摘要
由于城轨日常行车对区间运行时分的多样性需求,不同运行时分下区间列车运行策略的优化求解极为困难。通过丰富的实际列车运行数据拟合,获取运行时分−最优能耗间的函数关系。将区间内每一列车运行策略对应的运行时分和能耗值作为数据点,根据数据特点构建劣点判别规则从原始数据中剔除能耗值较高的不合理数据,筛选得到最优拟合数据点集。提出基于遗传算法的B样条拟合方法,结合遗传算法寻求B样条曲线的最优参数向量和控制顶点数,拟合得到运行时分-最优能耗值函数曲线。通过某城市地铁区间的实际列车运行数据进行验证,证明本文提出的拟合方法精度高、速度快,绘制出的运行时分-最优能耗值函数曲线能够为城轨列车运行组织提供依据。
Due to the complexity of the operation control of urban rail transit,it is very difficult to optimize the train operation strategy in different running times.This paper studied the functional relationship between running time and optimal energy consumption by abundant actual train operation data.Taking the running time and energy consumption corresponding to each train operation strategy as data points,the discriminant criterion of outliers was constructed to select the optimal fitting data set from the original data set.Furthermore,a B-spline curve fitting method combined with a Genetic Algorithm was proposed,which is proven to have high fitting accuracy and convergence speed.The validation of Guangzhou Metro’s actual operation data of one certain section shows that the energy-running-time function curve fitted by our method has lower energy and could be used for the operation of an urban rail train.
作者
邓连波
梅宏达
周文梁
静恩伟
DENG Lianbo;MEI Hongda;ZHOU Wenliang;JING Enwei(School of Traffic and Transportation Engineering,Central South University,Changsha 410075,China)
出处
《铁道科学与工程学报》
CAS
CSCD
北大核心
2021年第9期2461-2468,共8页
Journal of Railway Science and Engineering
基金
国家自然科学基金资助项目(U1934216,71871226)
广州地铁集团有限公司研究基金资助项目(HT170235)。
作者简介
通信作者:邓连波(1977-),男,辽宁昌图人,教授,博士,从事交通运输规划与管理研究,E−mail:lbdeng@csu.edu.cn。
