Energy Management of Multi-Microgrid Systems with the Presence of Renewable Resources an‎d Battery Energy Storage

12/22/2025 12:00:00 AM

In this study, a two-layer hierarchical optimization framework is proposed for the intelligent operation of multi-microgrid systems. In the proposed structure, each microgrid is equipped with distributed generation resources, including photovoltaic panels, wind turbines, diesel generators, an‎d energy storage systems. In addition, a deman‎d response program (DRP) is implemented to manage electricity consumption during peak load hours. The proposed model consists of two decision-making layers. In the first layer, each microgrid independently optimizes its internal resources using a local energy management system. In the second layer, a central energy management system is responsible for power allocation an‎d coordination among the microgrids an‎d the upstream grid, with the objective of minimizing the overall operational cost. To address uncertainties in generation an‎d load, a forecasting model based on a multilayer perceptron neural network is developed, which accurately predicts four critical variables: load deman‎d, solar irradiance, wind speed, an‎d ambient temperature. The developed model is implemented an‎d eva‎luated in the GAMS platform using real electricity market data from Ontario, Canada. Simulation results under multiple scenarios demonstrate that the proposed two-layer structure not only leads to a significant reduction in total system operating costs but also improves microgrid performance during critical hours through the application of deman‎d response programs an‎d optimal utilization of energy storage resources. Furthermore, by introducing an operational innovation, load peak shaving is modeled in the first microgrid, an‎d its effectiveness in reducing peak deman‎d an‎d enhancing operational flexibility is validated. From structural, computational, an‎d operational perspectives, the proposed model exhibits high flexibility an‎d can be employed as a practical an‎d implementable solution for intelligent an‎d economic operation of future microgrids.

ريزشبكه , مديريت انرژي , سيستم‌هاي تجديدپذير , شبكه‌هاي عصبي

Microgrid , Energy Management , Renewable Energy Systems , Neural Networks

Sepideh cheraghi

Dr mohsen kalantar