Adaptive Reconfiguration of Microgrid Based on Structured Loss Component Decomposition for Energy Efficiency Enhancement

The rapid development of distributed energy resources, renewable generation technologies, and energy storage systems has fundamentally transformed the architecture of modern power distribution networks. Traditional centralized power systems are increasingly being supplemented by decentralized energy infrastructures capable of operating both in grid-connected and islanded modes. Among these emerging solutions, Microgrids have attracted considerable attention due to their ability to improve energy reliability, resilience, operational flexibility, and integration of renewable energy sources.A typical Microgrid combines distributed generation units, photovoltaic systems, wind turbines, battery energy storage systems, controllable loads, and power electronic converters within a coordinated control framework. The growing penetration of converter-interfaced resources and nonlinear loads introduces additional challenges associated with power quality degradation, voltage unbalance, harmonic distortion, reactive power circulation, and neutral conductor currents. These phenomena increase technical power losses, reduce network efficiency, and negatively affect the operational performance of Microgrid infrastructures.One of the most effective approaches to improving the energy efficiency of Microgrids is network reconfiguration. Reconfiguration involves changing the status of switching devices to modify network topology and power flow distribution while maintaining operational constraints. Numerous studies have demonstrated that appropriate reconfiguration strategies can significantly reduce power losses, improve voltage profiles, balance feeder loading, and enhance system reliability. However, most existing reconfiguration methods employ optimization criteria based solely on total active power losses. Although such approaches provide useful performance indicators, they do not reveal the physical mechanisms responsible for the formation of losses within the electrical network.In modern three-phase four-wire systems, total power losses are influenced by several distinct factors. In addition to unavoidable conductive losses, significant contributions may originate from reactive power flows, oscillations of instantaneous active power caused by harmonic distortion, and currents circulating through the neutral conductor under unbalanced operating conditions. Conventional optimization techniques treat these phenomena collectively, making it difficult to identify the dominant source of inefficiency and select the most effective corrective actions.To overcome these limitations, this paper proposes an adaptive Microgrid reconfiguration approach based on structured decomposition of loss components. The proposed methodology utilizes a physically interpretable representation of total power losses as a combination of minimum achievable losses, losses associated with reactive power, losses caused by active power pulsations, and losses resulting from neutral conductor currents. Such decomposition enables a more detailed assessment of energy losses and supports the development of targeted control actions aimed at mitigating specific loss mechanisms.The proposed framework integrates structured loss evaluation with adaptive network reconfiguration procedures. By combining real-time monitoring, system state assessment, and topology optimization, the method provides the capability to continuously adapt Microgrid configuration according to changing operating conditions, generation patterns, and load demands. Unlike conventional approaches focused exclusively on integral loss minimization, the proposed method establishes a direct relationship between optimization objectives and the physical origin of energy losses.The objective of this research is to develop an adaptive reconfiguration methodology for Microgrids that improves energy efficiency through structured loss component decomposition and physically informed optimization. The study presents the mathematical foundations of the proposed approach, formulates the optimization problem, and develops a framework for adaptive network control capable of supporting intelligent energy management in modern distributed power systems.