Recently, distributed battery energy storage systems (BESSs) has gained profound importance due to the ever-growing penetration of distributed energy resources (e.g., rooftop solar, wind turbines, etc.) in power grid. Electric vehicles (EVs), driven by carbon emissions control and oil supply risks, are universally believed to be the future of transportation. Hence, recent years have witnessed an urgent demand of establishing advanced EV networks for supporting transportation electrification. The planning and operation of distributed BESSs and EV networks, though possess different technical and economic constraints, share a common bond through their dedication to the charging/discharging operation of batteries. It is therefore an important and urgent research task to investigate...[ Read more ]

Recently, distributed battery energy storage systems (BESSs) has gained profound importance due to the ever-growing penetration of distributed energy resources (e.g., rooftop solar, wind turbines, etc.) in power grid. Electric vehicles (EVs), driven by carbon emissions control and oil supply risks, are universally believed to be the future of transportation. Hence, recent years have witnessed an urgent demand of establishing advanced EV networks for supporting transportation electrification. The planning and operation of distributed BESSs and EV networks, though possess different technical and economic constraints, share a common bond through their dedication to the charging/discharging operation of batteries. It is therefore an important and urgent research task to investigate new models and algorithms for the operation of batteries, and further apply them to the planning and operation of distributed BESSs and EV networks in smart grid.

One of the central challenges confronting BESS investors/operators is that batteries have high capital cost and the degradation of batteries is a very complicated process, making it extremely difficult to estimate the economic value of a distributed BESS over its entire lifetime. The first part of this thesis focuses on the design of a novel stochastic model and algorithm that can efficiently quantify the exact relationship between batteries’ lifetime and specific operational trajectories, and further investigates the operational policy that can characterize the optimal trade-off between achieving better economic value and extending longer lifetime. The second part of this thesis focuses on queueing network modeling, quality-of-service analysis and optimal scheduling of EV networks. Our key contribution in the second part is the establishment of a mixed queueing network model for EV battery swapping and charging stations, of which both the steady-state and asymptotic performance are analytically derived. Based on this queueing model, we further propose a computationally-efficient optimal charging strategy for scheduling a centralized battery charging station that serves EVs based on battery swapping.