Abstract: Train control center and interlocking integrated system is in the experimental stage in China at present.It is of great significance to study the risk assessment of the train control center and interlocking integrated system for its reliable operation.To solve the problems of uncertainty and risk changing with time in the risk assessment of train control center and interlocking integrated system, a risk assessment method of train control center and interlocking integrated system based on evidence theory and dynamic Bayesian network is proposed.Firstly, the fault tree of train control center and interlocking integrated system is constructed and transformed into dynamic Bayesian network.Then, the fuzzy number is introduced to determine the failure rate of the basic event, and the state that expresses the cognitive uncertainty is introduced into the new state space through the evidence theory to determine the parameters of the dynamic Bayesian network.The key risk factors affecting the reliable operation of the system are found out by using the bidirectional reasoning, importance and sensitivity analysis of dynamic Bayesian network.Finally, the expert scoring method is used to quantify the risk loss, and the risk level of the train control center and interlocking integrated system is determined by drawing the risk matrix.The simulation results show that the risk level of the Train control center and interlocking integrated system belongs to the "IV（acceptable risk）" in the railway safety standard EN 50126.The communication interface with switch machine and track circuit and the object controller unit in the communication unit are the weak links of the system, so it is necessary to strengthen the risk control of the communication unit and the object controller unit.