Model Predictive Control Based Heavy Load Train Formation Tracking Method

Train formation control has gradually become an important means to improve railway transportation capacity. In response to the characteristics of heavy haul trains with large traction mass, long train groups, and uneven power distribution, this paper proposes a model predictive control based heavy haul train formation tracking method. Firstly, a multi particle train dynamics model considering the influence of coupler force was constructed; In the formation, the pilot train runs according to the predetermined speed trajectory, and subsequent trains use the previous train status as a reference input. The MPC controller predicts and optimizes their own operating status in real time. In the control objectives, factors such as operational impact, energy consumption, and speed error are comprehensively considered to improve the smoothness and energy efficiency of formation operation. To cope with the computational pressure brought by high-dimensional models, an event triggering mechanism is further introduced to dynamically adjust the control sampling frequency, thereby reducing the burden of online optimization. The simulation results have verified the effectiveness and practicality of the proposed method in achieving high-density and small interval safe formation operation, providing new solutions and technical support for intelligent control of heavy-duty trains.