Cooperative MPC-Based Energy Management for Networked Microgrids

TLDR

Microgrids are subsystems of the distribution grid that operate as a single controllable system, either connected to or isolated from the main grid. This paper proposes a cooperative model predictive control framework for urban districts with multiple microgrids sharing distributed energy resources. The distributed MPC coordinates each microgrid’s flexible loads, heating, and local generation—using forecasts, prices, and constraints—to optimize shared DER flexibility, minimize grid exchange and overall costs, while guaranteeing constraint satisfaction, cooperation, fairness, and scalability, and is implemented in a virtual testing environment. Numerical experiments demonstrate that the framework is feasible, computationally efficient, and effective, delivering agreed cost savings to each microgrid.

Abstract

Microgrids are subsystems of the distribution grid operating as a single controllable system either connected or isolated from the grid. In this paper, a novel cooperative model predictive control (MPC) framework is proposed for urban districts comprising multiple microgrids sharing certain distributed energy resources (DERs). The operation of the microgrids, along with the shared DER, are coordinated such that the available flexibility sources are optimised and a common goal is achieved, e.g., minimizing energy exchanged with the distribution grid and the overall energy costs. Each microgrid is equipped with an MPC-based energy management system, responsible for optimally controlling flexible loads, heating systems, and local generation devices based on end-user preferences, weather-dependent generation and demand forecasts, energy prices, and technical and operational constraints. The proposed coordination algorithm is distributed and guarantees constraints satisfaction, cooperation among microgrids and fairness in the use of the shared resources, while addressing the issue of scalability of energy management in an urban district. Furthermore, the proposed framework guarantees an agreed cost saving to each microgrid. The described method is implemented and evaluated in a virtual testing environment that integrates accurate simulators of the microgrids. Numerical experiments show the feasibility, the computational benefits, and the effectiveness of the proposed approach.