Dynamic Event-Triggered MPC Enhanced with Time-Varying Move Blocking Strategy for Open Channel Systems

This paper explores the Model Predictive Control (MPC) method for open channel systems. By employing spatial discretization, the Saint-Venant equations are transformed into an affine nonlinear model. Building upon this, a dynamic event-triggered MPC algorithm enhanced with time-varying move blocking strategy is proposed. This algorithm integrates the advantages of event-triggered and self-triggered, solving optimization problems only upon event triggers, reducing the solver update frequency. The time-varying move blocking strategy ensures that the obtained control sequence remains constant over a certain time, alleviating computational burdens at each step, and its integration with event-triggered control mitigates the potential impact of frequent gate openings on equipment wear. Recursive feasibility and stability of the Optimization Control Problem (OCP) are ensured. Simulation results validate the effectiveness and efficiency of the proposed algorithm.