密集城镇群客流分布双层最大熵模型

Double Maximum Entropy Distribution Model for Passenger Flow of Dense Urban Group

为了满足密集城镇群客流分布预测的需求,为密集城镇群规划及建设提供更加合理的参考,结合密集城镇群客流分布预测对应的研究区域空间跨度较大的特点,将研究区域分为市区、市区外区域2个层级,分别进行交通小区划分。采用适用性较广的最大熵模型进行客流分布预测,同时考虑到空间距离对密集城镇群客流分布的影响权重较大,借鉴重力分布模型的思想,引入广义出行费用反映交通阻抗,对最大熵模型进行修正;并分别从市区、市域2个层级构建双层最大熵模型。采用双层最大熵模型对东莞市2020年客流分布进行了预测。预测结果表明:双层最大熵模型能够减少空间范围过大的不利影响,同时双层最大熵模型相对独立,可以互为补充,更加适用于密集城镇群的客流分布预测。

In order to meet the demand of passenger flow prediction and provide reasonable reference for planning and construction of dense urban group, with consideration of large space span of study areas corresponding to passenger flow distribution prediction, the study areas were divided into city areas and outside city areas and their traffic zones were also divided respectively. Maximum entropy model was used to predict the distribution of passenger flow, and with consideration of the great impact of spatial distance on passenger flow distribution of dense urban group, the idea of gravity distribution model was referred and the generalized travel cost was introduced to reflect the traffic impedance and correct maximum entropy model. Double maximum entropy model was constructed from city and region. Double maximum entropy model was used to predict passenger flow distribution of Dongguan city in 2020. The results show that double maximum entropy model can reduce the adverse effect of large space range, and two maximum entropy models are relatively independent, so they can complement each other, which makes it more suitable for passenger distribution prediction of dense urban group.