Optimal Deployment of Energy Storage Technologies for Grid Service Provision and Microgrid Formulation

Abstract

The global transition towards deep decarbonization and the widespread integration of renewable energy resources introduce unprecedented operational complexities within modern power systems. Managing these dynamics necessitates massive flexibility across both transmission and distribution networks, which can be achieved through Energy Storage (ES) and decentralized microgrid architectures. Despite the rapid advancement of these technologies, significant barriers remain in their effective integration into the power grid, including the complex modeling of emerging technologies, stringent market regulations, and the lack of electrically informed decentralization strategies. This thesis presents comprehensive and novel optimization frameworks to address these critical gaps in the deployment of ES technologies for Grid Services (GSs) provision and the optimal formulation of sustainable microgrids. At the transmission level, a holistic techno-economic framework is proposed for the optimal sizing and multi-horizon scheduling of grid-connected Advanced Adiabatic Compressed Air Energy Storage (AA-CAES). By incorporating detailed thermodynamic constraints and diverse electricity pricing schemes, the model assesses AA-CAES participation across Energy Arbitrage (EA), Operating Reserve (OR), and Capacity Markets Demand Response (CMDR). Results indicate that while AA-CAES is economically viable for EA and OR primarily in high-variability markets, it remains highly profitable for CMDR across both stable and variable pricing markets. At the consumer level, the research develops an optimal energy resource sizing and management framework for Large Energy Facilities (LEFs) participating in CMDR. To overcome the severe computational complexities introduced by dynamic baseline evaluation rules, a novel convex relaxation of CMDR constraints is introduced. This approach reduces computation time from several hours to fractions of a second while ensuring global optimality with an optimality gap below 0.1%. The framework further demonstrates that aggregating short-duration (e.g., Battery Energy Storage (BES)) and long-duration (e.g., Hydrogen Energy Storage (HES) and Compressed Air Energy Storage (CAES)) technologies maximizes cost savings and environmental sustainability through enhanced arbitrage, emissions-cutting peak-demand reduction, and sustained capacity market obligations. At the distribution network level, a tri-stage optimization framework is designed to systematically decentralize Power Distribution Systems (PDSs) into self-sustaining microgrids. This framework introduces a hierarchical agglomerative clustering method based on electrical line impedance, followed by the stochastic planning of Distributed Energy Resources (DERs). A subsequent allocation strategy optimally places active and reactive power resources, significantly enhancing voltage regulation and reducing system losses under both islanded and cooperative operational modes. These proposed frameworks provide robust, computationally tractable solutions to maximize the economic and operational benefits of ES and decentralized microgrids in modern power systems.