考虑高速公路运行风险的雨天可变限速方法

为保证雨天环境下高速公路行驶安全，降低道路整体运行风险，结合雨天风险特征，开展考虑运行风险的雨天可变限速研究。首先应用随机森林模型，标定雨天环境下高速公路动静态风险因素的特征重要度，并结合熵值理论，建立高速公路风险模型、计算风险系数、划分风险等级;之后，以空域自适应算法中可变限速推演变化规律为基础，考虑大、小型车的行驶特征，结合预期风险、雨天停车视距、水膜厚度等因素，优化可变限速模型，细化大、小型车辆的初始控制值，进而提出不同降雨强度、不同能见度下的可变限速推荐值；在此基础上，利用驾驶模拟实景仿真系统、心理生理检测设备、微观交通流仿真软件，开展可变限速系统控制值合理性及驾驶人行车适应性与交通流运行状态的实证分析。研究结果表明：随着能见度降低和降雨量增大，在可变限速控制下，驾驶人呈现出交感神经兴奋性减弱、副交感神经兴奋性增强、心理紧张度降低的状态，其平均心率、心率变异性高频值、心率变异性低频值、心率变异性差异值分别由74.13、0.121、0.643、2.37变为78.23、0.192、0.567、2.01，驾驶人对限速方案的适应性良好；同时，可变限速可保证道路整体通行效率，不会造成交通流风险震动，在小型车、大型车限速分别为80、60 km·h^-1和小型车限速60 km·h^-1、大型车禁止驶入场景下，碰撞时间均值、中值大于未限速场景，各车道的行车安全性均能得到保障;提出的雨天可变限速控制方法合理，且具有一定工程适应性，能为异常天气高速公路宏观车流主动防控提供理论支撑。

Variable Speed Limit Method for Rainy Weather Considering Freeway Operation Risk

To ensure the safe driving of freeway traffic in rainy environments and reduce operation risk, a variable speed limit (VSL) management mode can be adopted to control the operation risks of freeways in rainy weather. First, the random forest analysis model was used to calibrate the feature importance of dynamic and static risk factors, and the entropy value theory was used to establish a risk analysis model of a freeway. Combining the model with the risk characteristics of dynamic traffic operations in rainy weather, the dynamic operation risk coefficient of expressways was calculated, and the risk level was determined. Subsequently, based on the variation law of VSL in Mainstream Traffic Flow Control and the characteristics of large and small vehicles, stopping sight-distance, water film thickness, expected risk, and other factors, the VSL model of freeways in rainy weather was optimized. The initial VSL control values for the large and small vehicles were refined. Then, a VSL for rainy days was proposed. Finally, a driving simulation system, a psychophysics-physiological testing equipment, and traffic flow simulation software VISSIM were used to verify the rationality of the control value based on the driver's driving adaptability and traffic flow. The results show that as visibility decreases and rainfall increases, under the VSL management of this study, the drivers' sympathetic nerve excitability weakened, whereas their parasympathetic nerve excitability enhanced, and their psychological tension reduced. The mean heart rate, high-frequency value, low-frequency value, and difference in heart rate variability of the drivers change from 74.13, 0.121, 0.643, and 2.37 to 78.23, 0.192, 0.567, and 2.01, respectively. These results indicate that drivers demonstrate good adaptability to the proposed speed. In addition, the verification results of the traffic flow operation state show that the VSL can ensure the overall traffic efficiency of the road and does not cause traffic flow risk vibrations. When the speed limit of small and large vehicles is 80 and 60 km·h^-1, or the speed limit of the small is 60 km·h^-1 and the driving of the large is forbidden, the mean and median time-to-collision values exceed those in a scenario without a speed limit, which shows that the safety performance of each lane can be guaranteed. In conclusion, the VSL control method proposed herein exhibits certain rationality and engineering adaptability, and can provide theoretical support for the active prevention and control of macro traffic flow on freeways in severe weather.