«上一篇/Previous Article|本期目录/Table of Contents|下一篇/Next Article»

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:

34

期数:

2024年06期

页码:

59-65

栏目:

移动与物联网络

出版日期:

2024-06-10

文章信息/Info

Title:

Traffic Volume Prediction and Tidal Lane Management System Based on Digital Twin

文章编号:

1673-629X(2024)06-0059-07

作者:

马征1; 胡冰2

1. 河北工业大学 人工智能与数据科学学院,天津 300131;2. 南京邮电大学 现代邮政学院,江苏 南京 210003

Author(s):

MA Zheng1; HU Bing2

1. School of Artificial Intelligence and Data Science,Hebei University of Technology,Tianjin 300131,China;2. School of Modern Posts,Nanjing University of Posts and Telecommunications,Nanjing 210003,China

关键词:

数字孪生; 车流量预测; 潮汐车道管理; 交通管理; 智能化技术

Keywords:

digital twin; traffic volume prediction; tidal lane management; traffic management; intelligent technology

分类号:

TP399;U491

DOI:

10.20165/j.cnki.ISSN1673-629X.2024.0084

摘要:

该文介绍了基于数字孪生的潮汐车道管理系统。 该系统通过实时采集交通数据,利用数字孪生技术建立车流量预测模型和潮汐车道管理模型,通过对车流量的精准预测,实现对潮汐车道的动态管理,从而完成对交通拥堵的准确预测和有效管理。 对车流影响因素进行深入分析,建立了车流量相关影响因素模型,并通过多模型对比,选定了效果最好的极端森林模型作为预测模型,同时,引入多种评价指标对极端随机森林模型进行评价。 结果显示,无论是在哪一种评价指标上,极端随机森林模型都能对车流量预测体系实现最高精度的预测,尤其是对突发性高峰的预测。 通过数字孪生技术对潮汐车道进行方案模拟,通过对不同方案的数字化仿真,有效降低了潮汐车道方案变更和验证的成本,为城市交通管理部门的工作提供了理论支持和数据支撑。 该系统的应用可以提高交通管理的科学性和效率性,为智慧交通的发展拓宽应用渠道,为城市交通拥堵问题的解决提供有力支持。

Abstract:

We introduce a tidal lane management system based on digital twins. The system collects real-time traffic data and utilizes digital twin technology to establish models for predicting traffic volume and managing tidal lanes. By accurately predicting traffic volume,the system achieves dynamic management of tidal lanes, enabling accurate forecasting and effective management of traffic congestion. We conduct an in-depth analysis of factors influencing traffic volume,establish a model for factors affecting traffic volume,and select the extreme random forest model as the best-performing prediction model through multi-model comparisons. Additionally,we introduce various evaluation metrics to assess the extreme random forest model. The results indicate that,across all evaluation metrics,the extreme random forest model achieves the highest accuracy in predicting traffic volume,especially for sudden peak events. Through digital twin technology, we simulate scenarios for tidal lanes and effectively reduce the cost of modifying and validating tidal lane scenarios through digital simulation of different scenarios. This provides theoretical support and data backing for the work of urban traffic management departments. The application of this system can enhance the scientific and efficient management of traffic, broaden the application channels for smart transportation,and provide robust support for addressing urban traffic congestion issues.

常用功能

更新日期/Last Update: 2024-06-10