Integrated energy scheduling for grid-connected microgrids using battery degradation-aware optimization and coordinated control strategies

Regional clusters of energy producers and consumers can be realized by integrating household Battery Energy Storage (BES) systems with Renewable Energy Sources (RES) and linking them to the main utility grid. These clusters, functioning as grid-connected microgrids (MGs), act as controllable units within the broader energy distribution network. As distribution systems evolve to include higher MG penetration, the need for efficient and scalable energy management becomes critical to ensure technical compatibility with grid objectives and operational constraints. Additionally, understanding the impact of battery usage patterns on degradation is essential for developing long-term, cost-effective energy management strategies. This paper presents a novel Grid-Connected Microgrid Energy Management (GCM-EM) model that incorporates both economic and technical constraints, with Battery Energy Storage (BES) as the central flexible resource. The proposed model supports both uncoordinated (microgrid-autonomous) and coordinated (DSO-integrated) scheduling schemes. The novelty lies in its ability to capture real-world BES degradation dynamics-including cycle aging and depth-of-discharge (DoD) effects-within an optimization-based energy scheduling framework. The scheduling model leverages mixed-integer programming, AC optimal power flow, and rolling-horizon control to achieve both local and system-level operational goals. The model's performance was validated using simulations on two representative test systems: a university campus distribution grid and a standardized 33-bus power network. Results demonstrate that localized MG optimization can reduce energy costs by up to 2%. At the same time, coordination with the Distribution System Operator (DSO) further enhances grid-level cost efficiency-though sometimes at the expense of local MG economic optimality. Importantly, the model preserves data privacy during coordination and maintains compliance with distribution grid constraints. Furthermore, the model was implemented in a real building-level microgrid (BMG), where it effectively minimized BES operational and degradation costs. Compared to conventional EMS frameworks that ignore battery wear, the proposed model achieved a 3% reduction in combined annual energy and degradation costs. Integration into actual EMS platforms also enabled optimized BES dispatch, reduced municipal grid dependence, enhanced MG operational flexibility, and lowered overall network operating expenses. This research provides a comprehensive and practically validated energy management architecture for BES-integrated microgrids. By combining advanced scheduling strategies with accurate degradation modeling and multi-agent coordination, the proposed system represents a significant advancement toward economically sustainable and technically robust distributed energy networks.