Modelling and control design for an electro-pneumatic braking system in trains with multiple locomotives

This paper focuses on the modelling and control design for an electro-pneumatic braking system used in multi-locomotives in order to achieve accurately and steadily braking control of heavy haul trains. To deal with various time delays, a T-S fuzzy model based on satisfaction degree is proposed to simplify the cylinder model construction and a fuzzy clustering algorithm with forgetting factors is deployed to achieve parameter selflearning in order to improve the fuzzy control accuracy. Then a fuzzy genetic algorithm is adopted as the rolling optimization method to reduce the effect of coupling noise, system disturbance and communication random interference such that the system robustness and controller response capability are improved. The effectiveness of the proposed method is verified by simulation and practical implementations.