Train Velocity Tracking Control Based On Sliding Mode

ABSTRACT:The demands of urban transportation are increasing along with rapid economy expansion. The population is getting larger and cities are becoming more crowded than before. Urban rail transit which has advantages of rapidity, safety and comfort can make contribution to the problems. It is convenient and easy for people to travel by taking urban rail transit and it is also affordable and energy-efficient for operating companies. Train operation control system is the key to ensure safety and effectivity of urban rail transit. Automatic train operation system can adjust train’s velocity in real time and track desired velocity accurately. As train operation velocity increasing and operation interval shortened, traditional control algorithm cannot meet the need of modern urban rail transit development. Hence, it is important and necessary to research on automatic train control algorithms. The paper investigates train operation velocity tracking algorithms and develops simulation system based on physical processor. Detailed contents are shown as follows:Firstly, train velocity tracking under uncertain conditions is studied. Taking advantage of sliding mode observer’s estimation performance, train velocity tracking algorithm is designed with single-point-mass model to estimate total resistance the train takes which can eliminate adverse factor of the uncertain. Controller parameters are optimized by particle swarm optimization algorithm. Simulation results show that the velocity tracking and position tracking are all well guaranteed, the system shows robustness when dealing with uncertain operational environments.Secondly, train velocity tracking under time-varying disturbance is studied. Train model with disturbance terms is established and the range of time-varying disturbance is determined. Train velocity tracking algorithm based on fast terminal sliding mode is designed, which utilizes the good performance of sliding mode control such as fast response and robustness. Simulation results show the robustness of control algorithm when applying it to train dynamic model with time-varying and mass-varying disturbance.Finally, train operation velocity tracking simulation system is designed by using physical processor as controller on which runs the algorithms. Single-point-mass model of train is built. Database of lines, trains and operation scenarios is designed to support simulations. The effect of system shows the simulation system is able to simulate with different lines and scenarios and to create testing environment for comparing different control algorithms. The systems are commonly flexible and are provided with actual application value.