高速公路追尾事故严重程度影响因素分析及预防对策研究

发布时间：2018-01-02 01:35

本文关键词：高速公路追尾事故严重程度影响因素分析及预防对策研究　出处：《吉林大学》2016年硕士论文　论文类型：学位论文

【摘要】：随着经济的快速发展,我国高速公路通车里程增长迅速。据统计,截止2015年底,我国的高速公路通车里程已突破12万公里,跃居世界第一。高速公路具有行车速度高、通行能力大、运输成本低等优点,对于沿线经济的发展具有重要的推动作用。然而,由于其全封闭、立体交叉、对向车流分离等特殊的交通环境,高速公路发生追尾碰撞事故的风险远高于其他道路。且由于车速高,车流量大,高速公路追尾事故一旦发生,其后果往往十分严重。因此,对高速公路追尾事故严重程度影响因素进行分析,并提出相应的预防对策,具有十分重要的现实意义。在此背景下,本论文通过对我国某地区407起高速公路追尾事故进行分析,分别构建了基于累积Logistic的高速公路追尾事故严重程度概率预测模型和基于SVM的高速公路追尾事故严重程度预测模型。在累积Logistic模型中,利用混合逐步选择法,筛选出了驾驶员驾龄、车辆安全状况、地形、路表状况、天气、能见度等因素与追尾事故严重程度显著相关。在SVM模型中,引入灵敏度分析方法,发现驾驶员驾龄、车辆安全状况、是否超载、路表状况、道路线形、天气、能见度等因素对高速公路追尾事故严重程度具有显著影响。最后,本论文根据累积Logistic和SVM模型分析结果,分别从驾驶员、车辆、道路、环境四个方面对高速公路追尾事故严重程度影响因素进行了分析,并提出了相应的对策来预防高速公路追尾事故的发生,降低追尾事故的严重程度。
[Abstract]:With the rapid development of economy, the mileage of expressway traffic in China is increasing rapidly. According to statistics, as of end of 2015, the mileage of expressway in China has exceeded 120,000 km. Freeway has the advantages of high speed, large capacity and low transportation cost, which plays an important role in promoting the development of economy along the route. However, because of its completely closed, three-dimensional intersection. For the special traffic environment, such as separation of traffic flow, the risk of rear-end collision on expressway is much higher than that on other roads, and because of the high speed and large traffic flow, the rear-end accident of expressway occurs once. The consequences are often very serious. Therefore, it is of great practical significance to analyze the factors affecting the severity of rear-end accidents and to put forward the corresponding preventive measures. This paper analyzes 407 accidents of freeway rear-end in a certain area of China. The probability prediction model of expressway rear-end accident severity based on cumulative Logistic and the prediction model of expressway rear-end accident severity based on SVM were constructed respectively. The cumulative Logisti model was used to predict the severity of rear-end accident. C in the model. The factors of driver's driving age, vehicle safety condition, terrain, road surface condition, weather, visibility and so on are significantly related to the severity of rear-end accident using mixed stepwise selection method. In SVM model. The sensitivity analysis method is introduced to find out the driver's driving age, vehicle safety condition, whether overload, road surface condition, road alignment, weather. Visibility and other factors have a significant impact on the severity of rear-end accidents. Finally, according to the cumulative Logistic and SVM model analysis results, from drivers, vehicles, roads, respectively. The factors influencing the severity of the rear-end accident of expressway are analyzed in four aspects of environment, and the corresponding countermeasures are put forward to prevent the occurrence of the rear-end accident and to reduce the serious degree of the rear-end accident.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：U491.31

【参考文献】