Two-Layer Model Predictive Control of Microgrids: Cost Optimization and Resilience Through Adaptive Setpoint Coordination

Nizamioǧlu T., Rahimi S., GÖL M.

2025 IEEE Kiel PowerTech, PowerTech 2025, Kiel, Germany, 29 June - 03 July 2025, (Full Text) identifier

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1109/powertech59965.2025.11180692
  • City: Kiel
  • Country: Germany
  • Keywords: battery storage, cost optimization, microgrid, mixed-integer linear programming, model predictive control, multi-agent layers, optimization, resilience, setpoints, two-layer control
  • Middle East Technical University Affiliated: Yes

Abstract

With the growing use of energy storage systems, storing surplus energy from renewable sources has become a profitable strategy. Given the rising trend in the use of renewable energy and the installation of battery storage systems, the significance of microgrids has increased. To ensure microgrid networks perform optimally, operators should maintain resilience for emergencies and cost-effectiveness during normal operating conditions. This study introduces a strategy that addresses these issues, focusing on customer flexibility and computational efficiency through a two-layer integrated model predictive controller. In this framework, setpoints from the upper-layer microgrid operator are issued every 15 minutes to lower-layer building managers, who adjust their operations on a minute basis in response to these setpoints, ensuring effective problem-solving. Additionally, the recommended battery setpoint dispatch substantially lessens the computational load of optimization, especially in larger systems, and achieves the required resilience. Throughout the paper, corresponding methodology and simulation results are presented and verified.