In order to improve the accuracy of the speed control of the shield hydraulic propulsion system and reduce the deviation between the mathematical model and the actual system, a dynamic system parameter identification method based on the time delay neural network and the upper and lower bounds of the system parameters is proposed. On the basis of establishing the speed control model of shield hydraulic propulsion system, the neural network is used to identify the indefinite parameters of the system within the range of parameters to be identified, and search for a set of parameters closer to the actual shield propulsion system to meet the system input. The actual system output can be accurately obtained for the actual signal. Finally, another set of sampled data of the shield simulation test bench is used to verify the system identification results. The results show that the model obtained by the identification method can approximate the real physical system.