Dissipativity-Based Data-Driven Decentralized Control of Interconnected Systems
This work provides a unified, data-driven framework for decentralized control of interconnected systems, addressing the challenge of stabilizing large-scale networks without full model knowledge.
The paper proposes data-driven decentralized control algorithms for interconnected discrete-time LTI systems, using dissipativity-based conditions to ensure global stability. The approach is validated on microgrid examples, demonstrating effectiveness and scalability.
We propose data-driven decentralized control algorithms for stabilizing interconnected discrete-time linear time-invariant systems. We first derive a data-driven condition to synthesize a local controller that ensures the dissipativity of the local subsystems. Then, we propose data-driven decentralized stability conditions for the global system based on the dissipativity of each local system. Since both conditions take the form of linear matrix inequalities and are based on dissipativity theory, this yields a unified pipeline, resulting in a data-driven decentralized control algorithm. As a special case, we also consider stabilizing systems interconnected through diffusive coupling and propose a control algorithm. We validate the effectiveness and the scalability of the proposed control algorithms in numerical examples in the context of microgrids.