交通事件下路段通行能力多层模糊估计方法

Multilayer Fuzzy Estimation Method for Road Traffic Capacity Affected by Traffic Incidents

交通事件常导致道路网络区域性拥堵。研究交通事件下路段通行能力估计方法,有助于精准施策,缓解交通拥堵,提高道路服务水平。针对交通事件下路段通行能力影响因素复杂、样本数据量少等问题,本文提出基于多层模糊的通行能力估计方法。首先,宏观分析路段通行能力影响因素间关联性,基于模糊逻辑,构建通行能力估计模糊子系统,并将各类交通事件影响量化为多项特征值。其次,研究以交通事件影响量化特征值为输入,构建路段通行能力多层模糊估计方法,输出路段通行能力折减率。最后,通过实际案例验证模型的有效性,对比分析交通事件下路段通行能力实测值、经典通行能力计算方法与本文方法。结果表明:本文方法适用于多种交通事件场景,通行能力估计总平均误差为5.43%,相较HCM折减车道法估计精度显著提高。在交通事件动态检测数据支撑下,研究成果可在线估计交通事件影响下的路段通行能力,可为交通管理部门实现非常态下的智能管理与控制提供支持。

Traffic incidents often lead to regional congestion of the road network.Researching methods to estimate the capacity of road sections during traffic incidents is crucial for implementing accurate policies,mitigating traffic congestion,and improving the level of service.This paper proposes a multi-layered fuzzy-based method for estimating capacity.The method addresses the complex factors that affect capacity during traffic incidents,as well as the shortage of sample data.Firstly,the interrelationships between factors affecting capacity are analyzed macroscopically.Based on these relationships,a fuzzy logic system is constructed to estimate capacity.The system quantifies various traffic incident impacts and converts them into multiple feature values.Secondly,a multi-layered fuzzy estimation method is developed for capacity using traffic incident impact quantification feature values as input and outputting a reduced section capacity rate.Finally,the effectiveness of the proposed model is verified through practical cases,which compared the measured capacity during traffic incidents,classical capacity calculation methods,and the proposed method.The comparison results show that this method is applicable in various traffic incident scenarios,with a total average estimation error of 5.43%,significantly improving precision compared to the traditional HCM lane reduction method.Supported by dynamic detection data during traffic incidents,this research can provide support for intelligent management and control during non-normal conditions for traffic management departments.