On the Distributed Energy Storage Investment and Operations

• 成果类型: Article
• 署名作者: Wu, Owen Q.; Kapuscinski, Roman; Suresh, Santhosh
• 署名单位: Indiana University System; IU Kelley School of Business; Indiana University Bloomington; University of Michigan System; University of Michigan
• 刊物名称: M&SOM-MANUFACTURING & SERVICE OPERATIONS MANAGEMENT
• ISSN/ISSBN: 1523-4614
• DOI: 10.1287/msom.2020.0652
• 发表日期: 2023
• 页码: 2277-2297
• 关键词: capacity investment; distributed energy storage; pooling; production smoothing

摘要

Problem definition: Energy storage has become an indispensable part of power distribution systems, necessitating prudent investment decisions. We analyze an energy storage facility location problem and compare the benefits of centralized storage (adjacent to a central energy generation site) versus distributed storage (localized at demand sites). This problem encompasses optimizing storage capacities across all locations, with the objective of minimizing the total storage investment and energy generation costs. Methodology/results: We employ a stylized model that captures essential features of an energy distribution system, including convex costs, stochastic demand, storage efficiency, and line losses. Using dynamic programming, we optimize storage operations and derive value function properties that are key to analyzing the storage investment decisions. We discern fundamental differences between centralization/localization decisions at the capacity investment stage and the centralization/localization decisions at the storage operations level. Operationally, centrally stored energy offers more flexibility, which is consistent with the conventional understanding of inventory pooling. However, we find that localized storage often emerges as the preferred option at the investment stage under various circumstances. Managerial implications: Storage investment should first be made at the demand locations with positive minimum demand regardless of the level of demand variability. Subsequent storage investment should consider the tradeoffs between centralized versus localized investment. Operationally, the relative magnitudes of storage and line losses drive different optimal storage policies. Despite the differences, these policies are guided by common principles such as pooling inventory and balancing local storage levels.