Study on high-speed train ATP based on fuzzy neural network predictive control

In the context that the target-speed control mode is commonly used in train control of high-speed railways in China, the predictive control based on fuzzy neural networks was applied to ATP of high-speed railways in view of the operation requirements of high-speed trains. The fuzzy neural network model for predictive control of high-speed train speeds by blocking sections was established. In a block section, the control information was sent to the train control center by communication between train and ground; from the acquired information, the automatic protection curve corresponding to the train speeds from the present position to the block section exit was obtained with the predictive control algorithm, and the train operation mode and control strategy were determined; within each communication period, optimization of the train speeds was achieved by rolling optimization and error correction. The simulation results show that, compared to traditional control methods, predictive control based on fuzzy neural networks brings about better performance of safety for automatic train protection of high-speed trains.