城市轨道交通延伸线客流预测研究

发布时间：2018-11-11 16:49

【摘要】：我国的城市轨道交通建设正处于高速发展的阶段,本着提高运营效益的原则,许多线路实行“分期建设、分期运营”的修建策略。随着线路后期工程的陆续开工建设,很多轨道交通线路将在原有工程的基础上逐步延伸,因此各个城市中轨道交通延伸线越来越多地出现。而线路延伸后,往往会对原有线路的客流分布产生较大影响,甚至需要调整线路运营方案。目前在轨道交通客流预测方面的研究多面向线网中长期的预测,而对线路延伸后客流变化的趋势以及延伸线的客流预测缺乏关注。针对这一现状,本文在已有研究的基础上,展开轨道交通延伸线的客流预测研究。本文首先分析了延伸线开通对既有线路客流影响的机理,并以西安地铁2号线为例对线路延伸后客流变化的规律进行了总结,研究了线路延伸后车站客流和站间客流分布的变化特征,为延伸线客流预测方法的建立提供理论依据。其次,本文提出了延伸线车站客流预测的方法,采用基于圈层变量的BP神经网路模型预测延伸线上新增站点进出站客流,并在新增站点客流预测结果的基础上对既有线路上的站点进出站客流进行预测,然后采用总量控制的方法调整全线站点的进出站客流量预测值。再次,本文通过引入非集计模型,提出了轨道交通乘客目的地站点选择模型,在分析站间OD分布影响因素的基础上确定了效用函数,建立了基于AFC数据的延伸线站间客流分布非集计预测方法,并研究了模型中参数的标定方法。最后,文章以西安地铁2号线为例,采用本文建立的车站客流预测模型和站间客流分布模型预测2号线南延伸线开通后的客流,将客流预测结果与实际运营客流进行对比分析,结果表明本文提出的预测模型相比传统方法能提高预测精度,对延伸线的客流预测有较好的适用性。
[Abstract]:The construction of urban rail transit in our country is in the stage of high speed development. In line with the principle of improving the operation benefit, many lines carry out the construction strategy of "construction by stages and operation by stages". With the construction of the railway line in the late stage, many rail transit lines will be extended gradually on the basis of the original project, so more and more rail transit extension lines appear in each city. After the extension of the line, the passenger flow distribution of the original line will often have a greater impact, and even need to adjust the line operation scheme. At present, the research on passenger flow prediction of rail transit is mostly oriented to the medium and long term forecast of line network, but the trend of passenger flow change after line extension and the passenger flow prediction of extension line are not paid attention to. In view of this situation, this paper, on the basis of the existing research, carries out the passenger flow prediction research of the rail transit extension line. This paper first analyzes the mechanism of the influence of the opening of the extended line on the passenger flow of the existing lines, and taking Xi'an Metro Line 2 as an example, summarizes the rules of passenger flow change after the extension of the line. The variation characteristics of passenger flow and passenger flow distribution between stations after line extension are studied in order to provide theoretical basis for the establishment of passenger flow prediction method. Secondly, this paper puts forward the method of passenger flow prediction of extension line station. The BP neural network model based on loop variables is used to predict the passenger flow of the new station in and out of the extension line. On the basis of the forecast result of the new station passenger flow, the passenger flow in and out of the station on the existing line is forecasted, and then the total quantity control method is adopted to adjust the forecast value of the incoming and outgoing station passenger flow of the whole line station. Thirdly, by introducing the disaggregate model, this paper puts forward the passenger destination station selection model of rail transit, and determines the utility function on the basis of analyzing the influencing factors of OD distribution between stations. A disaggregate forecasting method of passenger flow distribution between extended line stations based on AFC data is established, and the calibration method of parameters in the model is studied. Finally, taking Xi'an Metro Line 2 as an example, the passenger flow prediction model and the passenger flow distribution model between stations are used to predict the passenger flow after the opening of the south extension line of Line 2. The results of passenger flow prediction are compared with the actual passenger flow. The results show that the prediction model proposed in this paper can improve the prediction accuracy compared with the traditional method, and has a better applicability to the passenger flow prediction of the extended line.
【学位授予单位】：长安大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：U293.13

【相似文献】