Energy Management of Microgrids: An MPC-Based Techno-Economic Optimisation for RES Integration and ESS Utilisation

Abstract

The rapid growth in electricity demand poses significant challenges to power systems that require efficient energy management frameworks. This paper proposes a real-time energy management framework for DC residential microgrids based on model predictive control (MPC) and a mixed integer nonlinear programming (MINLP) optimisation approach. Unlike conventional methods relying on static datasets, the proposed framework integrates real-time optimisation and predictive data derived from microgrid simulations, which enhances adaptability, scalability and operational efficiency. A demand response (DR) programme for residential loads further reduces reliance on the upstream grid during peak hours. The performance of the framework is validated across multiple scenarios, like increased demand and fluctuations in renewable energy and demand. Results indicate significant economic benefits through optimised power exchange with the upstream grid, achieving cost reductions of 21.05%, 18.36% and 16.59% under real-time, three-tariff and fixed tariffs, respectively, compared to conventional MPC methods. Moreover, it minimises state of charge (SoC) changes to a maximum of 17.4%, significantly reducing energy storage depreciation and extending longevity as technical performance. These findings demonstrate the framework's ability to adapt dynamically to operational changes, optimise power exchanges and achieve technical and economic goals, making it a scalable solution for real-time residential microgrid energy management.