Organization Name:CRRC Industrial Research Academy Co., Ltd

Project introduction:

In the field of rail transit, the design standards and operating environment of different levels of lines are quite different, and the impact on train operation stability is very obvious. Therefore, it is of great significance to select appropriate vibration damping parameters according to the characteristics of the actual running line to maintain the stability of the train running across the line.

Based on the traditional suspension system, the adaptive suspension control system of rail vehicles realized the continuous adjustable damping of the suspension system. The biggest innovation of this system is that the reinforcement learning algorithm has been used to realize the adaptive control function of vehicle stability. Through the independent decision-making of the system, the optimal vibration damping parameters of the suspension system were given, and the vibration damping actuator was controlled to act to actively maintain the stable running state of the vehicle.

This product has been assembled in high-speed EMU and a large number of line operation tests have been carried out. By means of comparison test, the running stability index of the vehicle body under active control state was compared with that under passive state, and it has been verified that the comprehensive stability index of the vehicle was improved by about 11%.

The adaptive suspension control system adjusts the parameters adaptively according to the control effect, and finally adapts to the new operating environment. Therefore, the system has universality under the universal scenario. In addition to being used for the suspension system control of rail transit vehicles, it can also be synchronously promoted to multiple scenarios such as train traction control, braking control, vehicle suspension system control, and adaptive control of industrial manipulator action.