Train Braking Deceleration Control Based on Improved Smith Estimator

Abstract:

In view of the problem of the poor control in traditional proportion integration differentiation （PID） control and the classic Smith prediction method in train deceleration control， the feedback structure of Smith estimator was improved， and the uncertain factors that affect the actual deceleration of the train are classified from the perspective of control system. A single mass point simulation model for train braking was established， and its step response characteristics under the control of PID， Smith estimator and improved Smith estimator were analyzed. The simulation results show that the improved Smith estimator has a better control performance and a higher robustness. Finally， based on the software in the loop simulation， the control effects of non-deceleration control， deceleration control based on improved Smith estimator method， and deceleration control based on adaptive parameter estimation algorithm were compared. The results show that the deceleration control based on the improved Smith estimator can not only achieve accurate tracking of the target deceleration， but also significantly reduce the longitudinal impact of the train.