usage rights grant a user the ability to charge or discharge the storage system at a fixed power level for a specific time period. This is ideal for users with predictable and time-critical needs, such as an industrial facility that must perform a specific energy-intensive process within a narrow window. Conversely, capacity usage rights provide a user with a fixed amount of energy that can be charged or discharged at a flexible rate over a continuous period. This offers greater operational freedom, allowing users to optimize their own charging and discharging schedules based on their internal needs and real-time conditions, which is particularly valuable for entities like commercial buildings managing their HVAC systems.
To facilitate a dynamic and responsive market, the mechanism is built upon the foundation of a combinatorial auction. Unlike simple auctions where bidders submit offers for individual items, a combinatorial auction allows participants to bid on complex combinations of goods. In this context, users are not limited to bidding on a single right for a single hour. They can submit bids that express sophisticated, coupled demands. The researchers designed four primary bidding types to capture these complex preferences. An “atomic” bid is a straightforward offer for a single right in a single period. An “XOR” (exclusive or) bid allows a user to offer different options—such as charging in the morning or discharging in the evening—where only one of the options can be accepted, reflecting a user’s substitutable needs. An “OR” bid permits a user to request multiple rights, with the possibility of any or all being accepted, representing partially substitutable demands. Most significantly, an “AND” bid enables a user to request a bundle of rights across multiple time periods and types, with the condition that either the entire bundle is awarded or none of it is. This “all-or-nothing” structure is essential for users with complementary needs, such as an EV fleet operator that requires a guaranteed charging block across several consecutive hours to ensure all vehicles are ready for service the next day.
The sophistication of the mechanism extends to the seller side as well. The shared energy storage operator (SESO) is not a passive seller but an active participant whose bidding strategy is deeply informed by the physical and economic state of the aggregated storage resources. The researchers developed a novel bidding curve for the SESO that explicitly incorporates the state of charge (SoC) of the storage system. This is a critical advancement. The model dictates that the price for charging rights should be lower when the SoC is low, incentivizing charging to replenish the system. Conversely, the price for discharging rights should be higher when the SoC is high, encouraging discharge to utilize the stored energy. The price for capacity rights is also dynamically adjusted, increasing as the SoC rises because a highly charged system has less available capacity to absorb new energy. This intelligent pricing strategy ensures that the market signals are aligned with the physical health and operational efficiency of the storage assets, preventing scenarios where the system is driven to dangerously low or high SoC levels, which can accelerate degradation and increase operational risk.
The auction process itself is a carefully orchestrated sequence designed to maximize overall social welfare, defined as the total net benefit to all participants. After buyers and the SESO submit their sealed bids, an auctioneer runs a complex optimization model to determine the winning combination of bids. The primary goal is to maximize the difference between the total value users place on the services they receive and the total cost to the SESO of providing them. To ensure market liquidity and prevent deadlock when bid prices are closely matched, a small penalty cost is introduced to incentivize the clearing of trades. The model rigorously enforces all physical constraints of the GES system, ensuring that the final allocation of power and capacity rights does not violate the SoC limits, maximum charging/discharging rates, or total energy capacity. This comprehensive integration of market economics with physical system constraints is what sets this mechanism apart from previous, more abstract models.
The financial settlement is designed to be both fair and incentive-compatible. Upon determining the winners, the final transaction price for each matched bid is set as the average of the buyer’s bid price and the SESO’s ask price. This “high-low matching” approach ensures that the gains from trade— the social welfare—are shared equitably between the buyer and the seller. This balanced distribution of benefits is crucial for fostering trust and encouraging truthful bidding, as participants are less likely to engage in strategic behavior to manipulate the market when they know the outcome will be fair.
The research team conducted a detailed case study to validate their theoretical framework. The simulation involved one SESO and nine diverse users over a six-hour auction period. The results were compelling. The mechanism successfully allocated resources in a way that respected all physical constraints, with the SoC of the aggregated storage system remaining safely within its operational bounds throughout the simulation. The pricing was dynamic and responsive, with the SESO’s bid prices varying significantly across time periods based on the evolving SoC and the level of demand, accurately reflecting the scarcity of the service. The final outcome demonstrated a substantial social welfare gain, and the settlement mechanism ensured this gain was nearly equally split between buyers and sellers, validating the fairness of the design.
A particularly insightful part of the analysis was a sensitivity study on the parameters of the SESO’s pricing curve. The researchers found that the shape of this curve is critical for market stability and efficiency. If the curve is too steep—meaning prices change dramatically with small changes in demand—it can lead to large fluctuations in clearing prices and a significant reduction in the total volume of trades, ultimately harming both user access and operator revenue. Conversely, if the curve is too flat or constant, it fails to send accurate price signals about resource scarcity, potentially leading to inefficient allocation and an inability to manage the SoC effectively. This finding provides crucial practical guidance for operators, emphasizing the need for a carefully calibrated and balanced pricing strategy.
The implications of this research are far-reaching. It provides a robust, ready-to-implement blueprint for the next generation of shared energy storage markets. By enabling a diverse array of users—from residential prosumers with rooftop solar to large industrial complexes and EV charging networks—to efficiently and fairly access storage services, this mechanism can dramatically accelerate the integration of renewable energy. It enhances grid resilience by providing a more responsive and flexible pool of balancing resources. Economically, it lowers the barrier to entry for users who need storage but cannot afford a dedicated system, while providing a clear and profitable business model for storage operators.
While the current study presents a single-operator model, the authors note its strong extensibility to a multi-operator, multi-user environment, paving the way for a truly competitive market. Future work will focus on expanding the model to incorporate multiple service types and time scales, as well as refining the aggregation process to better account for the individual preferences and operational constraints of the underlying resource owners. Nonetheless, the foundational work presented here marks a significant milestone. It moves the conversation from the theoretical promise of shared storage to a concrete, practical, and economically sound mechanism that is poised to play a vital role in building a more flexible, efficient, and sustainable power system for the future.
Gao Ciwei, Lin Gujing, Song Meng, Zhang Zitao, Chen Tao, Zhang Weishi, Zhou Jianing, School of Electrical Engineering, Southeast University. Published in Proceedings of the CSEE, DOI: 10.13334/j.0258-8013.pcsee.231884