Dynamic Communication Topology Based Coordinated Control Of High-speed Trains

As a safe,reliable,fast and comfortable,and eco-friendly transportation system with high loading capacity,high-speed railways have been booming in recent years worldwide,and have already become a crucial part of the comprehensive transport system,and play important roles in people's livelihood.At present,the operation of high-speed trains is mainly based on the quasi-moving block signaling system mode,which uses manual driving mode and each train adjusts its running status individually.And un-der the current short-interval,high-density,dynamic-coupling operating conditions,it is difficult to attenuate the impacts of emergencies in the railway network on regular oper-ation of the trains,and further improve the operating efficiency of high-speed railways while ensuring operational safety.Coordinated control of high-speed trains helps to im-prove the operation efficiency and enhance the robustness of high-speed railway system to the emergencies,and has received significant attention in recent years.In view of the characteristics of time delay,discrete sampling,and time-varying switching topology of the train communication network,and considering the practical constraints of the train control system,this dissertation mainly focuses on coordinated control problems for the high-speed train system,and rapid composite fault diagnosis of the trains are also con-sidered to guarantee the stability of the train dynamics.The main work is summarized as follows,1.A composite fault diagnosis approach is proposed for dynamic systems of high-speed trains subject to multi-source disturbances.Considering the impacts of the gust disturbance,and external and internal interferences that are difficult to model,on the fault diagnosis results of the train,a fault detection filter is designed based on the unknown input decoupling mechanism and the H_?index in a finite frequency domain.The eigen-subspace assignment technique and the linear matrix inequality optimization method are used to derive parametric matrices and existing conditions of the fault detection filter.The fault detection filter guarantees that the residuals are decoupled from the multi-source interference without truncating the influence of fault signals such that rapid detection and isolation of composite faults is achieved.2.Distributed time-varying low-gain coordinated control laws for high-speed train-s under a communication network that subjects to information transmission delays are proposed.Aiming at the influence of information transmission delays on the operation of high-speed trains,a third-order nonlinear time-delay control model of a train is construct-ed based on the basic principles of train control system of the high-speed railway,and distributed coordinated control laws,which take the structure of the low-gain feedback controller,are established.Further,the time-varying and convergence characteristics of the parameter of the distributed coordinated low-gain controllers are exploited to guaran-tee rapid adjustment of train status and accurate tracking of the reference train operation curve under the constraint of information transmission delays.3.In view of the constraints on the updating frequency of the actuator of the trains,event-triggered coordinated control approaches of high-speed trains are proposed.Aim-ing at the sampled-data based control of trains,the tracking control problems are trans-formed into stabilization problems for control systems with time-varying bounded input delay by analyzing the update rules of the tracking/braking force,and power decentral-ized train oriented event-triggered low-gain control laws are established.Further,the results are extended to the coordinated control problem for multiple train system,and distributed event-triggered low-gain controllers,which can effectively reduce the number of controller updates and the communication frequency,are designed to guarantee the performance of coordinated control of trains.4.State-dependent topology weight-based coordinated control laws for high-speed trains under a time-varying switching communication network are proposed.Consider-ing the impact of state constraints on the coordinated operation process of trains,a multi-agent like model is first constructed for the dynamics of a train,and by taking account of the state constraints caused by the safety stretching/compressing range of the train couplers,a state-dependent topology weight-based control strategy is proposed.Further,the proposed state-dependent topology weight-based control strategy is extended to the coordinated control of multiple train system,and considering the constraints caused by the limited train-to-train wireless communication range and by the train control signaling system,state-dependent weight-based distributed coordinated control laws are designed such that consensus among trains at the desired operating curve is achieved while avoid-ing triggering emergency braking.5.Distributed anti-saturation coordinated control laws are designed for high-speed trains subject to actuator saturation.In view of the impact of switching communication network topology on the cooperative operation process of the trains,a topology switching rule for the communication network is established by analyzing the principles followed by the trains in the information exchange process thoroughly.Considering the single-particle nonlinear model of a train,distributed anti-saturation coordinated control laws are designed based on state-dependent topology weights and an anti-saturation observer like function.Further,considering the safety tracking interval constraints,a collision-avoidance condition that the initial states of trains should meet is proposed to ensure the safety of the train cooperative operation process.