Distributed and Privacy Preserving MPC With Global Constraints Over Time-Varying Communication

Distributed model predictive control (DMPC) has recently attracted extensive research attention due to its high flexibility and outstanding performance. However, most of those studies focus only on optimizing some subsystems with sets of local constraints, or on designing some schemes over time-invariant communication networks, by ignoring the cooperation of multiple subsystems with global constraints and also the privacy preservation of the subsystems in complex communication networks. Current research on DMPC is incapable to address all these challenges. Therefore, a novel distributed and privacy preserving MPC algorithm with global constraints over time-varying communication networks is proposed in this article. Firstly, we transform the MPC optimization problem that includes all the subsystems into its dual problem. Then, a fully distributed dual gradient algorithm with row-stochastic matrix is developed for solving that problem. For further reduction of the computational cost, an event-triggered communication protocol is designed, which is realized through a distributed trigger criterion. Under reasonable assumptions, we prove that the algorithm can converge to the optimal solution while assuring the recursive feasibility and exponential stability of a closed-loop system. In addition, considering that the sensitive information may be leaked or eavesdropped during the subsystem interaction, a distributed encryption algorithm with privacy preservation is presented in an updated version. We demonstrate that the proposed algorithm is secure and can effectively protect the privacy of the subsystems. Finally, the effectiveness and performance of the proposed approaches are substantiated through numerical simulations.