技术领域technical field
本发明涉及高速公路拥挤预测的技术领域，尤其是指一种融合历史数据和预约数据的高速公路拥挤预测方法。The invention relates to the technical field of highway congestion prediction, in particular to a highway congestion prediction method that combines historical data and reservation data.
背景技术Background technique
随着交通运输量的不断增加，我国高速公路的拥堵问题(尤其是在节假日高峰期)愈发严重。在2012年国庆首日，全国有16个省份的26条高速公路出现拥堵；而在2015年国庆首日，仅广东省就有31条高速公路出现拥堵。为了解决拥堵问题，高速公路管理部门会在拥堵发生之后采取一定的疏导措施。虽然这些措施能达到一定效果，但已发生的拥堵还是给出行者和管理者都造成了损失。因此，一些学者提出了在拥堵发生前进行拥堵疏导的设想。近年来逐渐成熟的移动互联网技术使得这一设想成为可能：将移动互联网和高速公路运营管理结合起来，通过预约通行和出行诱导实现拥堵的主动疏导。With the continuous increase of traffic volume, the congestion problem of expressways in my country (especially during the holiday peak period) is becoming more and more serious. On the first day of National Day in 2012, 26 expressways in 16 provinces across the country were congested; on the first day of National Day in 2015, 31 expressways in Guangdong Province alone were congested. In order to solve the congestion problem, the expressway management department will take certain dredging measures after the congestion occurs. Although these measures can achieve certain effects, the congestion that has occurred has caused losses to travelers and managers. Therefore, some scholars put forward the idea of congestion relief before the occurrence of congestion. The mobile Internet technology that has gradually matured in recent years has made this idea possible: combine the mobile Internet with expressway operation and management, and realize active congestion relief through reservation and travel guidance.
要实现这样的目标，对高速公路进行准确的拥挤预测是前提。国内外学者已经对高速公路拥挤预测进行了大量的研究。发达国家对拥挤预测模型的研究始于上世纪中叶。我国在道路交通拥挤预测方面的研究起步较晚且大多数研究都是通过短期交通量或其他指标来评价道路交通拥挤状态。目前应用在拥挤预测方面的主要方法有时间序列模型、卡尔曼滤波模型、神经网络模型、支持向量机模型等。这些方法各有所长，但是都有一个共同点——在历史数据的基础上进行预测。这使得现有方法难以应对高速公路的预约通行为带来的挑战：①预约数据将为拥挤预测工作提供新的数据源，在预测的过程中要考虑预约数据和历史数据的融合。②预约行为将改变交通量在高速公路路网的分布。如果拥挤预测过程中不对这些变化进行考虑而简单地套用现有方法，将会导致拥挤预测精度的下降。To achieve such a goal, accurate congestion prediction on expressway is the premise. Scholars at home and abroad have done a lot of research on expressway congestion prediction. Research on congestion prediction models in developed countries began in the middle of the last century. The research on road traffic congestion prediction in our country starts late and most of the researches evaluate the state of road traffic congestion through short-term traffic volume or other indicators. At present, the main methods used in congestion prediction include time series model, Kalman filter model, neural network model, support vector machine model and so on. These methods have their own strengths, but they all have one thing in common - making predictions based on historical data. This makes it difficult for existing methods to deal with the challenges brought about by reservation traffic on expressways: ① reservation data will provide a new data source for congestion prediction, and the fusion of reservation data and historical data should be considered in the prediction process. ②Reservation behavior will change the distribution of traffic volume in the expressway network. If these changes are not considered in the process of congestion prediction and the existing methods are simply applied, the accuracy of congestion prediction will decrease.
因此，本发明提出了一种融合预约数据和历史数据进行高速公路拥挤预测方法，为出行诱导和高速公路管理工作提供更可靠的依据。Therefore, the present invention proposes a method for predicting expressway congestion by fusing reservation data and historical data, so as to provide more reliable basis for travel guidance and expressway management.
发明内容Contents of the invention
本发明的目的在于克服现有技术的缺点和不足，提供一种融合历史数据和预约数据的高速公路拥挤预测方法，突破常规高速公路拥挤预测方法未考虑用户预约数据、无法反映实行预约通行后交通量在路网内重新分布等缺陷，实现了在高速公路预约通行条件下的拥挤预测，能够有效提高预测精度，为高速公路预约通行的实现奠定基础。The purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art, and provide a highway congestion prediction method that integrates historical data and reservation data, which breaks through the conventional highway congestion prediction method that does not consider user reservation data and cannot reflect traffic after reservation traffic is implemented. By eliminating defects such as redistribution of traffic in the road network, it realizes the congestion prediction under the condition of expressway reserved traffic, which can effectively improve the prediction accuracy and lay the foundation for the realization of expressway reserved traffic.
为实现上述目的，本发明所提供的技术方案为：一种融合历史数据和预约数据的高速公路拥挤预测方法，包括以下步骤：In order to achieve the above object, the technical solution provided by the present invention is: a method for predicting expressway congestion by fusing historical data and reservation data, comprising the following steps:
1)选定研究的高速公路，获取基础资料，总共包括四方面的资料：高速公路用户预约数据、高速公路路网基础数据、高速公路交通流历史数据、高速公路收费政策数据。1) Select the highway for research and obtain basic data, including four aspects of data in total: highway user reservation data, highway network basic data, highway traffic flow history data, highway toll policy data.
2)根据高速公路用户预约数据和高速公路路网基础数据，对高速公路路网进行时空划分处理并确定各路段的预测时段。2) According to the reservation data of expressway users and the basic data of expressway road network, the expressway road network is divided into time and space and the prediction period of each road section is determined.
3)对预测时段内高速公路路网各路段的交通量进行预测：首先，利用历史数据对交通总量进行预测；然后，将总交通量按照预约量和非预约量进行分配并用预约数据对非预约量进行修正；最后将分配后预约量和非预约量相加得到各路段的交通量。3) Predict the traffic volume of each section of the expressway network within the forecast period: first, use historical data to predict the total traffic volume; then, distribute the total traffic volume according to reserved volume and non-reserved volume The reservation volume is corrected; finally, the traffic volume of each road section is obtained by adding the reserved volume and non-reserved volume after allocation.
4)预测各路段的拥挤度：采用饱和度作为拥挤的评价指标，通过阈值判断各路段的拥挤状态。4) Predict the congestion degree of each road section: the saturation is used as the evaluation index of congestion, and the congestion state of each road section is judged by the threshold.
5)利用程序设计语言实现拥挤预测结果的可视化输出。5) Use the programming language to realize the visual output of the congestion prediction results.
在步骤1)中，所述高速公路用户预约数据包括用户输入的出发地、目的地、进站时段、车型、用户app或微信公众号的市场占有率，通过用户app后台或者微信公众号后台获取。所述高速公路的路网基础数据包括各道路的长度、车道数、车道宽度、设计通行能力、自由流条件下的车辆行驶时间，通过高速公路运营管理部门获取。所述高速公路交通流历史数据包括流量和车速数据，通过高速公路运营管理部门获取。所述高速公路收费政策数据包括基础费率、优惠幅度和旅行时间价值，通过高速公路运营管理部门获取。In step 1), the expressway user reservation data includes the departure place, destination, entry period, vehicle type, market share of the user app or WeChat official account input by the user, and is obtained through the background of the user app or the official account of WeChat . The road network basic data of the expressway includes the length of each road, the number of lanes, the width of the lane, the design capacity, and the vehicle travel time under free flow conditions, and is obtained through the expressway operation and management department. The historical traffic flow data of the expressway includes flow rate and vehicle speed data, and is obtained through the expressway operation and management department. The expressway toll policy data includes basic rate, preferential range and travel time value, and is obtained through the expressway operation and management department.
在步骤2)中，根据高速公路路网基础资料和用户预约数据，对高速公路路网进行时空划分处理并确定各路段的预测时段，具体包括以下步骤：In step 2), according to the basic information of the expressway road network and the user reservation data, the expressway road network is divided into time and space and the prediction period of each road section is determined, which specifically includes the following steps:
2.1)时间的划分：采用与用户app或微信公众号上进站时段划分相一致的单位时间间隔Δt，可根据高速公路运营管理部门的要求取10min，20min，30min，60min，把一天24小时分成m个时间片。2.1) Time division: use the unit time interval Δt that is consistent with the time division of the user's app or WeChat official account, and take 10min, 20min, 30min, 60min according to the requirements of the expressway operation and management department, and divide 24 hours a day into m time slices.
式中：T——时间片集合；In the formula: T——time slice set;
ti——第ti个时间片；ti ——the ti- th time slice;
m——时间片个数，个；m——the number of time slices;
Δt——单位时间间隔，min。Δt—unit time interval, min.
2.2)空间的划分：采用与高速公路运营管理部门采集交通流数据的路段划分相一致的单位空间间隔Δs，可根据高速公路运营管理部门的要求取1km，2km，5km，10km，将路网内的所有高速公路划分成个路段。2.2) Space division: use the unit space interval Δs that is consistent with the division of road sections collected by the expressway operation management department to collect traffic flow data. All highways in the segments.
式中：S——路段集合；In the formula: S——section set;
——第j条高速公路的第kj个路段； - the kjth section of the jth expressway;
kj——第j条高速公路的路段个数，个；kj ——the number of road sections of the jth expressway, number;
Lj——第j条高速公路的长度，km；Lj ——the length of the jth expressway, km;
Δs——单位空间间隔，km；Δs—unit space interval, km;
n——高速公路条数，条。n——the number of highways.
2.3)确定各路段的预测时段：用户到达每一路段终点的时间通过如下公式计算：2.3) Determine the forecast period of each road section: the time for the user to reach the end of each road section is calculated by the following formula:
式中：——用户达到第j条高速公路的第kj个路段终点的时间；In the formula: ——the time when the user reaches the end point of the kjth section of thejth expressway;
——用户达到第j条高速公路的入口的时间； ——the time when the user reaches the entrance of the jth expressway;
vj——第j条高速公路的平均行驶车速；vj ——the average driving speed of the jth expressway;
ta——用户预约的进站时段的起点；ta ——the starting point of the inbound time slot reserved by the user;
tb——用户预约的进站时段的终点。tb ——The end point of the inbound slot reserved by the user.
这样，就可以判断路段的预测时段：若则路段的预测时段为ti。In this way, it is possible to determine the road segment Forecast period for : if then section The prediction period of t i is ti .
在步骤3)中，对预测时段高速公路路网各路段的交通量进行预测，具体包括以下步骤：In step 3), the traffic volume of each section of the expressway road network during the forecast period is predicted, which specifically includes the following steps:
3.1)根据历史数据预测各路段的总交通量，采用BP神经网络进行预测，可直接在matlab中调用自带的函数实现。3.1) Predict the total traffic volume of each road section according to historical data, and use BP neural network to predict, which can be realized by directly calling the built-in function in matlab.
3.2)计算预约交通总量和非预约交通总量其计算公式为：3.2) Calculating the total amount of reserved traffic and total walk-in traffic Its calculation formula is:
式中：——预测时段ti时路网的总交通量；In the formula: ——the total traffic volume of the road network at the forecast period ti ;
——预测时段ti时路网的预约交通总量； ——the total amount of scheduled traffic on the road network at the forecast time period ti ;
——预测时段ti时路网的非预约交通总量； ——the total amount of non-reserved traffic on the road network at the time of prediction period ti ;
——预测时段ti时第j条高速公路的第kj个路段的交通量； ——the traffic volume of the k j-th section of the j-th expressway at the forecast time period ti ;
α——用户app或微信公众号的市场占有率。α——The market share of the user's app or WeChat official account.
其余各参数意义如前所述。The meanings of other parameters are as mentioned above.
3.3)按照车流量的历史分布规律，对非预约交通总量在路网范围内进行分配，得到各路段的非预约量，其计算公式为：3.3) According to the historical distribution law of traffic volume, the total amount of non-reserved traffic is allocated within the scope of the road network, and the non-reserved volume of each road section is obtained. The calculation formula is:
式中：——预测时段ti时第j条高速公路的第kj个路段的非预约交通量。In the formula: ——The non-reserved traffic volume of the kjth section of the jth expressway at the forecast time period ti .
3.4)采用用户均衡模型对预约交通总量在路网范围内进行分配，得到各路段的预约量。3.4) Use the user balance model to distribute the total amount of reserved traffic within the road network, and obtain the reserved amount of each road section.
式中：——预测时段ti时第j条高速公路的流量。In the formula: ——The flow rate of thejth expressway at the forecast time period ti .
——预约用户在路网中的路阻函数，其具体形式为： ——The road resistance function of the reservation user in the road network, its specific form is:
式中：to——自由流条件下的行驶时间，min；In the formula: to ——travel time under free flow condition, min;
——第j条高速公路的第kj个路段的道路设计通行能力，pcu/h； ——The road design capacity of the kjth road section of thejth expressway, pcu/h;
co——基础通行费用，元；co ——basic toll fee, yuan;
γ——优惠幅度，％；γ——preferential margin, %;
α,β——待标定的参数；α, β——parameters to be calibrated;
τ——旅行时间价值，元/h。τ——travel time value, yuan/h.
3.5)采用实际预约量对各路段的预约量进行修正，修正规则如下：3.5) Use the actual reservation quantity to correct the reservation quantity of each road section, and the correction rules are as follows:
①若则不变。① if but constant.
②若则② if but
式中：——预测时段ti时第j条高速公路的第kj个路段的实际预约交通总量；In the formula: ——The actual scheduled traffic volume of the kjth road section of thejth expressway at the time of prediction period ti ;
3.6)将各路段预约量和非预约量进行叠加，得到各路段的交通量。3.6) Superimpose the reserved volume and non-reserved volume of each road section to obtain the traffic volume of each road section.
式中：——预测时段ti时第j条高速公路的第kj个路段的预约交通量。In the formula: ——Reserved traffic volume of the kj -th section of the j- th expressway at the forecast time period ti .
在步骤4)中，预测各路段的拥挤度，具体包括以下步骤：In step 4), the degree of congestion of each road section is predicted, which specifically includes the following steps:
4.1)计算各路段的饱和度，计算公式为：4.1) Calculate the saturation of each road section, the calculation formula is:
式中：——预测时段ti时第j条高速公路的第kj个路段的饱和度。In the formula: ——The saturation of the kjth section of thejth expressway at the time period ti is predicted.
4.2)拥挤度等级的划分，通过饱和度阈值将拥挤度划分成不同的等级。可以根据各城市的交通调查数据来确定阈值。在无调查资料的情况下，建议使用下列表格进行分级。4.2) Classification of congestion degree, the degree of congestion is divided into different grades by saturation threshold. The threshold can be determined based on the traffic survey data of each city. In the absence of survey data, the following table is recommended for grading.
表1Table 1
在步骤5)中，利用程序设计语言实现拥挤预测结果的可视化输出。将不同的拥挤度等级与不同的颜色对应，在编写代码时将相应的颜色代码赋予各路段。运行代码，即可输出拥挤预测结果。In step 5), the visual output of the congestion prediction result is realized by using a programming language. Correspond to different congestion levels with different colors, and assign corresponding color codes to each road segment when writing codes. Run the code to output the congestion prediction result.
本发明与现有技术相比，具有如下优点与有益效果：Compared with the prior art, the present invention has the following advantages and beneficial effects:
1、将移动互联网技术运用到高速公路管理中，借助手机app或者微信公众号获取预约数据，为实现高速公路的预约通行提供基础。1. Apply mobile Internet technology to expressway management, obtain reservation data with the help of mobile app or WeChat official account, and provide a basis for realizing reserved passage on expressways.
2、将用户预约数据融合到预测算法中，对利用历史数据进行拥挤预测的结果进行修正，能够提高预测工作的精度。2. Integrating user reservation data into the forecasting algorithm, correcting the results of congestion forecasting using historical data, can improve the accuracy of forecasting work.
3、算法较简便，建模的过程简单，在训练的过程中不需要大量的学习样本，预测精度较高。3. The algorithm is relatively simple, the modeling process is simple, a large number of learning samples are not required in the training process, and the prediction accuracy is high.
4、便于高速公路运营管理部门通过价格杠杆，利用预约出行方式进行交通疏导，促进高速公路网络的用户均衡，尽量避免形成拥堵，也便于用户避开拥挤路段和拥堵时段错峰出行。4. It is convenient for the expressway operation and management department to use the price lever to use the reserved travel method to guide traffic, promote the user balance of the expressway network, avoid congestion as much as possible, and also facilitate users to avoid congested roads and travel at peak times during congested periods.
5、有利于高速公路管理部门及时掌握路网交通流信息和拥堵情况，对未来可能形成的拥堵进行预警，及时采取必要的疏解措施。5. It is beneficial for the expressway management department to grasp the road network traffic flow information and congestion situation in a timely manner, to give early warning of possible future congestion, and to take necessary relief measures in a timely manner.
附图说明Description of drawings
图1为本发明的逻辑流程示意图。Fig. 1 is a schematic diagram of the logic flow of the present invention.
图2为本发明实施例的路网示意图。Fig. 2 is a schematic diagram of a road network according to an embodiment of the present invention.
图3为本发明实施例的拥挤预测可视化输出结果。FIG. 3 is a visualization output result of congestion prediction according to an embodiment of the present invention.
具体实施方式Detailed ways
下面结合具体实施例对本发明做进一步的说明。The present invention will be further described below in conjunction with specific embodiments.
如图1所示，本实施例所述的融合历史数据和预约数据的高速公路拥挤预测方法，包括以下步骤：As shown in Figure 1, the expressway congestion prediction method of fusion historical data and reservation data described in the present embodiment comprises the following steps:
1)选定研究的高速公路，获取基础资料，总共包括四方面的资料：高速公路用户预约数据、高速公路路网基础数据、高速公路交通流历史数据、高速公路预约费用数据。其中，所述高速公路用户预约数据包括用户输入的出发地、目的地、进站时段、车型、用户app或微信公众号的市场占有率，通过用户app后台或者微信公众号后台获取。所述高速公路的路网基础数据包括各道路的长度、车道数、车道宽度、设计通行能力、自由流条件下的车辆行驶时间，通过高速公路运营管理部门获取。所述高速公路交通流历史数据包括流量和车速数据，通过高速公路运营管理部门获取。所述高速公路收费政策数据包括基础费率、优惠幅度和旅行时间价值，通过高速公路运营管理部门获取。1) Select the highway for research and obtain basic data, including four aspects of data in total: highway user reservation data, highway network basic data, highway traffic flow historical data, and highway reservation fee data. Wherein, the expressway user reservation data includes the departure place, destination, entry time period, vehicle type, market share of the user app or WeChat official account input by the user, and is obtained through the background of the user app or the WeChat official account. The road network basic data of the expressway includes the length of each road, the number of lanes, the width of the lane, the design capacity, and the vehicle travel time under free flow conditions, and is obtained through the expressway operation and management department. The historical data of expressway traffic flow includes flow rate and vehicle speed data, and is obtained through the expressway operation and management department. The expressway toll policy data includes basic rate, preferential margin and travel time value, and is obtained through the expressway operation and management department.
本实施例所采用的高速公路路网如图2所示。假定从用户app后台获得某用户输入的出发地为A、目的地为B、进站时段为8：00–9:00,车型为一类车；用户app的市场占有率是12％。The highway network used in this embodiment is shown in FIG. 2 . Assume that the starting point is A, the destination is B, the arrival time is 8:00–9:00, and the vehicle type is Class I. The market share of the user app is 12%.
根据上述某用户预约的出发地、目的地和进站时段，收集到其余的资料如下：高速公路路网基础数据如表1所示、高速公路1、2的流量历史数据和当前预约用户总量分别如表2、表3所示(由于篇幅所限，只列出了预测时段的历史数据)、高速公路收费政策数据如表4所示，两条高速公路的平均行驶车速为80km/h，自由流条件下的车辆行驶时间分别为1.1h和1.5h。According to the place of departure, destination and inbound time reserved by a certain user above, the rest of the data collected are as follows: the basic data of the expressway road network is shown in Table 1, the historical traffic data of expressways 1 and 2 and the total number of current reservation users As shown in Table 2 and Table 3 respectively (due to space limitations, only the historical data of the forecast period are listed), and the expressway toll policy data are shown in Table 4. The average driving speed of the two expressways is 80km/h, Vehicle travel times under free-flow conditions were 1.1 h and 1.5 h, respectively.
表1高速公路路网基础数据Table 1 Basic data of highway network
表2高速公路1的流量历史数据和当前预约用户总量Table 2 Historical flow data of expressway 1 and the total number of current reservation users
表3高速公路2的流量历史数据和当前预约用户总量Table 3 Historical traffic data and total number of current reservation users of Expressway 2
表4高速公路收费政策数据Table 4 Expressway Toll Policy Data
2)根据高速公路路网基础资料和用户预约数据，对高速公路路网进行时空划分处理并确定各路段的预测时段，具体包括以下步骤：2) According to the basic information of the expressway network and user reservation data, the time-space division of the expressway network is performed and the prediction period of each road section is determined, which specifically includes the following steps:
2.1)时间的划分：采用与用户app或微信公众号上进站时段划分相一致的单位时间间隔Δt，可根据高速公路运营管理部门的要求取10min，20min，30min，60min，把一天24小时分成m个时间片。2.1) Time division: use the unit time interval Δt that is consistent with the time division of the user's app or WeChat official account, and take 10min, 20min, 30min, 60min according to the requirements of the expressway operation and management department, and divide 24 hours a day into m time slices.
式中：T——时间片集合；In the formula: T——time slice set;
ti——第ti个时间片；ti ——the ti- th time slice;
m——时间片个数，个；m——the number of time slices;
Δt——单位时间间隔，min。Δt—unit time interval, min.
在本实施例中，单位时间间隔Δt＝60min，一天24小时分成m＝24个时间片，则T＝{t1,…,t12,…,t24}。In this embodiment, the unit time interval Δt=60 min, and 24 hours a day is divided into m=24 time slices, then T={t1 , . . . , t12 , . . . , t24 }.
2.2)空间的划分：采用与高速公路运营管理部门采集交通流数据的路段划分相一致的单位空间间隔Δs，可根据高速公路运营管理部门的要求取1km，2km，5km，10km，将路网内的所有高速公路划分成个路段。2.2) Space division: use the unit space interval Δs that is consistent with the division of road sections collected by the expressway operation management department to collect traffic flow data. All highways in the segments.
式中：S——路段集合；In the formula: S——section set;
——第j条高速公路的第kj个路段； - the kjth section of the jth expressway;
kj——第j条高速公路的路段个数，个；kj ——the number of road sections of the jth expressway, number;
Lj——第j条高速公路的长度，km；Lj ——the length of the jth expressway, km;
Δs——单位空间间隔，km；Δs—unit space interval, km;
n——高速公路条数，条。n——the number of highways.
在本实施例中，每条高速公路只进行一次流量采集工作，故每条高速公路即为1个路段。各条高速公路的长度即为单位空间间隔，则S＝{s11，s21}。In this embodiment, each expressway only performs traffic collection once, so each expressway is 1 section. The length of each expressway is the unit space interval, then S={s11 , s21 }.
2.3)确定各路段的预测时段：用户到达每一路段终点的时间通过如下公式计算：2.3) Determine the forecast period of each road section: the time for the user to reach the end of each road section is calculated by the following formula:
式中：——用户达到第j条高速公路的第kj个路段终点的时间；In the formula: ——the time when the user reaches the end point of the kjth section of thejth expressway;
——用户达到第j条高速公路的入口的时间； ——the time when the user reaches the entrance of the jth expressway;
vj——第j条高速公路的平均行驶车速；vj ——the average driving speed of the jth expressway;
ta——用户预约的进站时段的起点；ta ——the starting point of the inbound time slot reserved by the user;
tb——用户预约的进站时段的终点。tb ——The end point of the inbound slot reserved by the user.
这样，就可以判断路段的预测时段：若则路段的预测时段为ti。In this way, it is possible to determine the road segment Forecast period for : if then section The prediction period of t i is ti .
在本实施例中，进站时间为(表示8点30分，下同)。用户达到高速公路1各路段的终点时间为9.5，故高速公路1的预测时段为t9(8:00-9:00)和t10(9:00-10:00)。用户达到高速公路2各路段的终点时间为10.00。故高速公路2的预测时段为t9(8:00-9:00)和t10(9:00-10:00)。In this example, the entry time is (It means 8:30, the same below). The time when the user arrives at the end of each section of expressway 1 is 9.5, so the predicted time periods of expressway 1 are t9 (8:00-9:00) and t10 (9:00-10:00). The end time for the user to arrive at each section of Expressway 2 is 10.00. Therefore, the forecast period of expressway 2 is t9 (8:00-9:00) and t10 (9:00-10:00).
3)对预测时段高速公路路网各路段的交通量进行预测，具体包括以下步骤：3) Predict the traffic volume of each section of the expressway network during the forecast period, specifically including the following steps:
3.1)根据历史数据预测各路段的总交通量，采用BP神经网络进行预测，可直接在matlab中调用自带的函数实现。3.1) Predict the total traffic volume of each road section according to historical data, and use BP neural network to predict, which can be realized by directly calling the built-in function in matlab.
在本实施例中，预测结果为如表5所示。In this embodiment, the prediction results are shown in Table 5.
表5高速公路各路段总交通量预测值Table 5 Predicted value of total traffic volume in each section of expressway
3.2)计算预约交通总量和非预约交通总量其计算公式为：3.2) Calculating the total amount of reserved traffic and total walk-in traffic Its calculation formula is:
式中：——预测时段ti时路网的总交通量；In the formula: ——the total traffic volume of the road network at the forecast period ti ;
——预测时段ti时路网的预约交通总量； ——the total amount of scheduled traffic on the road network at the forecast time period ti ;
——预测时段ti时路网的非预约交通总量； ——the total amount of non-reserved traffic on the road network at the time of prediction period ti ;
——预测时段ti时第j条高速公路的第kj个路段的交通量； ——the traffic volume of the k j-th section of thej- th expressway at the forecast time period ti ;
α——用户app或微信公众号的市场占有率。α——The market share of the user's app or WeChat official account.
其余各参数意义如前所述。The meanings of other parameters are as mentioned above.
计算结果如表6所示。The calculation results are shown in Table 6.
表6预约交通量和非预约交通量计算结果Table 6 Calculation results of reserved traffic volume and non-reserved traffic volume
3.3)按照车流量的历史分布规律，对非预约交通总量在路网范围内进行分配，得到各路段的非预约量，其计算公式为：3.3) According to the historical distribution law of traffic volume, the total amount of non-reserved traffic is allocated within the scope of the road network, and the non-reserved volume of each road section is obtained. The calculation formula is:
式中：——预测时段ti时第j条高速公路的第kj个路段的非预约交通量。In the formula: ——The non-reserved traffic volume of the kjth section of the jth expressway at the forecast time period ti .
计算结果如表7所示。The calculation results are shown in Table 7.
表7非预约量分配结果Table 7 Distribution results of non-reserved quantities
3.4)采用用户均衡模型对预约交通总量在路网范围内进行分配，得到各路段的预约量。3.4) Use the user balance model to distribute the total amount of reserved traffic within the road network, and obtain the reserved amount of each road section.
式中：——预测时段ti时第j条高速公路的流量。In the formula: ——The flow rate of thejth expressway at the forecast time period ti .
——预约用户在路网中的路阻函数，其具体形式为： ——The road resistance function of the reservation user in the road network, its specific form is:
式中：to——自由流条件下的行驶时间，min；In the formula: to ——travel time under free flow condition, min;
——第j条高速公路的第kj个路段的道路设计通行能力，pcu/h； ——The road design capacity of the kjth road section of thejth expressway, pcu/h;
co——基础通行费用，元；co ——basic toll fee, yuan;
γ——优惠幅度，％；γ——preferential margin, %;
α,β——待标定的参数；α, β——parameters to be calibrated;
τ——旅行时间价值，元/h。τ——travel time value, yuan/h.
本实施例中，参数取值为α＝0.5,β＝1。认为走高速公路2的用户为错峰出行，高速公路1的用户非错峰出行。故两条高速公路的阻抗函数分别为：In this embodiment, the parameter values are α=0.5, β=1. It is considered that the users who take the expressway 2 travel on a peak-staggered trip, and the users on the expressway 1 travel on a non-peak-staggered trip. Therefore, the impedance functions of the two expressways are:
求解二元一次方程组得到预约量的分配结果如表8所示。Solve a system of linear equations in two variables The allocation results of the reserved amount are shown in Table 8.
表8预约量分配结果Table 8 Allocation Results of Appointment Amount
3.5)采用实际预约量对各路段的预约量进行修正，修正规则如下：3.5) Use the actual reservation quantity to correct the reservation quantity of each road section, and the correction rules are as follows:
①若则不变。① if but constant.
②若则② if but
式中：——预测时段ti时第j条高速公路的第kj个路段的实际预约交通总量；In the formula: ——The actual scheduled traffic volume of the kjth road section of thejth expressway at the time of prediction period ti ;
修正后的结果如表9所示：The corrected results are shown in Table 9:
表9预约量分配的修正结果Table 9 Correction results of reservation allocation
3.6)将各路段预约量和非预约量进行叠加，得到各路段的交通量。3.6) Superimpose the reserved volume and non-reserved volume of each road section to obtain the traffic volume of each road section.
式中：——预测时段ti时第j条高速公路的第kj个路段的预约交通量。In the formula: ——Reserved traffic volume of thekj- th section of the j-th expressway at the forecast time period ti .
最终的预测结果如表10所示：The final prediction results are shown in Table 10:
表10路段交通量最终预测结果Table 10 The final prediction results of road traffic volume
4)预测各路段的拥挤度：采用饱和度作为拥挤的评价指标，通过阈值判断各路段的拥挤状态。具体包括以下步骤：4) Predict the congestion degree of each road section: the saturation is used as the evaluation index of congestion, and the congestion state of each road section is judged by the threshold. Specifically include the following steps:
4.1)计算各路段的饱和度，计算公式为：4.1) Calculate the saturation of each road section, the calculation formula is:
式中：——用户选择出行时段ti时第j条高速公路的第kj个路段的饱和度。In the formula: ——the saturation of the kjth section of thejth expressway when the user selects the travel period ti .
饱和度计算结果如表11所示。The saturation calculation results are shown in Table 11.
表11饱和度计算结果Table 11 Saturation Calculation Results
4.2)拥挤度等级的划分，通过饱和度阈值将拥挤度划分成不同的等级。可以根据各城市的交通调查数据来确定阈值。在无调查资料的情况下，建议使用表12进行分级。4.2) Classification of congestion degree, the degree of congestion is divided into different grades by saturation threshold. The threshold can be determined based on the traffic survey data of each city. In the absence of survey data, it is recommended to use Table 12 for grading.
表22Table 22
5)利用程序设计语言实现拥挤预测结果的可视化输出。将不同的拥挤度等级与不同的颜色对应，在编写代码时将相应的颜色代码赋予各路段。运行代码，即可输出拥挤预测结果。输出结果如图3所示。5) Use the programming language to realize the visual output of the congestion prediction results. Correspond to different congestion levels with different colors, and assign corresponding color codes to each road segment when writing codes. Run the code to output the congestion prediction results. The output result is shown in Figure 3.
综上所述，在采用以上方案后，本发明为高速公路拥挤预测提供了新的方法，能够适应高速公路预约通行带来的新变化，在新的条件下有效地预测高速公路的拥挤情况，为高速公路的出行诱导和管理工作提供更可靠的基础，有效推动我国高速公路的发展，具有实际推广价值，值得推广。In summary, after adopting the above scheme, the present invention provides a new method for expressway congestion prediction, which can adapt to the new changes brought about by expressway reservation, and effectively predict the congestion situation of expressway under new conditions. It provides a more reliable foundation for the travel guidance and management of expressways, and effectively promotes the development of expressways in my country. It has practical promotion value and is worth promoting.
以上所述实施例只为本发明之较佳实施例，并非以此限制本发明的实施范围，故凡依本发明之形状、原理所作的变化，均应涵盖在本发明的保护范围内。The above-described embodiments are only preferred embodiments of the present invention, and are not intended to limit the scope of the present invention. Therefore, all changes made according to the shape and principles of the present invention should be covered within the protection scope of the present invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610806611.4ACN106327871B (en) | 2016-09-06 | 2016-09-06 | A kind of crowded prediction technique of highway of fusion historical data and reservation data |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610806611.4ACN106327871B (en) | 2016-09-06 | 2016-09-06 | A kind of crowded prediction technique of highway of fusion historical data and reservation data |
| Publication Number | Publication Date |
|---|---|
| CN106327871A CN106327871A (en) | 2017-01-11 |
| CN106327871Btrue CN106327871B (en) | 2018-09-14 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610806611.4AActiveCN106327871B (en) | 2016-09-06 | 2016-09-06 | A kind of crowded prediction technique of highway of fusion historical data and reservation data |
| Country | Link |
|---|---|
| CN (1) | CN106327871B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106920395B (en)* | 2017-04-21 | 2019-04-12 | 杭州市综合交通研究中心 | A kind of traffic impedance computation method based on parameter calibration |
| WO2018205067A1 (en)* | 2017-05-08 | 2018-11-15 | 深圳市卓希科技有限公司 | Method and system for detecting volume of traffic |
| CN107481533A (en)* | 2017-09-23 | 2017-12-15 | 山东交通学院 | A traffic flow forecasting system and method |
| CN108510091A (en)* | 2018-03-16 | 2018-09-07 | 厦门华方软件科技有限公司 | A method of the reservation right of way based on electronic license plate |
| CN108492558B (en)* | 2018-03-27 | 2021-08-17 | 深圳大学 | Expressway travel reservation method, storage medium and terminal |
| CN108986459B (en)* | 2018-07-06 | 2020-12-22 | 华南理工大学 | Expressway congestion dredging method based on reserved traffic |
| CN109118789B (en)* | 2018-08-24 | 2020-11-06 | 交通运输部规划研究院 | Multi-source data fusion method and device for highway dispatching station |
| CN110021164B (en)* | 2019-03-02 | 2020-09-04 | 合肥学院 | Analysis method of road network occupancy rate of online car-hailing based on travel time data |
| CN110021163B (en)* | 2019-03-02 | 2020-10-13 | 合肥学院 | Network appointment road network occupancy analysis method based on travel mileage data |
| CN112749825B (en)* | 2019-10-31 | 2024-08-02 | 华为云计算技术有限公司 | Method and device for predicting destination of vehicle |
| CN111985716B (en)* | 2020-08-21 | 2024-05-14 | 北京交通大学 | Passenger traffic volume prediction system with passenger traffic information visualization function |
| CN114882696B (en)* | 2020-10-28 | 2023-11-03 | 华为技术有限公司 | Road capacity determination method, device and storage medium |
| CN113077281B (en)* | 2021-03-12 | 2023-02-17 | 中山大学 | A method and device for predicting the distribution of subway passenger flow |
| CN114548554A (en)* | 2022-02-22 | 2022-05-27 | 西藏招商交建电子信息有限公司 | Method, device, equipment and storage medium for predicting effect of differentiated charging strategy |
| CN114550457A (en)* | 2022-03-11 | 2022-05-27 | 中天科技(清远)有限公司 | Traffic flow prediction method for expressway service area |
| CN115909766A (en)* | 2022-12-20 | 2023-04-04 | 同济大学 | Reservation travel method for urban expressway traffic |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101819717A (en)* | 2010-03-05 | 2010-09-01 | 吉林大学 | Road network performance judgment method based on traffic state space-time model |
| CN103106787A (en)* | 2012-12-21 | 2013-05-15 | 周晓东 | System for proactively solving urban traffic congestion |
| CN104200649A (en)* | 2014-08-25 | 2014-12-10 | 沈阳工业大学 | System and method for dispatching and distributing peak traffic hour route resources based on application in advance |
| CN105719019A (en)* | 2016-01-21 | 2016-06-29 | 华南理工大学 | Public bicycle peak time demand prediction method considering user reservation data |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4003828B2 (en)* | 2002-03-29 | 2007-11-07 | 富士通エフ・アイ・ピー株式会社 | Road control method, road control system, and recording medium |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101819717A (en)* | 2010-03-05 | 2010-09-01 | 吉林大学 | Road network performance judgment method based on traffic state space-time model |
| CN103106787A (en)* | 2012-12-21 | 2013-05-15 | 周晓东 | System for proactively solving urban traffic congestion |
| CN104200649A (en)* | 2014-08-25 | 2014-12-10 | 沈阳工业大学 | System and method for dispatching and distributing peak traffic hour route resources based on application in advance |
| CN105719019A (en)* | 2016-01-21 | 2016-06-29 | 华南理工大学 | Public bicycle peak time demand prediction method considering user reservation data |
| Publication number | Publication date |
|---|---|
| CN106327871A (en) | 2017-01-11 |
| Publication | Publication Date | Title |
|---|---|---|
| CN106327871B (en) | A kind of crowded prediction technique of highway of fusion historical data and reservation data | |
| CN102819955B (en) | Road network operation evaluation method based on vehicle travel data | |
| CN103956050B (en) | Road network postitallation evaluation methods based on vehicle travel data | |
| CN104157139B (en) | A traffic jam prediction method and visualization method | |
| CN105096643B (en) | Real-time public transport arrival time Forecasting Methodology based on multi-line front truck service data | |
| CN103745089A (en) | Multi-dimensional public transport operation index evaluation method | |
| CN105096615B (en) | Signalling-unit-based adaptive optimization control system | |
| CN103198658B (en) | Urban road traffic state non-equilibrium degree detection method | |
| CN109416879A (en) | A kind of preferential short berth classification Dynamic Pricing method stopped | |
| CN107895481B (en) | Regional road vehicle flow control method based on floating vehicle technology | |
| CN110288205B (en) | Traffic influence evaluation method and device | |
| CN110633558B (en) | A Modeling System of Urban Traffic System | |
| CN106816009A (en) | Highway real-time traffic congestion road conditions detection method and its system | |
| CN106157618B (en) | A kind of urban public transportation lane plans cloth network method | |
| CN106601005B (en) | An urban intelligent traffic guidance method based on RFID and WeChat platform | |
| CN107085943A (en) | A kind of road travel time short term prediction method and system | |
| CN106710216A (en) | Expressway real-time traffic jam road condition detection method and system | |
| CN105894814A (en) | Joint optimization method and system for multiple traffic management and control measures in consideration of environmental benefits | |
| CN104809112A (en) | Method for comprehensively evaluating urban public transportation development level based on multiple data | |
| CN103279669A (en) | Method and system for simulating calculation of transport capacity of urban rail transit network | |
| CN101777260B (en) | Harbour district traffic flow forecasting method under reserved harbour concentration mode | |
| CN115424432B (en) | Upstream diversion method based on multisource data under expressway abnormal event | |
| CN108986459A (en) | A kind of highway crowded leading method current based on reservation | |
| CN105741548A (en) | Method for generating traffic state cloud atlas | |
| CN105160867B (en) | Traffic message Forecasting Methodology |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |