Research on model predictive control of metro train based on input blocking strategy

The controller design of an automatic train operation system under multiple operating constraints is studied in this paper. Based on the nonlinear multi-particle model of metro trains, a predictive control algorithm of the automatic train operation model has been established to meet the requirements of train punctuality, energy saving, and passenger comfort. On this basis, the input blocking technology was introduced to adjust and reorganize the distribution of the control time domain in the prediction time domain of the (Model Predictive Control) (MPC)algorithm. At the same time, the simulation model was used to verify the effectiveness of this algorithm. The simulation results show that when the operational constraints are met, the calculation time of this algorithm is less than that of the standard model predictive control algorithm, when accuracy is ensured.