考虑服务水平的城市轨道交通牵引能耗研究
|
摘要
轨道交通牵引能耗除了受敷设方式、车型、操纵方式等因素影响外,还与列车满载率、开行间隔等服务水平有关。为明晰服务水平与牵引能耗之间的关系,在保证服务水平的前提下给列车节能开行方案合理编制提供参考,以轨道交通线路牵引能耗为研究对象,建立考虑服务水平的牵引能耗模型,并分析一定服务水平下线路的节能潜力。在定性分析服务水平与能耗关系的基础上,提取面向能耗评估相关的服务水平指标并进行等级划分;基于服务水平指标与能耗强度的相关性分析结果,选取环境温度和满载率作为基于支持向量回归的牵引能耗模型的输入,并应用北京地铁历史数据对模型进行标定与验证;以服务水平为约束条件,对线路的节能潜力进行分析。研究结果显示,与既有统计模型相比,提出的模型测算精度较高,平均绝对百分误差为1.5%;在保证服务水平的条件下,优化不同时段列车开行间隔可降低6.04%的牵引能耗。
Traction energy consumption(TEC) of urban rail transit is usually affected by line laying mode, train type, train operation mode, as well as its level of service(LOS)(e.g., load factor of train and interval between trains). In order to understand the relationship between LOS and TEC, and provide a reference for the rational development of energy-saving train operation plan without lowering LOS, this study proposed a TEC model considering LOS indicators and analyzed the energy saving potential under a certain LOS based on the study of TEC of urban rail transit line. First, the LOS indicators for energy assessments were extracted and classified on the basis of qualitative analysis of the relationship between LOS and energy consumption. Second, based on the analysis of the correlation between extracted LOS indicators and energy consumption strength, two indicators including environment temperature and load factor were selected as the inputs of support vector regression-based TEC model, which was estimated and validated with the data collected from Beijing urban rail system. Finally, the energy saving potential was analyzed with LOS as the constraint condition. The results show the proposed model has higher measurement accuracy with mean absolute percentage error of 1.5% compared with existing statistical models. Moreover, without lowering LOS, the TEC can be reduced by 6.04% through the optimization of train interval for different time periods in a day.
Traction energy consumption(TEC) of urban rail transit is usually affected by line laying mode, train type, train operation mode, as well as its level of service(LOS)(e.g., load factor of train and interval between trains). In order to understand the relationship between LOS and TEC, and provide a reference for the rational development of energy-saving train operation plan without lowering LOS, this study proposed a TEC model considering LOS indicators and analyzed the energy saving potential under a certain LOS based on the study of TEC of urban rail transit line. First, the LOS indicators for energy assessments were extracted and classified on the basis of qualitative analysis of the relationship between LOS and energy consumption. Second, based on the analysis of the correlation between extracted LOS indicators and energy consumption strength, two indicators including environment temperature and load factor were selected as the inputs of support vector regression-based TEC model, which was estimated and validated with the data collected from Beijing urban rail system. Finally, the energy saving potential was analyzed with LOS as the constraint condition. The results show the proposed model has higher measurement accuracy with mean absolute percentage error of 1.5% compared with existing statistical models. Moreover, without lowering LOS, the TEC can be reduced by 6.04% through the optimization of train interval for different time periods in a day.
引文
[1] 陈荣武,刘莉,郭进.基于遗传算法的列车运行能耗优化算法[J].交通运输工程学报,2012,12(1):108-114.CHEN Rongwu,LIU Li,GUO Jin.Optimization Algorithm of Train Operation Energy Consumption Based on Genetic Algorithm[J].Journal of Traffic and Transportation Engineering,2012,12(1):108-114.
[2] 杨利军,胡用生,孙丽霞.城市轨道交通节能线路仿真算法[J].同济大学学报(自然科学版),2012,40(2):235-240.YANG Lijun,HU Yongsheng,SUN Lixia.Energy-saving Track Profile of Urban Mass Transit[J].Journal of Tongji University(Natural Science),2012,40(2):235-240.
[3] 宿帅,唐涛.城市轨道交通ATO的节能优化研究[J].铁道学报,2014,36(12):50-55.SU Shuai,TANG Tao.Optimal Train Control for ATO System [J].Journal of the China Railway Society,2014,36(12):50-55.
[4] RUELLAND F,AL-HADDAD K.Reducing Subway’s Energy[C]//Electrical Power Conference.Montreal:IEEE,2007:261-267.
[5] 王子甲,陈峰,施仲衡.北京城市轨道交通中远期能耗预测研究[J].中国铁道科学,2013,34(3):133-136.WANG Zijia,CHEN Feng,SHI Zhongheng.Prediction on Medium and Long-term Energy Consumption of Urban Rail Transit Network in Beijing[J].China Railway Science,2013,34(3):133-136.
[6] 陈峰,杨洋,刘欧阳.城市轨道交通牵引能耗估算方法[J].都市快轨交通,2014,27(2):90-93.CHEN Feng,YANG Yang,LIU Ouyang.Estimation Methods for Traction Energy Consumption of Urban Rail Transit[J].Urban Rapid Rail Transit,2014,27(2):90-93.
[7] HICKMAN J,HASSEL D,JOUMARD R,et al.Methodology for Calculating Transport Emissions and Energy Consumption[R].Luxembourg:Office for Official Publications of the European Communities,1999.
[8] CORTES C,VAPNIK V.Support-vector Networks[J].Machine Learning,1995,20(3):273-297.
[9] SCH?LKOPF B,SMOLA A J.Learning with Kernels:Support Vector Machines,Regularization,Optimization,and Beyond[M].London:MIT Press,2002.
[10] 陈光,张宁,陈晖,等.城市轨道交通服务水平评价体系研究[J].都市快轨交通,2008,21(6):5-9.CHEN Guang,ZHANG Ning,CHEN Hui,et al.The Evaluation System of LOS for Urban Rail Transit [J].Urban Rapid Rail Transit,2008,21(6):5-9.
[11] 向兵,孙有望.城市轨道交通乘客服务水平评价方法研究[J].城市轨道交通研究,2008,11(9):15-18.XIANG Bing,SUN Youwang.An Evaluation Index System of Passenger Service Performance of URT[J].Urban Mass Transit,2008,11(9):15-18.
[12] 沈景炎.关于车辆定员与拥挤度的探析[J].都市快轨交通,2007,20(5):14-17.SHEN Jingyan.On the Carriage Passenger Capacity and the Crowdedness Involved[J].Urban Rapid Rail Transit,2007,20(5):14-17.
[13] 王凡.基于支持向量机的交通流预测方法研究[D].大连:大连理工大学,2010.
[14] 袁政.公共管理定量分析:方法与技术[M].重庆:重庆大学出版社,2009:76.
[15] 张厚粲,徐建平.现代心理学和统计学[M].北京:北京师范大学出版社,2009:150.
[16] OSUNA E,FREUND R,GIROSI F.An Improved Training Algorithm for Support Vector Machines[C]// Neural Networks for Signal Processing.Amecia Island:IEEE,1997:276-285.
[17] 王雪然.考虑服务水平的城市轨道交通节能开行方案优化研究[D].北京:北京交通大学,2015.
[2] 杨利军,胡用生,孙丽霞.城市轨道交通节能线路仿真算法[J].同济大学学报(自然科学版),2012,40(2):235-240.YANG Lijun,HU Yongsheng,SUN Lixia.Energy-saving Track Profile of Urban Mass Transit[J].Journal of Tongji University(Natural Science),2012,40(2):235-240.
[3] 宿帅,唐涛.城市轨道交通ATO的节能优化研究[J].铁道学报,2014,36(12):50-55.SU Shuai,TANG Tao.Optimal Train Control for ATO System [J].Journal of the China Railway Society,2014,36(12):50-55.
[4] RUELLAND F,AL-HADDAD K.Reducing Subway’s Energy[C]//Electrical Power Conference.Montreal:IEEE,2007:261-267.
[5] 王子甲,陈峰,施仲衡.北京城市轨道交通中远期能耗预测研究[J].中国铁道科学,2013,34(3):133-136.WANG Zijia,CHEN Feng,SHI Zhongheng.Prediction on Medium and Long-term Energy Consumption of Urban Rail Transit Network in Beijing[J].China Railway Science,2013,34(3):133-136.
[6] 陈峰,杨洋,刘欧阳.城市轨道交通牵引能耗估算方法[J].都市快轨交通,2014,27(2):90-93.CHEN Feng,YANG Yang,LIU Ouyang.Estimation Methods for Traction Energy Consumption of Urban Rail Transit[J].Urban Rapid Rail Transit,2014,27(2):90-93.
[7] HICKMAN J,HASSEL D,JOUMARD R,et al.Methodology for Calculating Transport Emissions and Energy Consumption[R].Luxembourg:Office for Official Publications of the European Communities,1999.
[8] CORTES C,VAPNIK V.Support-vector Networks[J].Machine Learning,1995,20(3):273-297.
[9] SCH?LKOPF B,SMOLA A J.Learning with Kernels:Support Vector Machines,Regularization,Optimization,and Beyond[M].London:MIT Press,2002.
[10] 陈光,张宁,陈晖,等.城市轨道交通服务水平评价体系研究[J].都市快轨交通,2008,21(6):5-9.CHEN Guang,ZHANG Ning,CHEN Hui,et al.The Evaluation System of LOS for Urban Rail Transit [J].Urban Rapid Rail Transit,2008,21(6):5-9.
[11] 向兵,孙有望.城市轨道交通乘客服务水平评价方法研究[J].城市轨道交通研究,2008,11(9):15-18.XIANG Bing,SUN Youwang.An Evaluation Index System of Passenger Service Performance of URT[J].Urban Mass Transit,2008,11(9):15-18.
[12] 沈景炎.关于车辆定员与拥挤度的探析[J].都市快轨交通,2007,20(5):14-17.SHEN Jingyan.On the Carriage Passenger Capacity and the Crowdedness Involved[J].Urban Rapid Rail Transit,2007,20(5):14-17.
[13] 王凡.基于支持向量机的交通流预测方法研究[D].大连:大连理工大学,2010.
[14] 袁政.公共管理定量分析:方法与技术[M].重庆:重庆大学出版社,2009:76.
[15] 张厚粲,徐建平.现代心理学和统计学[M].北京:北京师范大学出版社,2009:150.
[16] OSUNA E,FREUND R,GIROSI F.An Improved Training Algorithm for Support Vector Machines[C]// Neural Networks for Signal Processing.Amecia Island:IEEE,1997:276-285.
[17] 王雪然.考虑服务水平的城市轨道交通节能开行方案优化研究[D].北京:北京交通大学,2015.