当前位置:科学网首页 > 基金首页 > 基于迭代学习的城市轨道交通列车自动运行控制研究

国家自然科学基金项目查询

基于迭代学习的城市轨道交通列车自动运行控制研究

批准号61503180 学科分类微纳控制系统 ( F030209 )
项目负责人刘娣 负责人职称副教授 依托单位南京工程学院
资助金额19.00
万元
项目类别青年科学基金项目 研究期限2016 年 01 月 01 日 至
2018 年 12 月 31 日
中文主题词轨道交通;运行控制;迭代学习控制;满意控制;初态
英文主题词rail transit;operation control;iterative learning control;satisfactory control;initial state

摘要

中文摘要 完善的城市轨道交通列车控制系统是列车安全、舒适、低能耗运行的重要保证。本申请以列车运行速度跟踪控制系统为研究对象,以列车运行多指标优化控制为科学问题,提出一种基于迭代学习的前/反馈式列车速度跟踪控制算法。首先,研究一种基于趋近律的迭代初态偏差修正算法,为学习律的应用提供初始条件,同时,提出一种基于扰动观测器且增益可变的ILC算法,实现对初态误差的鲁棒性、抑制非重复性扰动及快速收敛;其次,对多指标约束下的列车速度满意控制算法进行研究,提取列车运行期望指标集并分析其相容性,研究控制器存在的充分性条件,求取满意解集;最后,实现两种控制算法的有效融合以弥补单一控制器的不足,并基于dSPACE半实物仿真系统对算法控制性能进行实验研究,评价算法有效性,为算法优化提供研究基础。本项目的理论研究紧密结合了研究对象的工程特性及指标要求,可为高速列车自动运行控制系统的实施提供方法论。
英文摘要 The perfect urban rail transit train control system is the important guarantee for the safety, comfortable and low energy consumption of the train operation. For the optimization of train operation indexes, feedforward-feedback control strategy of train speed tracking control system based on iterative learning control (ILC) is studied in this project. Firstly, the correction algorithm of initial state error based on reaching law is studied so as to provide the initial condition for the application of learning law. At the same time, an ILC algorithm with variable gain is proposed based on disturbance observer to obtain the robustness of initial state error, the inhibition of non repetitive disturbances and rapid convergence. Secondly, the train speed satisfactory control algorithm is studied in the multiple index constraints. The desired index set of train operation is obtained and its consistency is analyzed. The sufficient condition of a PID controller existing is studied and the satisfactory solution set is obtained. Finally, the effective integration of ILC algorithm and satisfactory PID algorithm is realized to make up for the deficiency of single controller. The performance of the control system is analyzed by dSPACE semi-physical simulation system and the algorithm effectiveness is evaluated to provide research foundation for algorithm optimization. The theoretical research of this project closely combines the engineering real characteristics and the index requirement so as to provide methodology for high speed train automatic operation control system implementation.
结题摘要

成果

序号 标题 类型 作者

关于我们| 网站声明| 服务条款| 联系方式| RSS| 中国科学报社 京ICP备14006957 京公网安备110402500057号
Copyright @ 2007- 中国科学报社 All Rights Reserved
地址:北京市海淀区中关村南一条乙三号   电话:010-62580783