Distributed MPC: Guaranteeing Global Stability from Locally Designed Tubes * *Work supported by Doctoral Scholarship from CONICYT-PFCHA/ Concurso para Beca de Doctorado en el Extranjero{ 72150125, and the Harry Nicholson PhD Scholarship, Department of Automatic Control &amp; Systems Engineering, University of Sheffild.

Hernandez, Bernardo; Baldivieso, Pablo E.; Trodden, Paul

doi:10.1016/j.ifacol.2017.08.1995

Cited by 9 publications

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

A distributed model predictive control with neighborhood state feedback invariant set for reconfigurable networked systems

Zheng

Hou

2022

Intl J Robust & Nonlinear

View full text Add to dashboard Cite

For a class of networked systems, this article proposed a distributed model predictive control (DMPC) design where the actual control law is composed of the solution of an independent optimization problem and a neighborhood states feedback part. By the proposed method, each subsystem‐based MPC is designed in a distributed way and the neighborhood feedback provides more potential for the designed DMPC to be applied to a wider class of systems. First, an LMI optimization problem is designed for determining the parameters of the neighborhood feedback matrix and corresponding invariant sets where the information of neighboring subsystems is explicitly involved. The designed feedback control law deals with the affection of the interacted subsystems and limits the deviation between the nominal model and the real system within the robust invariant set. Then, the independent suboptimal problem is developed using a predictive model without considering interconnections among subsystems, where the state is restricted by a tightened constraint which is related to the designed robust invariant set. The feasibility and stability of the closed‐loop system are analyzed. The proposed DMPC enables the direct removal or plugging‐in of subsystems. Simulation results demonstrated the effectiveness of the method.

show abstract