直接模糊神经控制及其在列车自动操纵中的应用(英)  

Direct Fuzzy Neural Control with Application to Automatic Train Operation

作　　者：王晶[1] 蔡自兴[2] 贾利民[3] 

机构地区：[1]中国科学院计算技术研究所,北京100080 [2]中南工业大学智能控制研究所,长沙410083 [3]中国铁道科学研究院通信信号研究所,北京100081

出　　处：《控制理论与应用》1998年第3期391-399,共9页Control Theory & Applications

摘　　要：本文基于模糊神经控制方法提出一种新的列车自动操纵策略．首先基于复杂动态过程划分的思想，建立列车运行过程的五个子过程近似工程数学模型；接着对所提的模糊神经控制器的结构、训练样本编码方法、学习算法和推理准则作了详细的描述；文末利用所提控制方法对列车在一段特定的线路上作了仿真研究，结果表明了所提方法的有效性．This paper proposes a novel methodology to perform the automatic train operation based on direct fuzzy neural control scheme which is functionally equivalent to the conventional fuzzy controller. Firstly,based on the idea of the partition of the complex process,the mathematic description of train traveling process is provided. Then, the structure, training sample coding method,learning algorithm and inference criteria of the proposed fuzzy neural controller are described in details. Finally,a group of simulation results for the train traveling procesS are compared with the human driver's control. The results have demonstrated the effectiveness of the proposed approach.

关 键 词：列车自动操纵 列车运行 模糊神经控制 

分 类 号：U284.482[交通运输工程—交通信息工程及控制] TP13[交通运输工程—道路与铁道工程]