Resilience-driven optimal sizing and pre-positioning of mobile energy storage systems in decentralized networked microgrids. (1st January 2022)

• Record Type: Journal Article
• Authors: Wang, Y.; Rousis, A. Oulis; Strbac, G.
• Is Part Of: Applied energy. Volume 305(2022)
• Publication Date: 2022-01-01
• Subjects: Resilience -- Networked microgrids -- Mobile energy storage systems -- AC optimal power flow -- Decentralized control -- Adaptive genetic algorithm; Power (Mechanics) -- Periodicals; Energy conservation -- Periodicals; Energy conversion -- Periodicals; 621.042
• DOI: 10.1016/j.apenergy.2021.117921
• Languages: English
• ISSNs: 0306-2619

Abstract

Abstract: Networked microgrids are considered an effective way to enhance resilience of localized energy systems. Recently, research efforts across the world have been focusing on the optimal sizing and pre-positioning problems of distributed energy resources for networked microgrids. However, existing literature on mobile energy storage systems mainly focused on single pre-positioning or operational problems rather than a comprehensive resilience-driven planning model capturing both optimal sizing and pre-positioning, especially in the presence of several MGs operating in a networked fashion. Additionally, centralized control is the method typically used to model networked microgrids that may be perceived as unrealistic in presence of high-impact extreme events. Therefore, this paper focuses on developing a three-level defender–attacker–defender model to solve resilience-driven optimal sizing and pre-positioning problems of mobile energy storage systems in networked microgrids with decentralized control. The upper level problem is formulated to obtain optimization results against a certain contingency, while the middle level problem and the lower level problem are merged as a subproblem to select a contingency that can cause the most severe damage. An adaptive genetic algorithm has been employed to search for sizing and positioning decisions and capture various potential attack plans, while a decentralized control approach based on consensus algorithm and linearized AC optimal power flow are utilized to model microgrid operations and capture technical constraints relating to voltage and power loss. Uncertainties relating to renewable energy sources and load profiles are incorporated into the model via stochastic programming. Extensive case studies considering meshed networks and load discrimination into essential/non-essential are developed to demonstrate the effectiveness of the proposed model on accurate decision making of capacities and initial locations. Highlights: A three-level model based on adaptive genetic algorithm is developed. Optimal sizing and pre-positioning of mobile energy storage units are considered. A decentralized control approach based on a consensus algorithm is developed. Internal uncertainties and external contingencies are considered. A linearized AC optimal power flow capturing network and technical constraints is utilized.