A scalable planning framework of energy storage systems under frequency dynamics constraints. (February 2023)

• Record Type: Journal Article
• Authors: Zhao, Tianqiao; Raghunathan, Niranjan; Yogarathnam, Amirthagunaraj; Yue, Meng; Luh, Peter B.
• Is Part Of: International journal of electrical power & energy systems. Volume 145(2023)
• Publication Date: 2023-02
• Subjects: Battery Energy Storage System -- Surrogate Absolute-Value Lagrangian Relaxation -- Frequency Dynamics -- Rolling-horizon -- Stochastic planning; Electrical engineering -- Periodicals; Electric power systems -- Periodicals; Électrotechnique -- Périodiques; Réseaux électriques (Énergie) -- Périodiques; Electric power systems; Electrical engineering; Periodicals; 621.3
• DOI: 10.1016/j.ijepes.2022.108693
• Languages: English
• ISSNs: 0142-0615

Abstract

Highlights: A new optimization model to size BESSs considering a new set of various frequency constraints. A realistic representation of wind power uncertainties considering both spatial and temporal evolutions of wind speed across a large region. Consideration of (N-1) generator contingency scenarios in the long-term ESS planning. A novel scalable solution algorithm, to address the increased complexity of a long-term planning problem with frequency constraints to enable practical solutions. Abstract: As the penetration of renewables increases in power systems, the declining system inertia can cause frequency stability issues. Battery energy storage systems (BESSs) respond fast and therefore can relieve the low inertia difficulty but need to be appropriately sized considering the associated cost. This paper presents a novel stochastic optimization model for economically planning BESS capacity while considering the spatial–temporal correlation of wind generation and generator outages under frequency stability constraints, which include the rate-of-change of frequency (RoCoF), frequency nadir (FN), and quasi-steady-state (QSS) frequency. A set of new FN constraints that can be easily linearized is developed. To account for renewable uncertainties, a realistic uncertainty modeling approach, Random Field, is adopted to generate wind generation scenarios by considering both spatial and temporal evolutions of wind speed profiles. The ESS sizing is formulated as a mixed-integer linear programming problem and solved by using a scalable decomposition-and-coordination approach, Surrogate Absolute Value Lagrangian Relaxation (SAVLR). To further improve the scalability and reduce computational burdens, a rolling-horizon-based update is developed and incorporated into SAVLR for providing a practical solution to the long-term planning of very large-scale power systems. A modified IEEE 118-bus system and the Polish system are used to validate the effectiveness and scalability of the model and solution methodology. © 2017 Elsevier Inc. All rights reserved.