摘要
当下我国城市轨道交通正处于兴旺发展的关键时期,然而我国尚未形成完善的城市轨道交通线网合理规模计算模型和理论,因此有必要深入探究城市轨道交通线网合理规模的计算方法。首先,从国家政策与交通战略、城市经济社会发展水平、城市规模与形态、交通需求与供给等方面,剖析城市轨道交通线网规模的影响因素,提取城镇常住人口、建成区面积、GDP、全网日均客运量、城市轨道交通占公共交通分担率和线网负荷强度6个量化指标。其次,利用我国21座城市的历年城市轨道交通线网规模影响指标数据,构建基于BP神经网络算法的城市轨道交通线网合理规模计算模型。最后,以上海市、深圳市、重庆市、长沙市和南宁市为例,应用该模型对五市2035年的线网合理规模进行预测,以期为我国城市轨道交通线网规划提供科学的依据,促进我国城市交通可持续发展。
At present,the urban rail transit is in the critical period of vigorous development in China.However,the rational scale calculation model and theory of urban rail transit network have not been formed yet in China.Therefore,it is necessary to further explore the calculation method of the rational scale of urban rail transit network.Firstly,the influence factors of urban rail transit network scale are expounded from the aspects of the national policy and transportation strategy,the urban economic and social development level,the urban scale and form,and the transportation demand and supply.The town population,urban built-up area,gross domestic product(GDP),average daily passenger volume of the whole network,share ratio of urban rail transit in public transport and network load intensity are extracted as six quantitative indicators.Then,employing the impact indicator data of urban rail transit network scale in 21 cities of China over the years,a rational scale calculation model of urban rail transit network based on BP neural network is established.Finally,taking Shanghai,Shenzhen,Chongqing,Changsha and Nanning as examples,the model is used to predict the rational scale of urban rail transit network in 2035 in order to provide the scientific evidence for urban rail transit network planning and to promote the sustainable development of urban transportation in China.
作者
张依婧
张毅
汪涛
谢乐龙
ZHANG Yijing;ZHANG Yi;WANG Tao;XIE Lelong
出处
《城市道桥与防洪》
2023年第3期230-235,M0021,M0022,共8页
Urban Roads Bridges & Flood Control
基金
国家社会科学基金一般项目(18BSH143)
上海市科技创新行动计划(20DZ1202900、21692106700)。
关键词
城市轨道交通
线网规模
影响指标
BP神经网络模型
urban rail transit
network scale
impact indicators
back propagation(BP)neural network model
