Method for determining congestion degree of target area and electronic equipment

Abstract

The invention discloses a method for determining the congestion degree of a target area and electronic equipment. The method comprises the following steps: determining the target area; the target area comprises M peripheral monitoring points and N boundary intersections; applying a first prediction model to predict a first in-transit vehicle number in the first time period based on the first parameter set; the first prediction model is a time-phased multiple linear regression model; using a second prediction model to predict a second number of vehicles in transit in the first time period based on the second parameter set; the second prediction model is a Holt-Winters model; selecting the first in-transit vehicle number and the second in-transit vehicle number as the target in-transit vehicle number of the first time period; and if the number of the target in-transit vehicles meets a preset congestion condition, determining that the target area is congested in the first time period. By applying the mode, the more accurate number of vehicles on the way can be obtained, the number of vehicles on the way is more intuitive as a judgment standard of regional congestion, and the number of vehicles needing to be controlled on the periphery of the region can be quantified during regional congestion slowing.

Description

Technical Field

[0001] The present application relates to the field of traffic technology, and in particular to a method and electronic device for determining the congestion level of a target area. Background Art

[0002] Intelligent transportation system is a comprehensive transportation management system that is real-time and accurate and plays a role in a wide range and in all directions. Traffic control and traffic induction are important means to relieve traffic congestion, and the current congestion is an influencing factor in the process of traffic control or traffic induction.

[0003] In related technologies, the indicators for measuring regional congestion are mainly the proportion of severely congested mileage in the road network, the travel time ratio of the road network, the delay time ratio of the road network and the regional traffic index, etc. However, the regional congestion situation determined by using these indicators is quite different from the actual congestion situation, which is inaccurate and not intuitive. Summary of the invention

[0004] The exemplary implementation of the present application provides a method and electronic device for determining the degree of congestion in a target area, which can obtain a relatively accurate number of vehicles on the road. It is more intuitive to use it as a criterion for judging regional congestion. When the regional congestion is relieved, the number of vehicles that need to be controlled outside the area can be quantified, thereby improving the effect of evacuating traffic congestion.

[0005] According to a first aspect of an exemplary embodiment, a method for determining a congestion level of a target area is provided, comprising:

[0006] Determine a target area; wherein the target area includes M peripheral monitoring points and N boundary intersections; M and N are integers respectively;

[0007] Applying a first prediction model, based on a first parameter set, predicting a first number of vehicles in transit in a first time period; wherein the first prediction model is a time-segmented multivariate linear regression model; the first parameter set includes at least one parameter that affects the first number of vehicles in transit; the first parameter set is determined based on first vehicle passing data monitored by M peripheral monitoring points and N boundary intersections;

[0008] Applying a second prediction model, based on a second parameter set, predicting the second number of vehicles on the way in the first time period; wherein the second prediction model is a Holt-Winters model, and the second parameter set is the number of vehicles on the way in the first historical time period before the current moment; the second parameter set is determined based on the second vehicle passing data monitored by M peripheral monitoring points and N boundary intersections;

[0009] Selecting a first number of vehicles in transit and a second number of vehicles in transit as a target number of vehicles in transit for a first time period;

[0010] If the target number of vehicles in transit meets the preset congestion condition, it is determined that the target area is congested in the first time period.

[0011] In the embodiment of the present application, the number of vehicles in transit is used as a criterion for determining whether there is congestion. In the process of predicting the number of vehicles in transit, the corresponding parameter set can be determined based on the vehicle passing data monitored by M peripheral monitoring points and N boundary intersections, and then the time-divided multivariate linear regression model and the Holt-Winters model are respectively applied for prediction to obtain the first number of vehicles in transit and the second number of vehicles in transit, and then one of them is selected as the target number of vehicles in transit. When the target number of vehicles in transit meets the preset congestion condition, it is determined that the target area is congested in the first time period. Using this method, a more accurate number of vehicles in transit can be obtained. Using the number of vehicles in transit as a criterion for judging regional congestion is more intuitive and the control target can be quantified. In addition, the predicted degree of congestion is more accurate, and the number of vehicles that need to be controlled outside the area can be quantified when the area is congested.

[0012] In an optional implementation, before applying the first prediction model to predict the first number of vehicles in transit in the first time period based on the first parameter set, the method further includes:

[0013] Constructing a first prediction model based on the historical parameter set;

[0014] Among them, the historical parameter set includes multiple groups of historical parameters, each group of historical parameters includes the number of vehicles entering the area within a first set time after the second historical time period, the number of vehicles leaving the area within a second set time before the second historical time period, the number of vehicles in transit in the second historical time period, and the average value of the number of vehicles in transit in each time interval with the same time attribute as the first time period.

[0015] In the above embodiment, when constructing the first prediction model, the historical parameter set takes into account the corresponding vehicle passing data before and after the second historical time period, and at the same time, also takes into account the attributes of the first time period to be predicted. The first prediction model constructed in this way is used to predict the number of vehicles in transit in the first time period, and the prediction result is more accurate.

[0016] In an optional implementation, applying the first prediction model to predict the first number of vehicles in transit in the first time period based on the first parameter set includes:

[0017] Inputting the first parameter set into the first prediction model to obtain a first number of vehicles in transit;

[0018] Among them, the first parameter set includes the number of vehicles entering the area within the first time period after the current moment, the number of vehicles leaving the area within a third set time period before the current moment that is consistent with the length of the first time period, the number of vehicles in transit at the current moment, and the average value of the number of vehicles in transit in each time interval with the same time attribute in the first time period.

[0019] In the above embodiment, each vehicle passing data at the current moment is input into the first prediction model as the first parameter set, so as to obtain the first number of vehicles in transit after the first time period after the current moment.

[0020] In an optional embodiment, the method further includes:

[0021] For each boundary intersection, the sum of the number of vehicles arriving at the boundary intersection from each peripheral monitoring point associated with the boundary intersection within the first time period is calculated as the number of vehicles entering the boundary intersection;

[0022] The sum of the number of vehicles entering the zone at each boundary intersection is determined to be the number of vehicles entering the zone in the first time period.

[0023] In the above embodiment, in the process of calculating the number of vehicles entering the zone in the first time period in the first parameter set, each boundary intersection is calculated separately and then superimposed. In the process of calculating the number of vehicles entering the zone at each intersection, each peripheral monitoring point associated with the boundary intersection is taken into account, and the calculation result is more accurate.

[0024] In an optional embodiment, the method further includes:

[0025] For each boundary intersection, determine a historical OD traffic flow regularity data table of the boundary intersection; wherein the historical OD traffic flow regularity data table includes P data items, each of which includes a characteristic day within a historical date range, and vehicle passing information from a peripheral monitoring point to a boundary intersection; the vehicle passing information includes the ratio of the number of vehicles arriving at the boundary intersection to the number of vehicles leaving the peripheral monitoring point, the departure time, the time from the peripheral monitoring point to the boundary intersection, and the average arrival time;

[0026] Match the first time period with the historical OD traffic pattern data table to determine the matched historical arrival time period;

[0027] Determine the departure time period based on the matched historical arrival time period and average arrival time;

[0028] The number of vehicles passing through the departure time period, the corresponding ratio, and the set coefficient are used to determine the predicted number of vehicles entering the area at a boundary intersection; wherein the set coefficient is the degree of overlap between the matched historical arrival time period and the first time period.

[0029] In the above embodiment, in the process of calculating the predicted number of vehicles entering the zone at a boundary intersection in the first parameter set, a historical OD traffic pattern data table for each boundary intersection is constructed, and the first time period to be predicted is matched with the OD traffic pattern data table, and then the predicted number of vehicles entering the zone at a boundary intersection is calculated, and the prediction result is more accurate.

[0030] In an optional implementation, applying the second prediction model to predict the second number of vehicles in transit in the first time period based on the second parameter set includes:

[0031] Obtain a second parameter set; wherein the second parameter set is the number of vehicles in transit in a third historical time period before the current moment;

[0032] The number of vehicles in transit in the third time period is input into the pre-constructed Holt-Winters model to obtain the second number of vehicles in transit in the first time period.

[0033] In the above embodiment, considering that the number of vehicles in transit has characteristics such as trend, seasonality and randomness, the Holt-Winters model is applied here to make full use of its common characteristics with traffic flow characteristics, and the predicted number of vehicles in transit in the second time period in the first time period is more in line with the actual situation of traffic flow.

[0034] In an optional implementation, selecting the first number of vehicles in transit and the second number of vehicles in transit as the target number of vehicles in transit for the first time period includes:

[0035] Compare the actual number of vehicles on the road detected at the current moment with the first historical number of vehicles on the road predicted at the first historical moment by applying the first prediction model to determine a first error;

[0036] Compare the actual number of vehicles on the road detected at the current moment with the second historical number of vehicles on the road predicted at the second historical moment by applying the second prediction model to determine a second error;

[0037] If the first error is less than the second error, selecting the first number of vehicles in transit as the target number of vehicles in transit for the first time period;

[0038] If the second error is smaller than the first error, the second number of vehicles in transit is selected as the target number of vehicles in transit for the first time period.

[0039] In the above embodiment, the first prediction model and the second prediction model have different characteristics. Therefore, in the actual application process, both prediction models can be used to perform predictions in real time, and the prediction results can be stored separately to determine which prediction result will be used as the final prediction result at the next moment, that is, the target number of vehicles in transit.

[0040] According to a second aspect of the exemplary implementation, there is provided an electronic device, including a processor and a memory;

[0041] Memory, configured to execute:

[0042] Store the peripheral monitoring points and boundary intersections corresponding to each area;

[0043] A processor configured to execute:

[0044] Determine a target area; wherein the target area includes M peripheral monitoring points and N boundary intersections; M and N are integers respectively;

[0045] Applying a first prediction model, based on a first parameter set, predicting a first number of vehicles in transit in a first time period; wherein the first prediction model is a time-segmented multivariate linear regression model; the first parameter set includes at least one parameter that affects the first number of vehicles in transit; the first parameter set is determined based on first vehicle passing data monitored by M peripheral monitoring points and N boundary intersections;

[0046] Applying a second prediction model, based on a second parameter set, predicting the second number of vehicles on the way in the first time period; wherein the second prediction model is a Holt-Winters model, and the second parameter set is the number of vehicles on the way in the first historical time period before the current moment; the second parameter set is determined based on the second vehicle passing data monitored by M peripheral monitoring points and N boundary intersections;

[0047] Selecting a first number of vehicles in transit and a second number of vehicles in transit as a target number of vehicles in transit for a first time period;

[0048] If the target number of vehicles in transit meets the preset congestion condition, it is determined that the target area is congested in the first time period.

[0049] According to a third aspect of an exemplary embodiment, there is provided a device for determining a congestion level of a target area, comprising:

[0050] The processing unit is used to: determine a target area; wherein the target area includes M peripheral monitoring points and N boundary intersections; M and N are integers respectively;

[0051] The prediction unit is used to: apply a first prediction model and predict a first number of vehicles in transit in a first time period based on a first parameter set; wherein the first prediction model is a time-segmented multivariate linear regression model; the first parameter set includes at least one parameter that affects the first number of vehicles in transit; the first parameter set is determined based on first vehicle passing data monitored by M peripheral monitoring points and N boundary intersections;

[0052] The prediction unit is further used to: apply a second prediction model, based on a second parameter set, to predict the number of second vehicles in transit in the first time period; wherein the second prediction model is a Holt-Winters model, and the second parameter set is the number of vehicles in transit in the first historical time period before the current moment; and the second parameter set is determined based on the second vehicle passing data monitored by M peripheral monitoring points and N boundary intersections;

[0053] The processing unit is further used to: select the first number of vehicles in transit and the second number of vehicles in transit as the target number of vehicles in transit for the first time period;

[0054] The processing unit is further used to: if the target number of vehicles in transit meets a preset congestion condition, determine that the target area is congested in the first time period.

[0055] According to a fourth aspect of the exemplary implementation, a computer storage medium is provided, in which computer program instructions are stored. When the instructions are executed on a computer, the computer executes the method for determining the congestion level of a target area as described in the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0057] Figure 1 A schematic diagram of a target area provided in an embodiment of the present application is exemplarily shown;

[0058] Figure 2 A flowchart of a method for determining the congestion level of a target area provided by an embodiment of the present application is exemplarily shown;

[0059] Figure 3 A flowchart of a method for determining the number of vehicles entering a zone in a first time period in a first parameter set provided by an embodiment of the present application is exemplarily shown;

[0060] Figure 4 A flowchart of a method for calculating the predicted number of vehicles entering a boundary intersection in a first parameter set provided by an embodiment of the present application is exemplarily shown;

[0061] Figure 5 A schematic diagram showing a comparison between a predicted result and an actual result of the first number of vehicles in transit on a certain day provided by an embodiment of the present application is exemplified;

[0062] Figure 6 A schematic diagram showing, by way of example, a decomposition result of a sequence of the number of vehicles in transit in a certain time period provided by an embodiment of the present application;

[0063] Figure 7 A flowchart of a method for determining the number of second vehicles in transit provided by an embodiment of the present application is exemplarily shown;

[0064] Figure 8 A flowchart of a method for determining the target number of vehicles in transit provided by an embodiment of the present application is exemplarily shown;

[0065] Fig. 9 A schematic diagram of the structure of a device for determining the congestion level of a target area provided in an embodiment of the present application is exemplarily shown;

[0066] Fig.10 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application is exemplarily shown. DETAILED DESCRIPTION

[0067] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0068] For ease of understanding, the terms involved in the embodiments of the present application are explained below.

[0069] (1) Short-term traffic flow prediction: Intelligent transportation system is a comprehensive transportation management system that functions in a wide range and in all directions, in real time and accurately. Traffic control and traffic induction are important areas of research in intelligent transportation systems. One of the key issues in achieving traffic control and induction is real-time and accurate short-term traffic flow prediction. Short-term traffic flow prediction refers to traffic flow prediction with a prediction time span of no more than 15 minutes. It is in the micro sense and is essentially different from the prediction time interval in the meso and macro sense (traffic flow prediction calculated in hours, days, or even months). The content of short-term traffic flow prediction generally includes traffic volume, traffic speed, traffic density (or occupancy), travel time, etc. Accurate and real-time traffic flow prediction is the key to transforming the intelligent transportation system from "passive reaction" to "active action". The quality of the prediction result is directly related to the effect of control and induction. Therefore, the short-term traffic flow prediction model should be real-time, accurate and reliable.

[0070] Indicators used in related technologies to measure regional congestion are mainly the proportion of severely congested mileage in the road network, the travel time ratio of the road network, the delay time ratio of the road network and the regional traffic index, etc. However, the regional congestion conditions determined by using these indicators are quite different from the actual congestion conditions, and are inaccurate and not intuitive.

[0071] To this end, an embodiment of the present application provides a method for determining the degree of congestion in a target area, in which the number of vehicles in transit is used as an indicator to measure whether the area is congested. When calculating the number of vehicles in transit, you can first apply a time-divided multivariate linear regression model to predict based on a first parameter set to obtain a first number of vehicles in transit; then apply the Holt-Winters model to calculate and predict based on a second parameter set to obtain a second number of vehicles in transit. In addition, the first number of vehicles in transit and the second number of vehicles in transit are selected as the target number of vehicles in transit for the first time period. Using this method, a more accurate number of vehicles in transit can be obtained, and using the number of vehicles in transit as a criterion for judging regional congestion is more intuitive and the control target can be quantified. In addition, when the area is relieved from congestion, the number of vehicles that need to be controlled outside the area can be quantified.

[0072] After introducing the design ideas of the embodiments of the present application, the following briefly introduces the application scenarios to which the technical solutions of the embodiments of the present application can be applied. It should be noted that the application scenarios introduced below are only used to illustrate the embodiments of the present application and are not limited. In specific implementation, the technical solutions provided by the embodiments of the present application can be flexibly applied according to actual needs.

[0073] refer to Figure 1 , shows a schematic diagram of a target area, which includes M peripheral monitoring points and N boundary intersections, where M and N are integers respectively.

[0074] Take a target area as an example. The target area includes 9 peripheral monitoring points and 6 boundary intersections. The boundary intersections refer to the intersections that must be passed through the target area. Therefore, the boundary intersections can be regarded as points D and the peripheral monitoring points as points O. Each point O to point D can be regarded as a channel.

[0075] Among them, vehicles entering the area are an important factor causing the number of vehicles in transit in the area to increase or decrease, especially when a large number of vehicles enter and exit the area, this phenomenon is more obvious. Therefore, the number of vehicles in each entry channel is first predicted, and on this basis, the number of vehicles in transit in the area is predicted. On the one hand, the number of vehicles entering the area is one of the factors affecting the number of vehicles in transit in the area. On the other hand, the number of vehicles entering the area can be used to accurately calculate the flow control of each channel, providing an accurate adjustment basis for the signal control scheme of each channel and each intersection. In an embodiment of the present application, the traffic flow prediction of the entry channel can be achieved based on a detailed OD traffic law.

[0076] To further illustrate the technical solution provided by the embodiment of the present application, this is described in detail below in conjunction with the accompanying drawings and specific implementation methods. Although the embodiment of the present application provides the method operation steps shown in the following embodiments or drawings, more or fewer operation steps may be included in the method based on routine or no creative labor. In the steps where there is no necessary causal relationship logically, the execution order of these steps is not limited to the execution order provided in the embodiment of the present application.

[0077] Combine the following Figure 1 The application scenario shown is shown in Figure 2 A flow chart of a method for determining the congestion level of a target area is shown to illustrate the technical solution provided in an embodiment of the present application.

[0078] S201: Determine the target area.

[0079] S202: Apply a first prediction model to predict a first number of vehicles in transit in a first time period based on a first parameter set.

[0080] S203: Apply a second prediction model to predict the second number of vehicles in transit in the first time period based on a second parameter set.

[0081] S204: Selecting a first number of vehicles on the way and a second number of vehicles on the way as a target number of vehicles on the way for the first time period.

[0082] S205: If the target number of vehicles in transit meets a preset congestion condition, it is determined that the target area is congested in the first time period.

[0083] In the embodiment of the present application, the number of vehicles in transit is used as a criterion for determining whether there is congestion. In the process of predicting the number of vehicles in transit, the corresponding parameter set can be determined based on the vehicle passing data monitored by M peripheral monitoring points and N boundary intersections, and then the time-divided multivariate linear regression model and the Holt-Winters model are respectively applied for prediction to obtain the first number of vehicles in transit and the second number of vehicles in transit, and then one of them is selected as the target number of vehicles in transit. When the target number of vehicles in transit meets the preset congestion condition, it is determined that the target area is congested in the first time period. Using this method, a more accurate number of vehicles in transit can be obtained. Using the number of vehicles in transit as a criterion for judging regional congestion is more intuitive and the control target can be quantified. In addition, the predicted degree of congestion is more accurate, and the number of vehicles that need to be controlled outside the area can be quantified when the area is congested.

[0084] In relation to S201, the target area in the embodiment of the present application can be found in Figure 1 The target area is shown.

[0085] In S202, a first prediction model is constructed according to a set of historical parameters. The first prediction model may be a time-divided multivariate linear regression model.

[0086] In actual situations, when the area is relatively large, the vehicles on the way in the area are composed of three parts: vehicles entering the area, vehicles leaving the area, and vehicles traveling inside the area. According to the actual operation situation, the influencing factors and characteristics of the vehicles on the way can be obtained: the number of vehicles entering the area, at any time, when a large number of vehicles from the periphery enter the core area, the number of vehicles on the way will increase rapidly; the trend of the vehicles on the way itself is consistent with the trend of the number of vehicles on the way in the area, which plays an important role in predicting the periodic trend of the number of vehicles on the way in the area; the historical law of the number of vehicles on the way; the number of vehicles leaving the area, at different times, the traffic volume in and out of the area accounts for a relatively stable proportion of the total number of vehicles on the way in the area.

[0087] Therefore, when constructing the first prediction model, the above-mentioned influencing factors and factor characteristics can be referred to to obtain multiple groups of historical parameters, which are called historical parameter sets. Each group of historical parameters is the historical data of a period of time. For example, according to the ratio of the number of vehicles entering and leaving the area to the number of vehicles in transit in the area, combined with the actual commuting rules, the whole day is divided into 7 time intervals: 0:00-5:00, 5:00-8:30, 8:30-11:30, 11:30-14:00, 14:00-17:00, 17:00-20:30, and 20:30-24:00. If the collection period of the historical parameters is 5 minutes, take any one of the time intervals (for example, 14:00-17:00) as an example, and one of the 5-minute time periods is called the second historical time period (for example, 14:35-14:40), for example.

[0088] Among them, the historical parameters corresponding to the second time period include the number of vehicles entering the area within the first set time (15 minutes) after the second historical time period (14:35-14:40), the number of vehicles leaving the area within the second set time (15 minutes) before the second historical time period (14:35-14:40), the number of vehicles in transit in the second historical time period (14:35-14:40), and the average value of the number of vehicles in transit in each time interval (8:30-11:30, 11:30-14:00, 14:00-17:00) with the same attribute as the first time period (15 minutes after the current time to be predicted) (the current time is, for example, 14:48 on Wednesday, then its attribute is daytime on weekdays).

[0089] The prediction model of the number of vehicles in transit is established by time period. Here, the prediction model of one time period is called the first prediction model. The model parameters are obtained by the least squares method:

[0090] y=b0+b1x1+b2x2+b3x3+b4x4;

[0091] Among them, y is the number of vehicles on the way in the next 15 minutes of the second historical time period, x1 is the number of vehicles entering the area in the next 15 minutes of the second historical time period, x2 is the number of vehicles leaving the area 15 minutes before the second historical time period, x3 is the number of vehicles on the way in the second historical time period, and x4 is the average number of vehicles on the way in each time interval with the same attribute as the first time period. If the attribute of the second historical time period is working day, then x4 is the average number of vehicles on the way in the same period of the five working days last week.

[0092] The above four characteristics are the key influencing factors of the number of vehicles on the road in the region. x1 greatly reduces the lag phenomenon in the prediction, x3 ensures that the prediction result does not deviate too much from the current number of vehicles on the road, and x4 can reduce the prediction lag when the traffic flow is relatively stable and similar on different days.

[0093] After obtaining the model parameters b0, b1, b2, b 3、 After b4, the construction of the first prediction model is completed. It should be noted that the first prediction model for each time interval is different, and can be constructed by referring to the above model construction process, which will not be repeated here.

[0094] After obtaining the first prediction model, the first parameter set is input into the first prediction model to obtain a first number of vehicles in transit.

[0095] The first time period may be 15 minutes after the current moment. The first parameter set is determined based on the first vehicle passing data monitored by M peripheral monitoring points and N boundary intersections. For example, the first parameter set includes at least one parameter that affects the first number of vehicles in transit, such as the number of vehicles entering the area within the first time period after the current moment, the number of vehicles leaving the area within a third set time period before the current moment that is consistent with the first time period, the number of vehicles in transit at the current moment, and the average value of the number of vehicles in transit in each time interval with the same time attribute of the first time period.

[0096] Among them, reference Figure 3 , the number of vehicles entering the zone in the first time period in the first parameter set can be calculated:

[0097] S301: For each boundary intersection, the sum of the number of vehicles arriving at the boundary intersection from each peripheral monitoring point associated with the boundary intersection within a first time period is calculated as the number of vehicles entering the boundary intersection.

[0098] Taking any boundary intersection as an example, the traffic volume of the boundary intersection in the first time period (next 15 minutes) can be calculated using the following formula:

[0099]

[0100] Entrance_flowi Monitor_flow is the traffic flow of the i-th boundary intersection in the next 15 minutes. j Monitor_flow is the traffic flow that will arrive at the i-th boundary intersection 15 minutes after the current j-th peripheral monitoring point. j The calculation method can be seen as follows Figure 4 .

[0101] S302: Determine the sum of the number of vehicles entering the zone at each boundary intersection as the number of vehicles entering the zone in the first time period.

[0102] Optionally, the number of vehicles entering the zone at each boundary intersection may be added together to obtain the number of vehicles entering the zone in the first time period.

[0103] Among them, reference Figure 4 , the predicted number of vehicles entering the zone at a boundary intersection in the first parameter set can be calculated:

[0104] S401: For each boundary intersection, determine a historical OD traffic flow pattern data table of the boundary intersection.

[0105] For example, the subdivision conditions for calculating the OD traffic flow pattern can be set. Among the vehicles passing through each peripheral monitoring point, there are vehicles with different license plate locations, such as local vehicles and non-local vehicles; on different characteristic days, such as working days and non-working days; at different time periods, such as morning rush hour and evening rush hour, there will be different proportions of vehicles entering the area, so multiple vehicle and date characteristics are set as the subdivision conditions of the OD pattern.

[0106] Table 1 OD rule subdivision conditions table

[0107]

[0108]

[0109] According to the subdivision conditions in Table 1, calculate the OD traffic flow pattern. Count the proportion of vehicles entering each boundary intersection (point D) after passing a certain peripheral monitoring point (point O) on different characteristic days, different time periods, and different number locations. Generate a traffic OD traffic flow pattern table. Optionally, Table 2 is an example of an OD traffic flow pattern table generated, called a historical OD traffic flow pattern data table, including P data, each of which includes a characteristic day within the historical date range, and the passing information from a peripheral monitoring point to a boundary intersection; the passing information includes the proportion of vehicles arriving at the boundary intersection to the number of vehicles leaving the peripheral monitoring point, the departure time, the time from the peripheral monitoring point to the boundary intersection, and the average arrival time. In this example, the boundary intersection D1 is taken as an example for illustration.

[0110] Table 2 Historical OD traffic flow data of boundary intersections

[0111]

[0112] It should be noted that the actual names of the peripheral monitoring points and the names of the boundary intersections in Table 2 are replaced by numbers, and Table 2 only shows the relevant data of some peripheral monitoring points and boundary intersections.

[0113] S402: Match the first time period with the historical OD traffic pattern data table to determine the matched historical arrival time period.

[0114] Among them, the first time period is the next 15 minutes. Taking the prediction of traffic flow in each channel in the next 15 minutes as an example, the traffic flow from the current time to the next 15 minutes can be divided into three 5-minute segments. Each five-minute segment is predicted separately, and then the three segments are added together.

[0115] Take the prediction process of the number of vehicles entering the zone at the boundary intersection of a 5-minute segment as an example. The 5-minute segment is just within a certain arrival time period. For example, the 5-minute segment is the arrival time [10:00, 10:05]. The 5-minute segment is used as the arrival time and matched to the arrival time period of the same characteristic day in the OD rule. If there is overlap, it can be considered a match. For example, if the future 5-minute segment is [10:00, 10:05], it can match the time period of arrival time [09:58, 10:13], or it can match the time period of arrival time [09:47, 10:02], or it can match the time period of [10:03, 10:18]. These matching time periods are all examples and do not form a specific limitation.

[0116] S403: Determine the departure time period according to the matched historical arrival time period and the average arrival time.

[0117] Among them, if the actual matching result is that the matched historical arrival time period is [09:58,10:13], when the average arrival time is 13 minutes, the departure time period is [09:45,10:00].

[0118] S404: Determine the predicted number of vehicles entering a boundary intersection by applying the number of vehicles passing through the departure time period, the corresponding ratio, and the set coefficient, wherein the set coefficient is the degree of overlap between the matched historical arrival time period and the first time period.

[0119] Optionally, the above 5-minute segment arrives at [10:00,10:05], and the arrival time is [09:58,10:13], then use [09:58,10:13]-the arrival time from O to D. For example, if the arrival time from O to D is 13 minutes, then the OD regular time period passing through point O is [09:45,10:00]. At this time, the number of vehicles passing at the departure time [09:45,10:00] × the OD distribution rule (corresponding proportion) should be selected as the number of vehicles passing from O to D. However, since the number of vehicles passing through point O is counted on a 5-minute hourly basis, at this time, select a time period with the largest overlap with [09:47,09:52] as the number of vehicles passing through point O, that is, [09:45,09:50]. However, when the 5-minute period during which the vehicle passes through point O has not been detected, for example, when predicting the number of vehicles arriving at D at [10:00,10:05], the arrival time is 10 minutes, and the vehicle passing through [09:50,09:55] has not been detected, then the number of vehicles passing through point O in the last 5 minutes of this period × OD distribution rule is used.

[0120] The above five-minute segment has an intersection with a certain arrival time period. For example, the next five-minute segment is [10:00, 10:05], and the matching arrival time is [09:47, 10:02], which overlaps by 2 minutes. The arrival time minus the travel time from O to D, [09:47, 10:02] - 17 minutes = [09:30, 09:45], and the overlapped part is [09:43, 09:45]. The distribution rule is [09:40, 09:45], which is the number of vehicles passing in the five minutes × 2 / 5 (set coefficient) × OD. If the vehicles passing in the overlapped part are not detected, the most recently detected vehicle data is applied.

[0121] In the actual prediction process, if too few peripheral monitoring points are set, not all vehicles entering the core area will be detected. If too many peripheral monitoring points are set, vehicles entering the core area will be detected repeatedly. Therefore, the prediction value based on the OD law needs to be adjusted to serve as the final prediction value. The adjustment is based on the current OD traffic law.

[0122] In an embodiment of the present application, a short-term traffic flow prediction method is provided. By acquiring the traffic flow of peripheral monitoring points in real time and predicting the traffic flow at regional boundary intersections, the problem of prediction lag is largely solved. By subdividing characteristic days, time periods, vehicle locations, etc., detailed OD rules are captured, making the prediction results more accurate and sensitive.

[0123] Figure 5 A schematic diagram comparing the predicted result of the first number of vehicles on the road on a certain day and the actual result provided in an embodiment of the present application shows that the predicted value is very close to the true value, and there is no prediction lag during the morning peak, evening peak and night peak, indicating that the model can achieve ideal prediction results when the traffic flow is relatively stable in the summer.

[0124] In the embodiment of the present application, in the method of applying the first prediction model, a time-divided linear regression method is applied, which not only takes into account the main influencing factors of the number of vehicles on the road in the future, but also takes into account the different degrees of influence of these influencing factors in different time periods. The model is divided into detailed parts from the time dimension and the influencing feature dimension to make the prediction result more accurate; the traffic flow forecast value of the entry channel is used as an influencing factor of the model, which greatly reduces the prediction lag of the model.

[0125] In S203, the second prediction model is the Holt-Winters model, and the second parameter set is the number of vehicles on the way in the first historical time period before the current moment.

[0126] Among them, the number of vehicles in transit has the characteristics of most traffic flows, mainly including the following characteristics: trend refers to the long-term increase or decrease of data. The trend can be any function, such as linear or exponential, and can change direction over time; seasonality refers to the cycle repeated in the series at a fixed frequency (hours of the day, weeks, months, years, etc.), and the seasonal pattern has a fixed known cycle; random noise refers to random changes in the series.

[0127] Figure 6 A schematic diagram of the decomposition results of the number of vehicles on the road in a certain time period provided by an embodiment of the present application. Among them, the trend term is reflected in the gradual increase in the number of vehicles on the road; the seasonal term is reflected in the number of vehicles on the road repeating in a daily cycle, with obvious morning peak, evening peak and night peak every day; the random noise is reflected in the random fluctuation of the number of vehicles on the road every five minutes.

[0128] Therefore, it can be seen that the Holt-Winters seasonal method is suitable for the statistics of such time series prediction, and these features can be combined to construct the Holt-Winters model. The Holt-Winters seasonal method includes a prediction equation and three smoothing equations, with two different seasonal components. When the seasonal changes remain roughly unchanged in the time series, an additive model is usually selected; and when the seasonal changes vary proportionally to the level of the time series, a multiplicative model is usually selected. In an embodiment of the present application, an additive model can be selected.

[0129] When constructing the Holt-Winters model, the number of vehicles in transit in the area in the last five days can be obtained to perform outlier processing and data smoothing. For example, when using the Savitzky-Golay filter to smooth the data, when the area is small, the number of vehicles in transit may fluctuate greatly, which is not conducive to the implementation of various signal control measures based on the number of vehicles in transit. The Savitzky-Golay filter is a data smoothing technology that is widely used in data stream smoothing and denoising. It is based on polynomials in the time domain and uses the least squares method through a moving window for optimal fitting. Compared with other similar averaging methods, this method can better retain distribution characteristics such as relative maximum values, minimum values, and width. The third step is to use the processed data to establish the Holt-Winters model, solve the optimal parameters, and make predictions.

[0130] Figure 7 A flow chart of a method for determining the number of second vehicles in transit provided in an embodiment of the present application, Figure 5 At least the following steps are included:

[0131] S701: Obtain a second parameter set.

[0132] The second parameter set is the number of vehicles on the road in the third historical time period before the current moment. The third time period may be 15 minutes. The second parameter set is determined based on the second vehicle passing data monitored by M peripheral monitoring points and N boundary intersections.

[0133] S702: Input the number of vehicles in transit in the third time period into a pre-built Holt-Winters model to obtain the second number of vehicles in transit in the first time period.

[0134] In the embodiment of the present application, when faced with sudden changes in traffic flow, such as a huge increase in traffic volume during holidays, the accuracy of the prediction method based on time-segmented multivariate linear regression will decrease. The prediction process of the Holt-Winters model can make up for this defect.

[0135] Regarding S204, since the first prediction model and the second prediction model each have their own advantages, the first number of vehicles in transit and the second number of vehicles in transit may be selected as the target number of vehicles in transit for the first time period.

[0136] In the actual application process, both of the above prediction models perform real-time prediction and store the prediction results respectively. The absolute percentage error between the two prediction values ​​and the true value is calculated in real time to determine which prediction model's prediction result to use for the next prediction value.

[0137] Figure 8 A flow chart of a method for determining the number of target vehicles in transit provided in an embodiment of the present application, Figure 6At least the following steps are included:

[0138] S801: Compare the actual number of vehicles on the road detected at the current moment with the first historical number of vehicles on the road predicted at the first historical moment using the first prediction model to determine a first error.

[0139] For example, the current time is 6:00, and the number of vehicles on the road at 6:15 is to be predicted. The first historical time is 5:45. The actual number of vehicles on the road at 6:00 is calculated and compared with the first historical number of vehicles on the road at 6:00 predicted at 5:45 using the first prediction model to determine the first error.

[0140] S802: Compare the actual number of vehicles on the road detected at the current moment with the second historical number of vehicles on the road predicted at the second historical moment using the second prediction model to determine a second error.

[0141] For example, the current time is 6:00, and the number of vehicles on the road at 6:15 is to be predicted. The first historical time is 5:45. The actual number of vehicles on the road at 6:00 is calculated and compared with the second historical number of vehicles on the road at 6:00 predicted at 5:45 using the second prediction model to determine the second error.

[0142] S803: If the first error is smaller than the second error, the first number of vehicles in transit is selected as the target number of vehicles in transit for the first time period.

[0143] S804: If the second error is smaller than the first error, the second number of vehicles in transit is selected as the target number of vehicles in transit for the first time period.

[0144] In addition, the prediction errors of multiple historical moments may be selected and weighted to obtain a comprehensive error, and the first number of vehicles on the way and the second number of vehicles on the way are selected as the target number of vehicles on the way according to the comprehensive error. For example, in the above embodiment, the error between the actual number of vehicles on the way at 5:55 and the predicted number of vehicles on the way at 5:40 is calculated, and the error between the actual number of vehicles on the way at 5:50 and the predicted number of vehicles on the way at 5:35 is calculated, and the multiple errors are weighted.

[0145] In an embodiment of the present application, OD traffic flow patterns are calculated based on vehicle passing data at peripheral monitoring points and regional boundary intersections; future traffic flow at each boundary intersection is predicted based on real-time vehicle passing and OD traffic flow patterns at peripheral monitoring points; a regional in-transit vehicle number prediction model (first prediction model) based on time-segmented multivariate linear regression is established based on the number of in-and-out vehicles, the current number of vehicles in transit, and the number of vehicles in transit during the same historical period; a time series prediction model (second prediction model) for the number of regional in-transit vehicles is established based on the Holt-Winters method; and the optimal result is selected as the prediction result for the number of vehicles in transit at the next time granularity based on the degree of proximity between the two prediction results and the most recent true value.

[0146] It involves S205: if the target number of vehicles on the way meets the preset congestion condition, it is determined that the target area is congested in the first time period.

[0147] Among them, congestion conditions are preset, for example, congestion is determined when the number of vehicles in transit is greater than a certain value. Therefore, when the target number of vehicles in transit meets the preset congestion conditions, it can be determined that the target area is congested in the first time period.

[0148] In the embodiment of the present application, the number of vehicles in transit in a region is used as the criterion for determining regional congestion. The criterion is more intuitive and the control target is quantifiable. When alleviating regional congestion, the number of vehicles that need to be controlled outside the region can be quantified. In the process of predicting the number of vehicles in transit, a prediction method based on OD law and time series is adopted, which can combine the advantages of different prediction methods to improve the prediction accuracy. By predicting the number of vehicles in transit in a region, a strategy solution that is close to the change of congestion is automatically recommended to achieve congestion relief control before congestion occurs.

[0149] Specifically, in order to achieve accurate prediction of the number of vehicles in transit in the region, at the physical measure level, multiple peripheral monitoring points for entering the region are established outside the region to monitor the traffic entering the region, so as to achieve more accurate and more reliable prediction of the traffic flow of each key entry channel. The traffic flow of the entry channel is an important factor affecting the increase or decrease of the number of vehicles in transit in the region. The combined prediction method and the hierarchical prediction method are adopted. The combined prediction method refers to the use of the multivariate linear regression method based on time periods and the time series prediction method to jointly predict the number of vehicles in the region in the future. The regression method ensures that the prediction is not delayed under the premise of strong seasonality of vehicles in transit, and the time series method ensures that when the number of vehicles in transit fluctuates greatly, the predicted value will not deviate significantly from the current value. The combined prediction has the advantages of different single prediction methods and is a common means to improve traffic flow prediction; hierarchical prediction refers to first predicting the number of vehicles in each entry channel in the future based on the OD law. This is used as a factor of the multivariate linear regression prediction method, combined with the number of vehicles leaving the region in the last 10 minutes, the number of vehicles in transit in the current region, and the number of vehicles in transit in the region in the same period in history, to achieve the prediction of the number of vehicles in transit in the region in the next 15 minutes.

[0150] like Fig. 9 As shown, based on the same inventive concept, an embodiment of the present application provides a device for determining the congestion level of a target area, including a processing unit 91 and a prediction unit 92.

[0151] The processing unit 91 is used to: determine a target area; wherein the target area includes M peripheral monitoring points and N boundary intersections; M and N are integers respectively;

[0152] The prediction unit 92 is used to: apply a first prediction model and predict a first number of vehicles in transit in a first time period based on a first parameter set; wherein the first prediction model is a time-segmented multivariate linear regression model; the first parameter set includes at least one parameter that affects the first number of vehicles in transit; and the first parameter set is determined based on first vehicle passing data monitored by M peripheral monitoring points and N boundary intersections;

[0153] The prediction unit 92 is further used to: apply a second prediction model and predict the second number of vehicles in transit in the first time period based on a second parameter set; wherein the second prediction model is a Holt-Winters model, and the second parameter set is the number of vehicles in transit in the first historical time period before the current moment; and the second parameter set is determined based on the second vehicle passing data monitored by M peripheral monitoring points and N boundary intersections;

[0154] The processing unit 91 is further configured to: select a first number of vehicles in transit and a second number of vehicles in transit as a target number of vehicles in transit for a first time period;

[0155] The processing unit 91 is further configured to: if the target number of vehicles in transit meets a preset congestion condition, determine that the target area is congested in the first time period.

[0156] In an optional implementation, before applying the first prediction model to predict the first number of vehicles in transit in the first time period based on the first parameter set, the processing unit 91 is further configured to:

[0157] Constructing a first prediction model based on the historical parameter set;

[0158] Among them, the historical parameter set includes multiple groups of historical parameters, each group of historical parameters includes the number of vehicles entering the area within a first set time after the second historical time period, the number of vehicles leaving the area within a second set time before the second historical time period, the number of vehicles in transit in the second historical time period, and the average value of the number of vehicles in transit in each time interval with the same time attribute as the first time period, and the second historical time period belongs to a time interval.

[0159] In an optional implementation, the prediction unit 92 is specifically configured to:

[0160] Inputting the first parameter set into the first prediction model to obtain a first number of vehicles in transit;

[0161] Among them, the first parameter set includes the number of vehicles entering the area within the first time period after the current moment, the number of vehicles leaving the area within a third set time period before the current moment that is consistent with the length of the first time period, the number of vehicles in transit at the current moment, and the average value of the number of vehicles in transit in each time interval with the same time attribute in the first time period.

[0162] In an optional implementation, the prediction unit 92 is further configured to:

[0163] For each boundary intersection, the sum of the number of vehicles arriving at the boundary intersection from each peripheral monitoring point associated with the boundary intersection within the first time period is calculated as the number of vehicles entering the boundary intersection;

[0164] The sum of the number of vehicles entering the zone at each boundary intersection is determined to be the number of vehicles entering the zone in the first time period.

[0165] In an optional implementation, the prediction unit 92 is further configured to:

[0166] For each boundary intersection, determine a historical OD traffic flow regularity data table of the boundary intersection; wherein the historical OD traffic flow regularity data table includes P data items, each of which includes a characteristic day within a historical date range, and vehicle passing information from a peripheral monitoring point to a boundary intersection; the vehicle passing information includes the ratio of the number of vehicles arriving at the boundary intersection to the number of vehicles leaving the peripheral monitoring point, the departure time, the time from the peripheral monitoring point to the boundary intersection, and the average arrival time;

[0167] Match the first time period with the historical OD traffic pattern data table to determine the matched historical arrival time period;

[0168] Determine the departure time period based on the matched historical arrival time period and average arrival time;

[0169] The number of vehicles passing through the departure time period, the corresponding ratio, and the set coefficient are used to determine the predicted number of vehicles entering the area at a boundary intersection; wherein the set coefficient is the degree of overlap between the matched historical arrival time period and the first time period.

[0170] In an optional implementation, the prediction unit 92 is specifically configured to:

[0171] Obtain a second parameter set; wherein the second parameter set is the number of vehicles in transit in a third historical time period before the current moment;

[0172] The number of vehicles in transit during the third time period is input into the pre-constructed Holt-Winters model to obtain the second number of vehicles in transit during the first time period.

[0173] In an optional implementation, the processing unit 91 is specifically configured to:

[0174] Compare the actual number of vehicles on the road detected at the current moment with the first historical number of vehicles on the road predicted at the first historical moment by applying the first prediction model to determine a first error;

[0175] Comparing the actual number of vehicles on the road detected at the current moment with the second historical number of vehicles on the road predicted at the second historical moment by applying the second prediction model to determine a second error;

[0176] If the first error is less than the second error, selecting the first number of vehicles in transit as the target number of vehicles in transit for the first time period;

[0177] If the second error is smaller than the first error, the second number of vehicles in transit is selected as the target number of vehicles in transit for the first time period.

[0178] Since the device is the device in the method in the embodiment of the present application, and the principle of solving the problem by the device is similar to that of the method, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0179] like Fig.10 As shown, based on the same inventive concept, an embodiment of the present application provides an electronic device, including a processor 101 and a memory 102 .

[0180] The memory 102 is configured to execute:

[0181] Store the peripheral monitoring points and boundary intersections corresponding to each area;

[0182] The processor 101 is configured to execute:

[0183] Determine a target area; wherein the target area includes M peripheral monitoring points and N boundary intersections; M and N are integers respectively;

[0184] Applying a first prediction model, based on a first parameter set, predicting a first number of vehicles in transit in a first time period; wherein the first prediction model is a time-segmented multivariate linear regression model; the first parameter set includes at least one parameter that affects the first number of vehicles in transit; the first parameter set is determined based on first vehicle passing data monitored by M peripheral monitoring points and N boundary intersections;

[0185] Applying a second prediction model, based on a second parameter set, predicting the second number of vehicles on the way in the first time period; wherein the second prediction model is a Holt-Winters model, and the second parameter set is the number of vehicles on the way in the first historical time period before the current moment; the second parameter set is determined based on the second vehicle passing data monitored by M peripheral monitoring points and N boundary intersections;

[0186] Selecting a first number of vehicles in transit and a second number of vehicles in transit as a target number of vehicles in transit for a first time period;

[0187] If the target number of vehicles in transit meets the preset congestion condition, it is determined that the target area is congested in the first time period.

[0188] In an optional implementation, before applying the first prediction model to predict the first number of vehicles in transit in the first time period based on the first parameter set, the processor 101 is further configured to execute:

[0189] Constructing a first prediction model based on the historical parameter set;

[0190] Among them, the historical parameter set includes multiple groups of historical parameters, each group of historical parameters includes the number of vehicles entering the area within a first set time after the second historical time period, the number of vehicles leaving the area within a second set time before the second historical time period, the number of vehicles in transit in the second historical time period, and the average value of the number of vehicles in transit in each time interval with the same time attribute as the first time period.

[0191] In an optional implementation, the processor 101 is configured to execute:

[0192] Inputting the first parameter set into the first prediction model to obtain a first number of vehicles in transit;

[0193] Among them, the first parameter set includes the number of vehicles entering the area within the first time period after the current moment, the number of vehicles leaving the area within a third set time period before the current moment that is consistent with the length of the first time period, the number of vehicles in transit at the current moment, and the average value of the number of vehicles in transit in each time interval with the same time attribute in the first time period.

[0194] In an optional implementation, the processor 101 is further configured to:

[0195] For each boundary intersection, the sum of the number of vehicles arriving at the boundary intersection from each peripheral monitoring point associated with the boundary intersection within the first time period is calculated as the number of vehicles entering the boundary intersection;

[0196] The sum of the number of vehicles entering the zone at each boundary intersection is determined to be the number of vehicles entering the zone in the first time period.

[0197] In an optional implementation, the processor 101 is further configured to:

[0198] For each boundary intersection, determine a historical OD traffic flow regularity data table of the boundary intersection; wherein the historical OD traffic flow regularity data table includes P data items, each of which includes a characteristic day within a historical date range, and vehicle passing information from a peripheral monitoring point to a boundary intersection; the vehicle passing information includes the ratio of the number of vehicles arriving at the boundary intersection to the number of vehicles leaving the peripheral monitoring point, the departure time, the time from the peripheral monitoring point to the boundary intersection, and the average arrival time;

[0199] Match the first time period with the historical OD traffic pattern data table to determine the matched historical arrival time period;

[0200] Determine the departure time period based on the matched historical arrival time period and average arrival time;

[0201] The number of vehicles passing through the departure time period, the corresponding ratio, and the set coefficient are used to determine the predicted number of vehicles entering the area at a boundary intersection; wherein the set coefficient is the degree of overlap between the matched historical arrival time period and the first time period.

[0202] In an optional implementation manner, the processor 101 is specifically configured to:

[0203] Obtain a second parameter set; wherein the second parameter set is the number of vehicles in transit in a third historical time period before the current moment;

[0204] The number of vehicles in transit during the third time period is input into the pre-constructed Holt-Winters model to obtain the second number of vehicles in transit during the first time period.

[0205] In an optional implementation, the processing unit 91 is specifically configured to:

[0206] Compare the actual number of vehicles on the road detected at the current moment with the first historical number of vehicles on the road predicted at the first historical moment by applying the first prediction model to determine a first error;

[0207] Comparing the actual number of vehicles on the road detected at the current moment with the second historical number of vehicles on the road predicted at the second historical moment by applying the second prediction model to determine a second error;

[0208] If the first error is less than the second error, selecting the first number of vehicles in transit as the target number of vehicles in transit for the first time period;

[0209] If the second error is smaller than the first error, the second number of vehicles in transit is selected as the target number of vehicles in transit for the first time period.

[0210] The embodiment of the present application further provides a computer storage medium, in which computer program instructions are stored. When the instructions are executed on a computer, the computer executes the steps of the method for determining the congestion level of a target area.

[0211] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0212] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0213] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0214] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0215] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A method for determining the congestion level of a target area, characterized in that: include: Determine a target area; wherein the target area includes M peripheral monitoring points and N boundary intersections; M and N are integers respectively; Applying a first prediction model, based on a first parameter set, predicting a first number of vehicles in transit in a first time period; wherein the first prediction model is a time-segmented multivariate linear regression model; the first parameter set includes at least one parameter that affects the first number of vehicles in transit; the first parameter set is determined based on first vehicle passing data monitored by the M peripheral monitoring points and the N boundary intersections; Applying a second prediction model, based on a second parameter set, predicting the second number of vehicles on the way in the first time period; wherein the second prediction model is a Holt-Winters model, and the second parameter set is the number of vehicles on the way in the first historical time period before the current moment; the second parameter set is determined based on the second vehicle passing data monitored by the M peripheral monitoring points and the N boundary intersections; selecting the first number of vehicles in transit and the second number of vehicles in transit as the target number of vehicles in transit during the first time period; If the target number of vehicles in transit meets a preset congestion condition, it is determined that the target area is congested in the first time period.

2. The method according to claim 1, characterized in that Before applying the first prediction model to predict the first number of vehicles in transit in the first time period based on the first parameter set, the method further includes: Constructing the first prediction model according to the historical parameter set; Among them, the historical parameter set includes multiple groups of historical parameters, each group of historical parameters includes the number of vehicles entering the area within a first set time after the second historical time period, the number of vehicles leaving the area within a second set time before the second historical time period, the number of vehicles in transit in the second historical time period, and the average value of the number of vehicles in transit in each time interval with the same time attribute as the first time period, and the second historical time period belongs to a time interval.

3. The method according to claim 2, characterized in that The applying the first prediction model to predict the first number of vehicles in transit in the first time period based on the first parameter set includes: Inputting the first parameter set into the first prediction model to obtain the first number of vehicles in transit; Among them, the first parameter set includes the number of vehicles entering the area within the first time period after the current moment, the number of vehicles leaving the area within a third set time length before the current moment that is consistent with the length of the first time period, the number of vehicles in transit at the current moment, and the average value of the number of vehicles in transit in each time interval with the same time attribute of the first time period.

4. The method according to claim 3, characterized in that The method further comprises: For each boundary intersection, the sum of the number of vehicles arriving at the boundary intersection from each peripheral monitoring point associated with the boundary intersection within the first time period is calculated as the number of vehicles entering the boundary intersection; The sum of the number of vehicles entering the zone at each boundary intersection is determined to be the number of vehicles entering the zone in the first time period.

5. The method according to claim 4, characterized in that The method further comprises: For each boundary intersection, determine a historical OD traffic flow regularity data table of the boundary intersection; wherein the historical OD traffic flow regularity data table includes P data, each data includes a characteristic day within a historical date range, and vehicle passing information from a peripheral monitoring point to a boundary intersection; the vehicle passing information includes the ratio of the number of vehicles arriving at the boundary intersection to the number of vehicles leaving the peripheral monitoring point, the departure time, the time from the peripheral monitoring point to the boundary intersection, and the average arrival time; Matching the first time period with the historical OD traffic flow pattern data table to determine a matched historical arrival time period; Determine a departure time period according to the matched historical arrival time period and the average arrival time; The number of vehicles passing through the departure time period, the corresponding ratio, and the set coefficient are used to determine the predicted number of vehicles entering the zone at the boundary intersection; wherein the set coefficient is the degree of overlap between the matched historical arrival time period and the first time period.

6. The method according to claim 1, characterized in that: The applying the second prediction model to predict the second number of vehicles in transit during the first time period based on the second parameter set includes: Obtain a second parameter set; wherein the second parameter set is the number of vehicles in transit in a third historical time period before the current moment; The number of vehicles in transit during the third time period is input into a pre-constructed Holt-Winters model to obtain a second number of vehicles in transit during the first time period.

7. The method according to any one of claims 1 to 6, characterized in that: The first number of vehicles in transit and the second number of vehicles in transit are selected as the target number of vehicles in transit for the first time period, including: Comparing the actual number of vehicles on the road detected at the current moment with the first historical number of vehicles on the road predicted at the first historical moment by applying the first prediction model to determine a first error; Comparing the actual number of vehicles on the road detected at the current moment with the second historical number of vehicles on the road predicted at the second historical moment by applying the second prediction model to determine a second error; If the first error is less than the second error, selecting the first number of vehicles in transit as the target number of vehicles in transit for the first time period; If the second error is smaller than the first error, the second number of vehicles in transit is selected as the target number of vehicles in transit for the first time period.

8. An electronic device, characterized in that: including a processor and a memory; The memory is configured to execute: Store the peripheral monitoring points and boundary intersections corresponding to each area; The processor is configured to execute: Determine a target area; wherein the target area includes M peripheral monitoring points and N boundary intersections; M and N are integers respectively; Applying a first prediction model, based on a first parameter set, predicting a first number of vehicles in transit in a first time period; wherein the first prediction model is a time-segmented multivariate linear regression model; the first parameter set includes at least one parameter that affects the first number of vehicles in transit; the first parameter set is determined based on first vehicle passing data monitored by the M peripheral monitoring points and the N boundary intersections; Applying a second prediction model, based on a second parameter set, predicting the second number of vehicles on the way in the first time period; wherein the second prediction model is a Holt-Winters model, and the second parameter set is the number of vehicles on the way in the first historical time period before the current moment; the second parameter set is determined based on the second vehicle passing data monitored by the M peripheral monitoring points and the N boundary intersections; selecting the first number of vehicles in transit and the second number of vehicles in transit as the target number of vehicles in transit during the first time period; If the target number of vehicles in transit meets a preset congestion condition, it is determined that the target area is congested in the first time period.

9. The electronic device according to claim 8, characterized in that: Before applying the first prediction model to predict the first number of vehicles in transit in the first time period based on the first parameter set, the processor is further configured to execute: Constructing the first prediction model according to the historical parameter set; Among them, the historical parameter set includes multiple groups of historical parameters, each group of historical parameters includes the number of vehicles entering the area within a first set time after a second historical time period, the number of vehicles leaving the area within a second set time before the second historical time period, the number of vehicles in transit in the second historical time period, and the average value of the number of vehicles in transit in each time interval with the same time attribute as the first time period.

10. The electronic device according to claim 9, characterized in that: The processor is configured to execute: Inputting the first parameter set into the first prediction model to obtain the first number of vehicles in transit; Among them, the first parameter set includes the number of vehicles entering the area within the first time period after the current moment, the number of vehicles leaving the area within a third set time length before the current moment that is consistent with the length of the first time period, the number of vehicles in transit at the current moment, and the average value of the number of vehicles in transit in each time interval with the same time attribute of the first time period.

Images