Resilient distributed model predictive control for cooperative microgrids under communication loss with demand...

Area of Science:

• Electrical Engineering
• Control Systems
• Energy Systems

Background:

• Networked microgrids require robust control for reliable energy management.
• Integrating demand response (DR) enhances grid flexibility but complicates control.
• Distributed model predictive control (MPC) offers decentralized coordination but faces communication challenges.

Purpose of the Study:

• To introduce a Resilient Distributed Model Predictive Control (RDMPC) framework.
• To enable coordinated energy management in networked microgrids with DR integration.
• To ensure safe and feasible operation under communication impairments.

Main Methods:

• An Alternating Direction Method of Multipliers (ADMM)-based distributed MPC formulation is used for coordination.
• Tie-line reciprocity is maintained via a shared consensus schedule.
• Communication failure resilience is achieved by treating tie-line mismatch as bounded disturbances with reserve margins and local feasibility repair.

Main Results:

• The RDMPC framework ensures feasible reciprocal execution even under packet loss and outages.
• Local feasibility repair with slack variables provides anytime feasibility.
• Demand response ablation reveals significant increases in energy not served and costs when flexibility is removed.

Conclusions:

• The proposed RDMPC framework effectively coordinates networked microgrids with DR under communication constraints.
• The resilience mechanisms ensure safe operation during communication failures.
• Demand response is crucial for maintaining microgrid performance and economic efficiency.