城市轨道交通客流短时预测方法与运营编组优化设计

发布时间：2018-11-26 08:03

【摘要】：随着我国城市化进程的加快,城市人口急剧增长,交通压力不断增大,城市道路拥挤不堪,给市民的正常出行带来了极大不便,城市交通问题日益突出。要解决这一难题,不能把希望仅仪寄托在公路建设上,发展城市轨道交通才是应对城市交通拥堵的好方法。目前我国的城市轨道交通建设正处在一个前所未有的蓬勃发展时期,快速发展的同时也带来了很多问题,主要有城市轨道交通客流预测不准确以及城市轨道交通列车编组形式不合适等。由于以往常规的客流预测不准确使得以其为基础的运营编组设计不合适,从而导致了现阶段城市轨道交通拥挤不堪或者运能浪费,进而引起城市轨道交通运营成本的增加。针对此问题本文在已有研究的基础上,总结出城市轨道交通客流具有时变性、均衡性以及周期性变化等特点,构造了基于灰色预测模型和神经网络模型的城市轨道交通客流短时预测组合模型。利用神经网络模型来修正灰色预测模型的残差,两种模型互补对于城市轨道交通断面客流短时预测具有一定的合理性和参考性,可以作为城市轨道交通运营编组设计优化的基础。参考智能交通信号灯的原理,根据城市轨道交通短时预测的实时断面客流量进行运营编组设计,更贴近客流的实际客流情况,具有实时性、灵活性和快速响应性。将城市轨道交通客流以一周为一个周期,使用最近一周的历史断面客流作为训练样本,应用嵌入式灰色神经网络组合模型进行短时预测,即可得到下一周期的断面客流短时预测量,得到的断面客流量更符合客流不断变化的趋势。在此基础上进行运营编组设计,使得城市轨道交通更能适应客流量的实时变化,可以满足不断变化的客流需求。然后将按照计划运营所得到的实际客流归入历史客流,进行更新修正,作为下一周期客流短时预测及运营编组设计的基础。通过基于短时预测的城市轨道交通运营编组优化,在一定程度上可以提高城市轨道交通系统运能,提升运营效率,降低运营成本。
[Abstract]:With the acceleration of urbanization in China, the rapid growth of urban population, increasing traffic pressure, urban road congestion, to the normal travel of citizens has brought great inconvenience, urban traffic problems are increasingly prominent. In order to solve this problem, the hope should not only be placed on the highway construction, but also the development of urban rail transit is a good way to deal with urban traffic congestion. At present, the construction of urban rail transit in our country is in an unprecedented period of vigorous development. The rapid development has also brought many problems at the same time. The main problems are that the forecast of urban rail transit passenger flow is not accurate and the form of train formation is not suitable. Because of the inaccuracy of the routine passenger flow prediction in the past, the operational marshalling design based on it is not suitable, which leads to the overcrowded or wasteful urban rail transit at the present stage, which leads to the increase of the operation cost of the urban rail transit. In this paper, based on the existing research, the characteristics of urban rail transit passenger flow are summarized, such as time-varying, equilibrium and periodic change, etc. Based on grey prediction model and neural network model, the combined model of short time forecast of urban rail transit passenger flow is constructed. The neural network model is used to correct the residual error of the grey prediction model. The two models complement each other and have some rationality and reference for the short-term passenger flow prediction of urban rail transit section. It can be used as the basis for the optimization of urban rail transit operation marshalling design. According to the principle of intelligent traffic signal light and according to the real-time section passenger flow forecast of urban rail transit, the operation marshalling design is carried out, which is closer to the actual passenger flow situation of passenger flow, and has the characteristics of real-time, flexibility and quick response. Taking the urban rail transit passenger flow as a cycle, using the historical section passenger flow of the last week as the training sample, the embedded grey neural network combination model is used for short-term prediction. The short-term prediction of cross-section passenger flow in the next cycle can be obtained, and the obtained cross-section passenger flow is more in line with the changing trend of passenger flow. On this basis, the operation marshalling design is carried out to make the urban rail transit more adaptable to the real-time change of the passenger flow and to meet the changing demand of the passenger flow. Then the actual passenger flow according to the planned operation is classified into the historical passenger flow and updated and revised as the basis for the short-term prediction of passenger flow in the next cycle and the design of operational marshalling. Through the optimization of urban rail transit operation organization based on short-term prediction, the operation capacity of urban rail transit system can be improved to a certain extent, the operation efficiency can be improved, and the operation cost can be reduced.
【学位授予单位】：大连交通大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：U293.5;U293.13

【参考文献】