| The application of automatic driving technology of heavy-haul trains can save energy and reduce equipment consumption,reduce the difference between drivers’ driving level and habits,improve the reliability of heavy-haul trains,reduce the work intensity of drivers and crew members,and improve line utilization rate and locomotive turnover rate.Automatic train operation(ATO)system is the core to realize automatic driving of heavy-hual trains.ATO system on the main control locomotive receives information from train operation monitoring and recording device(LKJ),central control unit(CCU)and brake control unit(BCU)as well as the wind pressure data at the tail of the train.Based on the automatic driving control algorithm inside ATO system,the corresponding traction/electric braking command and air braking command are calculated.It is implemented by CCU and BCU on the main-control locomotive,and transmitted to the slave control locomotive through wireless reconnection synchronous control system for synchronous control,so as to realize automatic train opreation for heavyhaul trains.This paper focuses on the target curve tracking control algorithm,by analyzing the traction/braking calculation model of the heavy-hual trains,considering the scattered output of the traction/braking force of each locomotive and the coupling relationship between carriages,the multi-particle dynamics model of the heavy train was established.Taking the linear discretization model of the multi-particle dynamics model of heavy-hual trains at the origin of coordinates as the prediction model,considering the relevant time-domain hard constraints,and taking the accuracy and energy saving of speed tracking as the control objective,a closed-loop control algorithm of constrained model predictive control was designed,and the feasibility and stability of the closed-loop control system based on the algorithm are proved.The closed-loop control algorithm was built in Matlab and modeled and simulated in Simulinik.The influence of prediction horizon,control horizon and output weight matrix on the closed-loop control system was compared,and the optimal control parameters were determined.Aiming at the problem that the prediction ability of the original prediction model is limited due to the nonlinear characteristics of the multi-particle dynamic model of heavy-haul trains,the linear parameter-varing model is obtained by linearly discretizing the multi-particle dynamic model of heavy haul trains at different speeds,updating the internal prediction model of the model prediction controller,introducing the terminal-cost function and considering the relevant time-domain hard constraints,and optimizing the original closed-loop control algorithm of constrained model prediction control,The closed-loop control of algorithm for constrained adaptive model predictive control with terminal-cost function is obtained,and the feasibility and stability of the closed-loop control system based on the algorithm are proved.The closed-loop control algorithm was built in Matlab and modeled and simulated in Simulink to verify the superiority of the algorithm. |
