Fuzzy uncertainty and dynamics are characteristics of failure rate in actual working environment of the motor train brake system. In this paper, the triangular fuzzy numbers and the extension principle are employed to represent the bottom event fault rate. By combining the dynamic discrete time Bayesian network model with the fault tree model, the fuzzy fault rate and the posterior probability of the root nodes of the dynamic Bayesian network nodes are obtained, solving the problem of reliability analysis of dynamic system under uncertain conditions. The method has been applied to analyze the operation of braking air supply system of Lanzhou passenger dedicated lines and the reliability prediction curves and the weak line of the system are obtained, which provides a theoretical basis for the maintenance plan of the braking system according to the fault characteristics.