多场景智慧高速公路的交通气象观测站点布设

Locating road weather information system stations on multi-scenario smart expressway

在交通强国战略和公路行业高质量发展背景下,我国开展了新一代智慧高速公路试点工程的建设。作为智能交通环境感知设施的一部分,交通气象观测站的科学选址对于提高智慧高速公路的通行能力和服务水平有着不可忽视的作用。针对车路协同和自动驾驶、全天候通行和主动交通管控以及伴随式信息服务3种典型的智慧高速公路应用场景,以及可维护性和交通需求等影响因素,提出由建设运营总成本最小、所选站点组成系统的可靠性最大和所覆盖路段的总车辆里程数最大组成的3个目标。以规划期内允许建设的最大站点数、两站点间的建议最短距离作为关键约束条件,构建多目标优化模型,并采用带精英策略的非支配排序遗传算法进行求解。将正在建设的潍青高速主体部分交通气象观测站的布设作为算例,在PyCharm中进行编程计算,得到了一个非劣解集。以其中一个非劣解作为布设方案,在12个候选点中选出7个布设位置并为每个站点配置一种类型的传感器,对应的总成本为109.175万元,可靠性为0.993 79,总车辆里程数为31 809.75百万km。最后讨论最大布设个数对目标函数值的影响,结果表明,3个目标值随着最大布设个数的增加会有不同的增加趋势。当最大布设个数由9增加到10和11时,对可靠性的增加作用较小。

In the context of the strategy of a powerful country in transportation and the high-quality development of the highway industry, our country has launched the construction of a new generation of smart expressway pilot projects. As a part of intelligent transportation environment perception facility, scientific location of road weather information station plays an important role in improving smart expressway capacity and service level. Smart expressway has three typical application scenarios like vehicle-infrastructure cooperation and automatic driving,all-weather running and active traffic control, and location based service. In regards of those scenarios and influencing factors such as maintainability and traffic demand, this study proposed three objectives composed of the minimum total construction and operation cost, the maximum reliability of the system composed of selected stations, and the maximum total vehicle mileage of covered sections. Based on those objectives as well as key constraints like the maximum number of stations allowed to be constructed during the planning period and the recommended shortest distance between two stations, a multi-objective optimization model was constructed. A non-dominated sorting genetic algorithm with elite strategy was used to solve that model. Furthermore, this investigation took the layout of road weather information system stations on the main part of Weiqing Expressway under construction as an example. A Pareto solution set was obtained by programming in PyCharm.For example, in one Pareto solution 7 deployment locations are selected from 12 candidate points and one type of sensor was configured for each station. Its total cost is 1.091 75 million yuan, reliability 0.993 79, and total vehicle mileage traveled 31 809.75 million kilometers. Finally, the influence of the maximum number of deployments on the value of the objective function was discussed. The results show that three objective values have different increasing trends as the maximum number of deployments increases. When the maximum number of deployments increases from 9 to 10 and 11, the increase in reliability is relatively slight.