Road congestion prediction method

Abstract

The invention discloses a road congestion prediction method, relates to the technical field of urban road traffic, and solves the technical problem of low accuracy of road congestion prediction in the prior art. According to the invention, a target road is divided into a plurality of road units based on intersection positions; obtaining road basic information and historical vehicle passing information of each road unit; calculating a road congestion coefficient of the target road based on the road basic information and the historical vehicle traffic data; obtaining a road congestion coefficient prediction value of the target road in the prediction period; and comparing the predicted value of the road congestion coefficient with a preset threshold value of a road congestion range to obtain the congestion degree. According to the method, the road congestion coefficient of each road unit is calculated by performing formulation processing on the congestion point density, the traffic light period and the historical vehicle passing data, so that the influence of the road congestion coefficient on the congestion degree of the road unit can be accurately reflected in a multi-dimensional manner, and the accuracy of predicting the road congestion can be improved.

Description

Technical Field

[0001] The invention belongs to the technical field of urban road traffic, and in particular is a road congestion prediction method. Background Art

[0002] With the continuous improvement of living standards, the number of vehicles has increased rapidly, and the contradiction between urban transportation infrastructure and people's transportation needs has become increasingly prominent. Therefore, urban traffic congestion prediction has become increasingly important.

[0003] The existing technology calculates the road congestion value through parameters such as road width, traffic volume and vehicle speed, and predicts the road congestion situation based on the road congestion value; however, the number of congestion points on different roads is not the same, and the existing technology is difficult to predict the road congestion based on the number of congestion points on the road, resulting in low accuracy in the prediction of road congestion; in addition, there are differences in the degree of congestion of different road units on the road, and the existing technology is difficult to comprehensively evaluate the congestion situation of the road based on the congestion degree of different road units in the road, resulting in low accuracy in the prediction of road congestion.

[0004] The present invention proposes a road congestion prediction method to solve the above technical problems. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a road congestion prediction method to solve the problem that the prior art is difficult to predict road congestion based on the number of congestion points on the road, resulting in low accuracy in road congestion prediction; in addition, there are differences in the degree of congestion of different road units on the road, and the prior art is difficult to comprehensively evaluate the congestion status of the road based on the congestion degree of different road units in the road.

[0006] To achieve the above object, a first aspect of the present invention provides a road congestion prediction method, comprising:

[0007] S1: Divide the target road into several road units based on the intersection location;

[0008] S2: Obtain basic road information and historical vehicle traffic information of each road unit; the basic information includes road width, road length, number of congestion points and traffic light cycle; the historical vehicle traffic information includes vehicle travel time, vehicle travel speed and number of waiting vehicles;

[0009] S3: pre-processing the historical vehicle traffic information to obtain historical vehicle traffic data;

[0010] S4: Calculate the road congestion coefficient of the target road based on the basic road information and historical vehicle traffic data;

[0011] S5: Obtaining a predicted value of a road congestion coefficient of a target road during a prediction period;

[0012] S6: Compare the predicted road congestion coefficient value with a preset road congestion range threshold to obtain a congestion degree; wherein the congestion degree includes no congestion, mild congestion and severe congestion.

[0013] It should be noted that the vehicle travel time, vehicle travel speed and number of vehicles waiting for each road unit are the average values ​​of the corresponding historical vehicle travel data; among them, the vehicle travel time is the average time required for all vehicles to travel from one side of the road unit to the other side.

[0014] Preferably, dividing the target road into a plurality of road units based on the intersection position includes:

[0015] Obtain the intersection position of the target road; divide the road between two adjacent intersection positions in the target road into road units, and mark the road unit as i; wherein i=1, 2, ..., n, and n is the total number of road units.

[0016] Preferably, the congestion point is a location in the road unit where the number of congestion occurrences within a set time is greater than a set threshold.

[0017] Preferably, the preprocessing of historical vehicle traffic information includes:

[0018] A1: Extract historical vehicle traffic information for several consecutive periods;

[0019] A2: Eliminate abnormal data from historical vehicle traffic information;

[0020] A3: Fill in the missing data in historical vehicle traffic information;

[0021] A4: Mark the pre-processed historical vehicle traffic information as historical vehicle traffic data.

[0022] Preferably, the method of calculating the road congestion coefficient of the target road based on the basic road information and historical vehicle traffic data includes:

[0023] B1: Obtain basic road information and historical vehicle traffic data for each road unit;

[0024] B2: Calculate the congestion point density of each road unit based on road width, road length and number of congestion points;

[0025] B3: Calculate the road congestion coefficient of each road unit based on the density of congestion points, traffic light cycle and historical vehicle traffic data;

[0026] B4: By formula Calculate the road congestion coefficient DYX of the target road; where ki is the weight coefficient corresponding to the road unit i and DYXi is the road congestion coefficient corresponding to road unit i.

[0027] The present invention calculates the road congestion coefficients of different road units in the target road separately, and then obtains the road congestion coefficient of the target road by weighted average calculation of the road congestion coefficients of each road unit. This can comprehensively evaluate the impact of each road unit on the congestion of the target road, which is beneficial to improving the accuracy of predicting road congestion.

[0028] Preferably, the calculation of the congestion point density of each road unit based on the road width, the road length and the number of congestion points includes:

[0029] Extract road width, road length and number of congestion points; through the formula Calculate the congestion point density YDMi of road unit i; where YDSi is the number of congestion points of road unit i, DKi is the road width of road unit i, and DCi is the road length of road unit i.

[0030] The present invention calculates the density of congestion points of each road unit by substituting the road width, road length and number of congestion points into a formula, which can accurately reflect the probability of congestion occurring in each road unit and is conducive to improving the accuracy of subsequent calculation of the road congestion coefficient of the road unit.

[0031] Preferably, the method of calculating the road congestion coefficient of each road unit based on the density of congestion points, traffic light cycle and historical vehicle traffic data includes:

[0032] Extract basic road information and historical vehicle traffic data of each road unit;

[0033] By formula Calculate the road congestion coefficient DYXi of road unit i;

[0034] Wherein, HLZi is the traffic light cycle of road unit i, HBZi is the standard traffic light cycle of road unit i, SCi is the vehicle travel time of road unit i, SDi is the vehicle travel speed of road unit i, SLi is the number of waiting vehicles of road unit i; ai, bi, ci, di, ei are the proportional coefficients corresponding to road unit i and are all greater than 0; h is a constant set to prevent division by zero, and 0 <h<1。

[0035] The present invention calculates the road congestion coefficient of each road unit by formulating the congestion point density, traffic light cycle, and historical vehicle passing data, enabling the road congestion coefficient to accurately and multi-dimensionally reflect the influence of the congestion point density, traffic light cycle, and historical vehicle passing data on the congestion degree of the road unit, which is beneficial to improving the accuracy of predicting road congestion.

[0036] Preferably, obtaining the predicted value of the road congestion coefficient of the target road during the prediction period includes:

[0037] Extracting the road congestion coefficients of several consecutive recent periods of the target road and marking them as the standard data set, and inputting the standard data set into the road congestion prediction model to obtain the predicted value of the road congestion coefficient of the target road during the prediction period; wherein, the road congestion prediction model is obtained by training an artificial intelligence model.

[0038] Preferably, the road congestion prediction model is obtained by training an artificial intelligence model, including:

[0039] Extracting the road congestion coefficients of several consecutive periods and integrating them into several groups of training data and test data; training the artificial intelligence model with the training data; testing the trained artificial intelligence model with the test data, and adjusting the artificial intelligence model according to the test results; finally obtaining a road congestion prediction model with the input being the road congestion coefficients of several consecutive recent periods and the output being the road congestion coefficient of the prediction period; wherein, the artificial intelligence model includes a BP neural network model or an RBF neural network model.

[0040] The present invention extracts the road congestion coefficient for training the artificial intelligence model, and after training, obtains the road congestion prediction model. By inputting the road congestion coefficients of several consecutive periods into the road congestion prediction model, the road congestion coefficient of the prediction period is obtained, which is beneficial to predicting the road congestion degree in advance and taking corresponding measures, facilitating people to adjust the travel plan in advance, and is beneficial to improving the vehicle passing efficiency and reducing the probability of congestion occurrence.

[0041] Preferably, comparing the predicted value of the road congestion coefficient with a preset road congestion range threshold includes:

[0042] Z1: Extract the predicted value DYX of the road congestion coefficient 预测 ;

[0043] Z2: Set the road congestion range threshold [YDY1, YDY2]; wherein, YDY1 < YDY2, and the road congestion range threshold is set according to the experience of experts;

[0044] Z3: Judge the predicted value DYX of the road congestion coefficient 预测Is it within the road congestion range threshold? If yes, the congestion level is marked as light congestion; if no, jump to Z4;

[0045] Z4: Determine the predicted value of road congestion coefficient DYX 预测 Is it less than YDY1? If yes, the congestion level is marked as no congestion; if no, the congestion level is marked as severe congestion.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] 1. The present invention calculates the road congestion coefficients of different road units in the target road separately, and then calculates the road congestion coefficient of each road unit by weighted average to obtain the road congestion coefficient of the target road. This can comprehensively evaluate the impact of each road unit on the congestion of the target road, which is beneficial to improving the accuracy of predicting road congestion.

[0048] 2. The present invention calculates the density of congestion points of each road unit by substituting the road width, road length and number of congestion points into a formula, which can accurately reflect the probability of congestion occurring in each road unit, and is conducive to improving the accuracy of subsequent calculation of the road congestion coefficient of the road unit.

[0049] 3. The present invention formulates the density of congestion points, traffic light cycles and historical vehicle traffic data to calculate the road congestion coefficient of each road unit, so that the road congestion coefficient can accurately and multi-dimensionally reflect the impact of the density of congestion points, traffic light cycles and historical vehicle traffic data on the congestion level of the road unit, which is conducive to improving the accuracy of predicting road congestion.

[0050] 4. The present invention extracts the road congestion coefficient for training the artificial intelligence model, and obtains the road congestion prediction model after the training is completed. The road congestion coefficient of several consecutive periods is input into the road congestion prediction model to obtain the road congestion coefficient of the prediction period, which is conducive to estimating the degree of road congestion in advance and taking corresponding measures, making it convenient for people to adjust their travel plans in advance, and is conducive to improving vehicle traffic efficiency and reducing the probability of congestion. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0052] Figure 1 is an overall flow chart of the road congestion prediction method of the present invention;

[0053] Figure 2 A flow chart of calculating the road congestion coefficient of a target road in the present invention;

[0054] Figure 3 The present invention is a flow chart for comparing a predicted road congestion coefficient value with a preset road congestion range threshold. DETAILED DESCRIPTION

[0055] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] See also Figure 1-Figure 3 The first embodiment of the present invention provides a road congestion prediction method, comprising:

[0057] S1: Divide the target road into several road units based on the intersection location;

[0058] S2: Obtain basic road information and historical vehicle traffic information of each road unit; the basic information includes road width, road length, number of congestion points and traffic light cycle; the historical vehicle traffic information includes vehicle travel time, vehicle travel speed and number of waiting vehicles;

[0059] S3: pre-processing the historical vehicle traffic information to obtain historical vehicle traffic data;

[0060] S4: Calculate the road congestion coefficient of the target road based on the basic road information and historical vehicle traffic data;

[0061] S5: Obtaining a predicted value of a road congestion coefficient of a target road during a prediction period;

[0062] S6: Compare the predicted road congestion coefficient value with a preset road congestion range threshold to obtain a congestion degree; wherein the congestion degree includes no congestion, mild congestion and severe congestion.

[0063] It should be noted that the vehicle travel time, vehicle travel speed and number of vehicles waiting for each road unit are the average values ​​of the corresponding historical vehicle travel data; among them, the vehicle travel time is the average time required for all vehicles to travel from one side of the road unit to the other side.

[0064] In this embodiment, the target road is divided into several road units based on the intersection location, including:

[0065] Obtain the intersection position of the target road; divide the road between two adjacent intersection positions in the target road into road units, and mark the road unit as i; wherein i=1, 2, ..., n, and n is the total number of road units.

[0066] Exemplarily, the intersection positions of the target roads are set to LK1, LK2, LK3, and LK4 respectively; the road between LK1 and LK2 is marked as road unit 1, the road between LK2 and LK3 is marked as road unit 2, and the road between LK3 and LK4 is marked as road unit 3.

[0067] In this embodiment, the congestion point is a location in the road unit where the number of congestion occurrences within a set time is greater than a set threshold.

[0068] Exemplarily, locations in the road unit where congestion occurs more than 5 times per month are marked as congestion points.

[0069] In this embodiment, the historical vehicle traffic information is preprocessed, including:

[0070] A1: Extract historical vehicle traffic information for several consecutive periods;

[0071] A2: Eliminate abnormal data from historical vehicle traffic information;

[0072] A3: Fill in the missing data in historical vehicle traffic information;

[0073] A4: Mark the pre-processed historical vehicle traffic information as historical vehicle traffic data.

[0074] Exemplarily, abnormal data in historical vehicle traffic information include data that are obviously inconsistent with common sense, such as vehicle traffic time less than 2s, vehicle traffic speed and number of waiting vehicles, etc.; missing data in historical vehicle traffic information is filled by the median of the corresponding historical vehicle traffic information.

[0075] In this embodiment, the road congestion coefficient of the target road is calculated based on the basic road information and historical vehicle traffic data, including:

[0076] B1: Obtain basic road information and historical vehicle traffic data for each road unit;

[0077] B2: Calculate the congestion point density of each road unit based on road width, road length and number of congestion points;

[0078] B3: Calculate the road congestion coefficient of each road unit based on the density of congestion points, traffic light cycle and historical vehicle traffic data;

[0079] B4: By formula Calculate the road congestion coefficient DYX of the target road; where ki is the weight coefficient corresponding to the road unit i and DYXi is the road congestion coefficient corresponding to road unit i.

[0080] Exemplarily, the weight coefficient k1 corresponding to road unit 1 is set to 0.3, the weight coefficient k2 corresponding to road unit 2 is set to 0.4, and the weight coefficient k3 corresponding to road unit 3 is set to 0.3; the road congestion coefficient DYX1 corresponding to road unit 1 is set to 3.25, the road congestion coefficient DYX2 corresponding to road unit 2 is set to 3.52, and the road congestion coefficient DYX3 corresponding to road unit 3 is set to 2.84; the road congestion coefficient DYX=3.235 of the target road is calculated by the formula.

[0081] The present invention calculates the road congestion coefficients of different road units in the target road separately, and then obtains the road congestion coefficient of the target road by weighted average calculation of the road congestion coefficients of each road unit. This can comprehensively evaluate the impact of each road unit on the congestion of the target road, which is beneficial to improving the accuracy of predicting road congestion.

[0082] In this embodiment, the congestion point density of each road unit is calculated based on the road width, road length and number of congestion points, including:

[0083] Extract road width, road length and number of congestion points; through the formula Calculate the congestion point density YDMi of road unit i; where YDSi is the number of congestion points of road unit i, DKi is the road width of road unit i, and DCi is the road length of road unit i.

[0084] Exemplarily, the number of congestion points YDS1 of road unit 1 is set to 2, the road width DK1 of road unit 1 is set to 0.032 km, and the road length DC1 of road unit 1 is set to 0.2 km; the congestion point density YDM1 of road unit 1 is calculated to be 312.5 through the formula.

[0085] The present invention calculates the density of congestion points of each road unit by substituting the road width, road length and number of congestion points into a formula, which can accurately reflect the probability of congestion occurring in each road unit and is conducive to improving the accuracy of subsequent calculation of the road congestion coefficient of the road unit.

[0086] In this embodiment, the road congestion coefficient of each road unit is calculated based on the density of congestion points, traffic light cycle and historical vehicle traffic data, including:

[0087] Extract basic road information and historical vehicle traffic data of each road unit;

[0088] By formula Calculate the road congestion coefficient DYXi of road unit i;

[0089] Wherein, HLZi is the traffic light cycle of road unit i, HBZi is the standard traffic light cycle of road unit i, SCi is the vehicle travel time of road unit i, SDi is the vehicle travel speed of road unit i, SLi is the number of waiting vehicles of road unit i; ai, bi, ci, di, ei are the proportional coefficients corresponding to road unit i and are all greater than 0; h is a constant set to prevent division by zero, and 0 <h<1。

[0090] Exemplarily, set a1=0.1, b1=3, c1=2, d1=3, e1=1.5, h=0.01; the congestion point density of road unit 1 YDM1=312.5, the traffic light cycle HLZi=80s of road unit 1, the standard traffic light cycle HBZi=70 of road unit 1, the vehicle travel time SC1=20s of road unit 1, the vehicle travel speed SD1=30km / h of road unit 1, and the number of waiting vehicles SL1=15 of road unit 1; the road congestion coefficient DYX1≈3.25 of road unit 1 is calculated by the formula.

[0091] The present invention calculates the road congestion coefficient of each road unit by formulating the density of congestion points, traffic light cycles and historical vehicle traffic data, so that the road congestion coefficient can accurately and multi-dimensionally reflect the influence of the density of congestion points, traffic light cycles and historical vehicle traffic data on the congestion degree of the road unit, which is conducive to improving the accuracy of predicting road congestion.

[0092] In this embodiment, obtaining the predicted value of the road congestion coefficient of the target road during the prediction period includes:

[0093] The road congestion coefficients of the target road in the most recent several consecutive periods are extracted and marked as a standard data set, and the standard data set is input into the road congestion prediction model to obtain the predicted value of the road congestion coefficient of the target road in the prediction period; wherein, the road congestion prediction model is obtained through artificial intelligence model training.

[0094] In this embodiment, the road congestion prediction model is obtained by training an artificial intelligence model, including:

[0095] Extract the road congestion coefficients for several consecutive periods and integrate them into several groups of original data. Use 80% of the original data as training data and 20% as test data. Train an artificial intelligence model using the training data. Test the trained artificial intelligence model using the test data and adjust the artificial intelligence model according to the test results. Finally, obtain a road congestion prediction model with the road congestion coefficients for the most recent several consecutive periods as the input and the road congestion coefficient for the prediction period as the output. Among them, the artificial intelligence model includes a BP neural network model or an RBF neural network model.

[0096] In the present invention, by extracting the road congestion coefficients for training the artificial intelligence model, after training, a road congestion prediction model is obtained. By inputting the road congestion coefficients for several consecutive periods into the road congestion prediction model, the road congestion coefficient for the prediction period is obtained, which is beneficial to estimating the road congestion degree in advance and taking corresponding measures, facilitating people to adjust their travel plans in advance, and is beneficial to improving the vehicle passing efficiency and reducing the probability of congestion occurrence.

[0097] In this embodiment, comparing the predicted value of the road congestion coefficient with a preset road congestion range threshold includes:

[0098] Z1: Extract the predicted value DYX of the road congestion coefficient 预测 ;

[0099] Z2: Set the road congestion range threshold [YDY1, YDY2]; where YDY1 < YDY2, and the road congestion range threshold is set according to the experience of experts;

[0100] Z3: Determine whether the predicted value DYX of the road congestion coefficient 预测 is within the road congestion range threshold; if yes, mark the congestion degree as mild congestion; if not, jump to Z4;

[0101] Z4: Determine whether the predicted value DYX of the road congestion coefficient 预测 is less than YDY1; if yes, mark the congestion degree as no congestion; if not, mark the congestion degree as severe congestion.

[0102] Exemplarily, set the predicted value DYX prediction of the road congestion coefficient to 3.34, and the road congestion range threshold [YDY1, YDY2] = [3.1, 3.4]; since the predicted value DYX prediction of the road congestion coefficient is within the road congestion range threshold [YDY1, YDY2], the congestion degree is marked as mild congestion.

[0103] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0104] Working principle of the present invention:

[0105] The present invention divides a target road into a plurality of road units based on the intersection position; obtains basic road information and historical vehicle traffic information of each road unit; pre-processes the historical vehicle traffic information to obtain historical vehicle traffic data; calculates a road congestion coefficient of the target road based on the basic road information and the historical vehicle traffic data; obtains a predicted value of the road congestion coefficient of the target road during a prediction period; and obtains a congestion degree based on a comparison between the predicted value of the road congestion coefficient and a preset road congestion range threshold.

[0106] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A road congestion prediction method, characterized in that: include: S1: Divide the target road into several road units based on the intersection location; S2: Obtain basic road information and historical vehicle traffic information of each road unit; the basic information includes road width, road length, number of congestion points and traffic light cycle; the historical vehicle traffic information includes vehicle travel time, vehicle travel speed and number of waiting vehicles; S3: pre-processing the historical vehicle traffic information to obtain historical vehicle traffic data; S4: Calculate the road congestion coefficient of the target road based on the basic road information and historical vehicle traffic data; S5: Obtaining a predicted value of a road congestion coefficient of a target road during a prediction period; S6: Compare the predicted road congestion coefficient value with a preset road congestion range threshold to obtain a congestion degree; wherein the congestion degree includes no congestion, mild congestion and severe congestion.

2. A road congestion prediction method according to claim 1, characterized in that: The target road is divided into a plurality of road units based on the intersection position, including: Obtain the intersection position of the target road; divide the road between two adjacent intersection positions in the target road into road units, and mark the road unit as i; wherein i=1, 2, ..., n, and n is the total number of road units.

3. A road congestion prediction method according to claim 1, characterized in that: The congestion point is a location in a road unit where the number of congestion occurrences within a set time is greater than a set threshold.

4. A road congestion prediction method according to claim 1, characterized in that: The preprocessing of historical vehicle traffic information includes: A1: Extract historical vehicle traffic information for several consecutive periods; A2: Eliminate abnormal data from historical vehicle traffic information; A3: Fill in the missing data in historical vehicle traffic information; A4: Mark the pre-processed historical vehicle traffic information as historical vehicle traffic data.

5. A road congestion prediction method according to claim 1, characterized in that: The method of calculating the road congestion coefficient of the target road based on the basic road information and historical vehicle traffic data includes: B1: Obtain basic road information and historical vehicle traffic data for each road unit; B2: Calculate the congestion point density of each road unit based on road width, road length and number of congestion points; B3: Calculate the road congestion coefficient of each road unit based on the density of congestion points, traffic light cycle and historical vehicle traffic data; B4: By formula Calculate the road congestion coefficient DYX of the target road; where ki is the weight coefficient corresponding to the road unit i and DYXi is the road congestion coefficient corresponding to road unit i.

6. A road congestion prediction method according to claim 5, characterized in that: The method of calculating the congestion point density of each road unit based on the road width, the road length and the number of congestion points includes: Extract road width, road length and number of congestion points; through the formula Calculate the congestion point density YDMi of road unit i; where YDSi is the number of congestion points of road unit i, DKi is the road width of road unit i, and DCi is the road length of road unit i.

7. A road congestion prediction method according to claim 5, characterized in that: The road congestion coefficient of each road unit is calculated based on the density of congestion points, traffic light cycle and historical vehicle traffic data, including: Extract basic road information and historical vehicle traffic data of each road unit; By formula Calculate the road congestion coefficient DYXi of road unit i; Wherein, HLZi is the traffic light cycle of road unit i, HBZi is the standard traffic light cycle of road unit i, SCi is the vehicle travel time of road unit i, SDi is the vehicle travel speed of road unit i, SLi is the number of waiting vehicles of road unit i; ai, bi, ci, di, ei are the proportional coefficients corresponding to road unit i and are all greater than 0; h is a constant set to prevent division by zero, and 0 <h<1。 8. A road congestion prediction method according to claim 1, characterized in that: The step of obtaining a predicted value of a road congestion coefficient of a target road during a prediction period includes: The road congestion coefficients of the target road in the most recent several consecutive periods are extracted and marked as a standard data set, and the standard data set is input into the road congestion prediction model to obtain the predicted value of the road congestion coefficient of the target road in the prediction period; wherein, the road congestion prediction model is obtained through artificial intelligence model training.

9. A road congestion prediction method according to claim 8, characterized in that: The road congestion prediction model is obtained through artificial intelligence model training, including: Extract road congestion coefficients for several consecutive periods and integrate them into several groups of training data and test data; use the training data to train the artificial intelligence model; use the test data to test the trained artificial intelligence model, and adjust the artificial intelligence model according to the test results; finally obtain a road congestion prediction model whose input is the road congestion coefficients of the most recent several consecutive periods and whose output is the road congestion coefficient of the prediction period; wherein the artificial intelligence model includes a BP neural network model or an RBF neural network model.

10. A road congestion prediction method according to claim 1, characterized in that: The comparing the predicted value of the road congestion coefficient with a preset road congestion range threshold value includes: Z1: Extract the predicted value of road congestion coefficient DYX 预测 ; Z2: Set the road congestion range threshold [YDY1, YDY2]; where YDY1 <YDY2; Z3: Determine the predicted value of road congestion coefficient DYX 预测 Is it within the road congestion range threshold? If yes, the congestion level is marked as light congestion; if no, jump to Z4; Z4: Determine the predicted value of road congestion coefficient DYX 预测 Is it less than YDY1? If yes, the congestion level is marked as no congestion; if no, the congestion level is marked as severe congestion.

Images