城市快速路交织区速度预测模型改进

发布时间：2018-03-02 14:14

本文选题：交通工程　切入点：交织区速度　出处：《中国公路学报》2017年01期 　论文类型：期刊论文

【摘要】：针对HCM2010交织区速度预测模型存在的3个典型问题,分别提出了交织区速度预测分段模型和改进的回归模型。选取上海市快速路4个典型交织区作为研究对象,利用遗传算法分别对HCM2010模型、分段模型和改进的回归模型参数进行了标定。为进一步测试所提出的模型,利用仿真试验的方法,对所提出的改进回归模型进行了敏感性测试。结果表明:改进的回归模型预测精度最高,交织车流速度预测平均误差为9.52%,非交织车流速度预测平均误差仅为6.64%,且改进模型不会出现交织车流速度大于非交织车流速度这种明显违背现实的预测结果;新的3个交织区的验证结果表明,改进的回归模型交织车流速度预测误差为10.79%,非交织车流速度预测误差为10.45%,所提出的模型具有良好的适用性。
[Abstract]:Aiming at the three typical problems in the velocity prediction model of HCM2010 interleaving region, the paper puts forward the interleaved zone velocity prediction segmentation model and the improved regression model, and selects four typical interleaved areas of Shanghai expressway as the research object. The parameters of HCM2010 model, piecewise model and improved regression model are calibrated by genetic algorithm. The sensitivity test of the improved regression model is carried out. The results show that the improved regression model has the highest prediction accuracy. The average error of interleaved vehicle flow velocity prediction is 9.52 and that of non-interleaved vehicle flow velocity prediction is only 6.64, and the improved model will not show that the interleaved vehicle flow velocity is larger than that of non-interleaved vehicle flow velocity, which is obviously contrary to the actual forecast results. The verification results of the three new interleaving regions show that the prediction error of interleaved vehicle flow velocity of the improved regression model is 10.79 and the prediction error of non-interleaved vehicle flow velocity is 10.45. The proposed model has good applicability.
【作者单位】：同济大学道路与交通工程教育部重点实验室;
【基金】：“十二五”国家科技支撑计划项目(2014BAG01B01)
【分类号】：U491

【相似文献】