山东省横向干扰路段实际通行能力模型研究  

Modelling of actual capacity of horizontally disturbed road sections in Shandong Province

作　　者：董昌乐 王修光 王雄 DONG Changle;WANG Xiuguang;WANG Xiong(School of Traffic and Logistics Engineering,Shandong Jiaotong College,Jinan 250357,Shandong,China;Shandong Zhengqu Traffic Engineering Co.,Ltd.,Jinan 250000,Shandong,China)

机构地区：[1]山东交通学院交通与物流工程学院,山东济南250357 [2]山东正衢交通工程有限公司,山东济南250000

出　　处：《农业装备与车辆工程》2024年第10期161-168,共8页Agricultural Equipment & Vehicle Engineering

基　　金：山东省科学技术厅新一代信息技术项目“混合增强智能‘交通大脑’新架构及关键技术研究”(2021TSGC1011);山东省工业和信息化厅新一代信息技术项目“混合增强人工智能‘交通大脑’新架构及关键技术研究”(202160101683);山东省工业和信息化厅新一代信息技术项目“数据驱动的城市快速路交通一体化诱导与协同控制”(202250101842)。

摘　　要：为明确横向干扰因素对公路路段通行能力的影响,实现通行能力计算模型本地化,针对山东省公路交通现状,以山东省横向干扰路段为研究对象,从交通流因素、道路基础设施因素、土地利用类型3个方面对公路基本路段通行能力横向干扰影响因素进行定性分析;采用皮尔逊相关性分析方法对横向干扰影响因素与影响系数进行相关性分析;根据相关性分析结果,选取建模自变量,分别通过SPSS、MATLAB分析软件构建基于多元线性回归和基于BP神经网络的公路基本路段通行能力横向干扰影响机理本地化模型。结果表明:基于BP神经网络的横向干扰路段通行能力本地化模型具有更高的精确性。In order to clarify the influence of lateral interference factors on the capacity of highway road sections and achieve the localisation of the capacity calculation model,in view of the current situation of highway traffic in Shandong Province,and taking the lateral interference road sections in Shandong Province as the research object,the lateral interference influencing factors of the capacity of basic road sections were qualitatively analysed from the three aspects of the traffic flow factors,the road infrastructure factors,and the type of land use,and Pearson's correlation was used to analyse the influence factors of horizontal interference on the capacity of basic road sections.The Pearson correlation analysis method was used to correlate the influence factors and influence coefficients of lateral interference,and according to the results of the correlation analysis,the independent variables of modelling were selected,and the localization model of the influence mechanism of lateral interference of the capacity of the basic road sections based on multivariate linear regression and BP neural network was constructed through the analytical software of SPSS and MATLAB,respectively.The results showed that the BP neural network-based localisation model of the capacity of the transverse interference section has higher accuracy.It was of great significance to predict the actual capacity accurately to solve the traffic congestion problem and improve the road service level.

关 键 词：横向干扰影响因素 公路路段通行能力 多元线性回归 BP神经网络 

分 类 号：U491.114[交通运输工程—交通运输规划与管理]