论文部分内容阅读
以高速公路基本路段上车道变换过程中后侵入时间(PET)的界定、测量和分析为基础,通过极值统计方法建立了车道变换PET的超阈值模型,进而结合模型参数估计构建了百万次车辆变道事故重现水平指标,并将该指标与交通事故数据次数的关系进行了分析。结果表明:百万次变道事故重现水平与交通事故次数之间的皮尔逊相关性系数高达0.77,决定性系数也达到了0.60,二者都明显优于PET=3.0 s所界定的车道变换冲突次数与交通事故次数之间的相关性。这说明通过对PET信息的进一步挖掘和利用,本文所得到的完全不依赖于历史事故数据的安全分析指标具有和交通事故指标近似等效的评价效果,也从侧面进一步证明了交通冲突技术的有效性。
Based on the definition, measurement and analysis of post-intrusion time (PET) in the lane change process of expressway section, an over-threshold model of lane-changing PET was established by extreme value statistics. Then, with the model parameter estimation, Vehicle lane change accidents to reproduce the level of indicators, and the index and the number of traffic accidents data were analyzed. The results show that Pearson’s correlation coefficient between the recurrence level of a million lane-changing accidents and the number of traffic accidents is as high as 0.77 and the coefficient of decisiveness is 0.60, both of which are obviously better than the lane-changing conflicts defined by PET = 3.0 s The correlation between the frequency and the number of traffic accidents. This shows that through the further mining and utilization of PET information, the safety analysis indicators completely independent of the historical accident data obtained in this paper have the same equivalent evaluation results as the traffic accident indicators, and further prove the effectiveness of the traffic conflict technology from the side Sex.