Researchers from the University of Michigan, including Jingguan Liu, Cong Chen, Xiaomeng Ai, Jiakun Fang, Jinsong Wang, and Jinyu Wen, have developed a novel approach to managing distributed energy storage systems. Their work, published in the journal IEEE Transactions on Power Systems, addresses the challenges of pooling and coordinating large numbers of energy storage devices for power system operations and market participation.
The team’s research focuses on creating a simplified, or “surrogate,” model that can represent the collective behavior of many energy storage devices. This is a complex task due to the diversity of storage technologies, their non-linear operating characteristics, and the large number of devices involved. The researchers’ solution is a mean-field learning framework, which interprets the aggregate behavior of storage devices as the average behavior of a large population.
As the number of devices in the population grows, the researchers found that the aggregate performance of the storage population converges to a unique, convex mean-field limit. This convergence enables tractable population-level modeling, which is crucial for practical applications in power system operations and market clearing. The convexity of the mean-field limit also allows for a price-responsive characterization of aggregate storage behavior, which is essential for market participation.
The researchers constructed a convex surrogate model that approximates the aggregate behavior of large storage populations. This model can be directly embedded into power system operations and market clearing processes. To identify the parameters of the surrogate model, the researchers formulated an optimization problem using historical market price-response data and adopted a gradient-based algorithm for efficient learning.
Case studies validated the theoretical findings and demonstrated the effectiveness of the proposed framework in approximation accuracy, data efficiency, and profit outcomes. This research provides a practical tool for energy storage aggregators to participate in power system operations and markets, potentially improving the integration of renewable energy sources and enhancing grid stability.
The research was published in the IEEE Transactions on Power Systems, a peer-reviewed journal dedicated to the dissemination of original and high-quality technical papers in the area of electric power systems.
This article is based on research available at arXiv.