高速列车模糊RBF神经网络速度跟踪控制

Speed Tracking Control for High-Speed Trains with Fuzzy RBF Neural Network

速度跟踪控制是列车自动驾驶的一个重要研究方向,但现有速度跟踪控制的性能尚需改进。为此,本文探究了基于模糊RBF神经网络(FRNN)的高速列车速度跟踪控制。建立了高速列车单位移多质点模型,结合反馈控制和FRNN设计了高速列车速度跟踪控制器,首先探究了该网络的层次结构和每层功能,再用该网络持续更新PID参数,并由此计算牵引力和制动力。最后进行仿真试验,并对三种方法进行了仿真对比,通过仿真证明了该方法优越的速度跟踪性能,该方法用在列车跟踪控制中效果良好。

Speed tracking control is an important direction of studying in the field of automatic train operation,but the perfor⁃mance of existing speed tracking control needs to be improved.Therefore,this paper explores a speed tracking control for a high-speed train based on a Fuzzy RBF Neural Network(FRNN).A multi-particle unit-displacement model for the high-speed train is built,and a speed tracking controller for high-speed trains is designed by combining feedback control and FRNN.The four-layer architecture of the FRNN and the function of each layer are firstly explored.Then,the PID parameters are constantly modified,and the traction force and braking force are determined by the FRNN.Lastly,numeral simulation tests are conducted with a scientific computer.Also,com⁃parative studies of three approaches are completed.Simulation results prove that this approach has superior speed tracking perfor⁃mance.The method is quite effective for train tracking control.