Cooperative Tracking Control Method For Multiple Trains

High-speed railways,boasting remarkable safety and significantly reduced travel time,have emerged as a favored mode of transportation between major cities.However,the rapid expansion of these rail networks has led to bottlenecks,limiting the capacity for further growth.One approach to achieving high-density operation under current conditions is through the cooperation of multiple trains,shortening the interval while maintaining safety.This paper investigates multi-train cooperative tracking control based on a single-point-mass train dynamics model and bidirectional train communication topology.The main contents of this paper are as follows:1)An adaptive cooperative tracking control method is proposed for addressing challenges stemming from model uncertainties and external disturbances in multi-train cooperation.A coupled sliding mode surface is constructed based on the position,velocity,and acceleration information of adjacent trains.The radial basis function(RBF)neural network and adaptive techniques are employed to estimate and compensate for these uncertainties and disturbances,minimizing tracking errors.An observer is utilized to obtain velocity and acceleration information from neighboring trains,and an output feedback tracking controller is designed to reduce the required state information.Lyapunov stability theory is employed to prove the stability and convergence of both single and multiple trains,with simulation results validating the effectiveness of the proposed control strategy.2)A neuro-adaptive fault-tolerant cooperative tracking control method is designed for multi-train cooperation in the presence of actuator faults,saturation limits,and unknown operation resistance.Within a bidirectional information flow topology framework,adaptive techniques are employed to estimate and compensate for actuator failures with partial loss of effectiveness.An RBF neural network is introduced to approximate unknown running resistance,and an adaptive law for neural network weights is provided.To address the actuator saturation issue,an anti-saturation auxiliary structure is designed,reducing tracking errors caused by saturation.Lyapunov stability analysis ensures the uniform ultimate boundedness of all closed-loop system signals,and numerical simulation results confirm the effectiveness of the proposed control scheme.3)A multi-train event-triggered neuro-adaptive control method based on minimal learning parameters is developed for resource-constrained cooperative tracking control.An eventtriggered mechanism is introduced,designing a fixed threshold strategy based on control input signal errors to reduce communication transmission information.The minimal parameter learning method of the neural network is employed for online estimation of the train’s basic resistance,reducing the number of online learning parameters and conserving computational resources.Theoretical analysis demonstrates that all closed-loop system signals are uniformly ultimately bounded and the Zeno behavior is avoided.Numerical simulation results validate the effectiveness of the proposed control approach.