A Dual‐Stage Optimisation Model for Multi‐Microgrid Energy Management: Balancing Economic, Environmental and Social Objectives

This paper introduces a novel two‐stage energy management approach to optimise the operation of multi‐microgrid systems in uncertain conditions. The first stage employs a multi‐objective optimisation model for day‐ahead scheduling, focusing on minimising costs and emissions while maximising social welfare. To generate accurate forecasts for renewable generation and demand, a long short‐term memory (LSTM) neural network is employed for time‐series forecasting, providing reliable inputs to the optimisation framework. The economic welfare maximisation framework accommodates diverse stakeholder interests. In the second stage, the day‐ahead schedule is updated every 5 min based on real‐time conditions to mitigate imbalance costs. The proposed approach integrates real‐time information, enabling efficient adaptation to changing circumstances. Case studies evaluate the approach, showing a significant 15% reduction in emissions compared to a conventional cost minimisation model. Social welfare is enhanced by approximately 12%. These findings highlight the economic viability of integrating high levels of renewable energy by coordinating multiple microgrids and leveraging distributed energy resources. The paper emphasises the environmental and social benefits of a multi‐objective microgrid management strategy within the emerging transactive energy systems paradigm. The proposed two‐stage energy management approach offers a robust framework for optimising multi‐microgrid systems, contributing to a sustainable and efficient energy future.