基于DP-SQP方法的列车动态编组优化控制  

Train Dynamic Coupling Optimization Control Based on Dynamic Programming-Sequential Quadratic Programming Method

作　　者：李建雄 王海峰[1,2,3] 张启鹤 柴铭 LI Jianxiong;WANG Haifeng;ZHANG Qihe;CHAI Ming(School of Electronics and Information Engineering,Beijing Jiaotong University,Beijing 100044,China;National Engineering Research Center of Rail Transportation Operation and Control System,Beijing Jiaotong University,Beijing 100044,China;Beijing Laboratory of Urban Rail Transit,Beijing Jiaotong University,Beijing 100044,China)

机构地区：[1]北京交通大学电子信息工程学院,北京100044 [2]北京交通大学轨道交通运行控制系统国家工程研究中心,北京100044 [3]北京交通大学城市轨道交通北京实验室,北京100044

出　　处：《铁道标准设计》2022年第4期168-174,187,共8页Railway Standard Design

基　　金：先进轨道交通专项课题(I18B500050);北京市自然科学基金“轨道交通联合”(L181005);国家自然科学基金项目(61473029);国家重点基础研究计划973课题(2014CB340703);城市轨道交通北京实验室科研课题(2012-07)。

摘　　要：动态编组逐渐成为列车运行控制领域的研究热点之一,低牵引能耗、高编组精确性、强抗干扰能力的多目标控制是实现动态编组的难题。针对这一问题,提出一种基于动态规划(DP)-序列二次规划(SQP)的优化控制方法。首先,对动态编组过程建立数学模型;然后,建立以低能量消耗和精确编组为目标的动态编组优化问题,在离线状态下使用动态规划算法计算速度曲线;最后,以离线速度结果为基础,使用序列二次规划算法求解最优控制策略在线控制列车。仿真实验表明,使用本文方法控制编组过程时,牵引能耗降低5.6%,编组误差下降至3 m,SQP算法求解的迭代次数下降到12次;在有外界干扰状态下,该方法可在线控制列车校正速度偏离,消除干扰带来的影响。Dynamic coupling has gradually become one of the research hotspots in the field of train operation control.Multi-objective control with low traction energy consumption,high coupling accuracy and strong anti-interference capability is a challenge to realize dynamic coupling.To solve this problem,this paper proposes an optimization control method based on dynamic programming-sequential quadratic programming.Firstly,the dynamic coupling process is described by mathematical model.Secondly,dynamic coupling optimization with low energy consumption and high coupling accuracy is fulfilled,and the speed curve is calculated offline using dynamic programming algorithm.Finally,based on the offline speed results,the optimal online control strategy which is calculated by sequential quadratic programming algorithm is used for train control.The simulation results show that,with this method to control coupling process,the traction energy consumption is lowered by 5.6%,the coupling error is reduced to 3 meters and the iterations of SQP algorithm is cut down to 12.Moreover,under the condition of external interference,the method can control the train online,correct the speed deviation and eliminate the influence of interference.

关 键 词：列控系统 动态编组 多目标优化 动态规划 序列二次规划 

分 类 号：U283[交通运输工程—交通信息工程及控制] U279[交通运输工程—道路与铁道工程]