A traffic demand prediction method for a new gas station project

Abstract

The application discloses a new gas station project traffic demand prediction method, comprising the following steps: obtaining the initial selected gas stations in the target area according to the traffic location conditions, calculating the Mahalanobis distance between the initial selected gas stations and between the initial selected gas stations and the new gas station based on the multi-dimensional characteristic data of the initial selected gas stations, and screening N analog gas stations according to the Mahalanobis distance; obtaining the annual average daily sales of each oil product, the peak hour sales of each oil product and the peak hour traffic generation of each oil product of each vehicle type of the N analog gas stations, calculating the peak hour sales coefficient of each oil product and the peak hour traffic generation rate of each oil product of each vehicle type of the analog gas stations, and predicting the peak hour traffic generation of each vehicle type of the new gas station based on the peak hour sales coefficient of each oil product and the peak hour traffic generation rate of each oil product of each vehicle type of the analog gas stations. The application solves the technical problem of how to accurately predict the traffic demand of the new gas station project.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent transportation planning, and particularly relates to a traffic demand prediction method for a new gas station project. BACKGROUND

[0002] As an important part of urban transportation infrastructure, the planning layout and traffic organization of urban gas stations directly affect the efficiency and safety of regional traffic operation. With the acceleration of urbanization and the continuous growth of motor vehicle ownership, gas stations as key nodes for vehicle energy supply have a significant impact on the traffic capacity, traffic order and traffic safety of the surrounding road network. If not properly planned, gas stations may become a cause of traffic congestion and even cause traffic accidents, reducing the overall service level of the urban transportation system. During the site selection and detailed planning of a gas station construction project, conducting a scientific and systematic traffic impact assessment is essential to ensure its coordinated development with the urban transportation system. According to the Technical Standards for Traffic Impact Assessment of Construction Projects, traffic demand prediction is one of the core contents of traffic impact assessment, which quantitatively analyzes the impact of the newly added traffic volume on the surrounding road network after the project is completed, providing data support for planning decisions. However, in the existing technology, traffic demand prediction for gas stations relies heavily on traditional models and does not fully consider the differences in road network structure, land use, parking spaces, and traffic flow around the gas station, resulting in a large deviation between the predicted traffic demand and the actual situation. This deviation directly affects the reliability of the TIA conclusion, which in turn misleads the planning decision, making the gas station itself or the surrounding road network a new traffic bottleneck or safety hazard source. Therefore, there is an urgent need for a traffic demand prediction method for new gas station projects to address the technical problem of accurately predicting traffic demand for new gas station projects. SUMMARY

[0003] The main purpose of the present application is to provide a traffic demand prediction method for a new gas station project, aiming to solve the technical problem of accurately predicting traffic demand for new gas station projects.

[0004] To achieve the above-mentioned purpose, the present application provides a traffic demand prediction method for a new gas station project, wherein the traffic demand prediction method for the new gas station project comprises the following steps:

[0005] S1, obtaining initial selected gas stations in a target area according to traffic location conditions, calculating Mahalanobis distances between the initial selected gas stations and between the initial selected gas stations and the new gas station based on multi-dimensional feature data of each initial selected gas station, and selecting N analogous gas stations according to the Mahalanobis distances;

[0006] S2, obtaining the annual average daily sales of each oil product, the peak hour sales of each oil product, and the peak hour traffic generation of each oil product of N analog gas stations, calculating the peak hour sales coefficient of each oil product and the peak hour traffic generation rate of each oil product of the analog gas stations, and predicting the peak hour traffic generation of each vehicle type of the new gas station based on the peak hour sales coefficient of each oil product and the peak hour traffic generation rate of each oil product of the analog gas stations.

[0007] In one preferred embodiment, step S1 obtains the initial selected gas stations in the target area according to the traffic site conditions, specifically:

[0008] In one preferred embodiment, step S1 obtains the initial selected gas stations in the target area according to the traffic site conditions, specifically: In one preferred embodiment, step S1 obtains the initial selected gas stations in the target area according to the traffic site conditions, specifically:

[0009] In one preferred embodiment, the first feature data set includes the operating land area of the gas station, the number of oil guns, the annual sales, the number of motor vehicle parking spaces within the service range of the gas station, and the road network traffic within the service range.

[0010] In one preferred embodiment, the second feature data set includes the annual average daily sales of each oil product, the peak hour sales of each oil product, and the peak hour traffic generation of each oil product of the gas station.

[0011] In one preferred embodiment, step S1 calculates the Mahalanobis distance between the initial selected gas stations and between the initial selected gas stations and the new gas station, specifically:

[0012] In one preferred embodiment, step S1 calculates the Mahalanobis distance between the initial selected gas stations and between the initial selected gas stations and the new gas station, specifically:

[0013] In one preferred embodiment, the feature covariance matrix is:

[0014]

[0015] wherein, * the feature covariance matrix of is the covariance of the i-th feature and the j-th feature of the initial selected gas station;

[0016] the covariance of the i-th feature and the j-th feature of the initial selected gas station is: ​​​​​

[0017]

[0018] wherein, N is the total number of the initial selected gas stations, is the Mahalanobis distance between the th feature observation of the th initial selected gas station, is the mean of the th feature observation of the initial selected gas station, is the mean of the th feature observation of the initial selected gas station.

[0019] In one of the preferred embodiments, the Mahalanobis distance between the initial selected gas stations is:

[0020]

[0021] wherein, is the Mahalanobis distance between the th initial selected gas station, , , is the feature vector of the th initial selected gas station, is the feature vector of the th initial selected gas station, is the regularization matrix of the feature covariance matrix, is the inverse of the regularization matrix; , is the th feature observation of the th initial selected gas station, , is the th feature observation of the th initial selected gas station;

[0022] The Mahalanobis distance between the initial selected gas stations and the newly built gas station is:

[0023]

[0024] wherein, is the Mahalanobis distance between the th initial selected gas station and the newly built gas station, is the feature vector of the newly built gas station, , is the th feature observation of the newly built gas station, is the feature vector of the th initial selected gas station, , is the The first gas station selected in the preliminary round Each characteristic observation value .

[0025] One preferred embodiment is that the Mahalanobis distance is standardized, specifically as follows:

[0026] Calculate the mean and standard deviation of the Mahalanobis distances between the initially selected gas stations; the mean is:

[0027]

[0028] in, for The mean of the Mahalanobis distances between the initial selected gas stations; the standard deviation is:

[0029]

[0030] in, for Standard deviation of Mahalanobis distance between the initial selected gas stations;

[0031] Based on the mean and standard deviation of the Mahalanobis distances between the initially selected gas stations, the standardized Mahalanobis distance between the initially selected gas stations and the newly built gas stations is calculated; the standardized Mahalanobis distance is:

[0032]

[0033] in, The standardized Mahalanobis distance between the initially selected gas stations and the newly built gas stations.

[0034] One preferred embodiment is that the peak hour sales coefficients for each type of fuel at the analog gas station are:

[0035]

[0036] in, To compare with gas stations oil products Peak hour sales coefficient To compare with gas stations oil products Peak hourly sales volume To compare with gas stations oil products The average daily sales volume per year;

[0037] The peak hour traffic generation rates for each fuel type and vehicle type are as follows:

[0038]

[0039] in, To compare with gas stations oil products Model Traffic generation rate To compare with gas stations Peak hour oil prices Model Traffic generation volume.

[0040] One preferred embodiment is that step S2 predicts the peak-hour traffic generation volume of the newly built gas station based on the peak-hour sales coefficient of each fuel type at the gas station and the peak-hour traffic generation rate of each fuel type by vehicle type, including:

[0041] Based on the peak hour sales coefficients and peak hour traffic generation rates of each fuel type at comparable gas stations, combined with the Mahalanobis distance between comparable gas stations and newly built gas stations, the peak hour sales coefficients and peak hour traffic generation rates of each fuel type at newly built gas stations are predicted.

[0042] The peak-hour sales coefficients for each type of fuel at the newly built gas station are as follows:

[0043]

[0044] in, For the new gas station fuel Peak hour sales coefficient To compare with gas stations Mahal distance from the newly built gas station;

[0045] The peak hour traffic generation rates for each fuel type and vehicle type are as follows:

[0046]

[0047] in, For peak hour oil products Model Traffic generation rate.

[0048] One preferred embodiment is that step S2 predicts the peak-hour traffic generation volume of the newly built gas station based on the peak-hour sales coefficient of each fuel type at the gas station and the peak-hour traffic generation rate of each fuel type by vehicle type, including:

[0049] Based on the peak-hour sales coefficients of each fuel type at the newly built gas station and the peak-hour traffic generation rates of each fuel type by vehicle type, the peak-hour traffic generation volume of the newly built gas station by vehicle type is predicted as follows:

[0050]

[0051] in, For peak-hour vehicle types at newly built gas stations Traffic generation volume, For the peak hour fuel quality of newly built gas stations Model Traffic generation volume;

[0052] The newly built gas station's peak-hour fuel quality Model Traffic generation volume is:

[0053]

[0054] in, For the new gas station fuel Annual sales volume This refers to the number of operating days per year for newly built gas stations.

[0055] The above-described technical solution of this invention provides a method for predicting traffic demand for a newly constructed gas station project, comprising the following steps: obtaining preliminary gas stations within a target area based on traffic location conditions; calculating Mahalanobis distances between the preliminary gas stations and between the preliminary gas stations and the newly constructed gas station based on multi-dimensional feature data of each preliminary gas station; selecting N analog gas stations based on these Mahalanobis distances; obtaining the average daily annual sales volume of each fuel type, peak hour sales volume of each fuel type, and peak hour traffic generation volume of each fuel type by vehicle type for each of the N analog gas stations; calculating the peak hour sales coefficient and peak hour traffic generation rate of each fuel type by vehicle type for each of the analog gas stations; and predicting the peak hour traffic generation volume of the newly constructed gas station by vehicle type based on the peak hour sales coefficient and peak hour traffic generation rate of each fuel type by vehicle type for each of the analog gas stations. This invention solves the technical problem of how to accurately predict traffic demand for a newly constructed gas station project. Attached Figure Description

[0056] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0057] Figure 1 This is a schematic diagram of a traffic demand forecasting method for a newly built gas station project according to an embodiment of the present invention.

[0058] The realization of the objective, functional characteristics and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0059] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0060] The technical solutions of the various embodiments of the present invention can be combined with each other, but this must be based on the ability of a person skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0061] See Figure 1 According to one aspect of the present invention, the present invention provides a method for predicting traffic demand for a newly constructed gas station project, wherein the method for predicting traffic demand for a newly constructed gas station project includes the following steps:

[0062] S1. Based on the traffic location conditions, obtain the preliminary gas stations in the target area. Based on the multi-dimensional feature data of each preliminary gas station, calculate the Mahalanobis distance between the preliminary gas stations and between the preliminary gas stations and the newly built gas stations. Then, select N analog gas stations based on the Mahalanobis distance.

[0063] S2. Obtain the annual average daily sales volume of each oil product, the peak hour sales volume of each oil product, and the peak hour traffic generation volume of each oil product by vehicle type for N analog gas stations. Calculate the peak hour sales coefficient and peak hour traffic generation rate of each oil product by vehicle type for the analog gas stations. Based on the peak hour sales coefficient and peak hour traffic generation rate of each oil product by vehicle type for the analog gas stations, predict the peak hour traffic generation volume of the newly built gas station by vehicle type.

[0064] Specifically, in this embodiment, step S1, which involves obtaining preliminary gas stations within the target area based on traffic location conditions, specifically comprises: obtaining gas stations within the target area based on traffic location conditions. A preliminary selection of gas stations, all of which sell the same type of fuel; and acquisition of multi-dimensional feature data of the preliminary selection of gas stations to obtain a first feature dataset and a second feature dataset; in this invention... =7, This invention does not impose specific limitations; the specific settings can be configured as needed. The following text will refer to... =7 For example, the first feature dataset includes the operating land area of ​​the gas station, the number of fuel nozzles, annual sales, the number of parking spaces for motor vehicles within the service area of ​​the gas station, and the traffic flow of the road network within the service area; the number of feature categories in the first feature dataset is 5, and the second feature dataset includes the annual average daily sales volume of each oil product at the gas station, the peak hour sales volume of each oil product, and the traffic generation volume of each oil product by vehicle type during peak hours.

[0065] Specifically, in this embodiment, the newly built gas station is numbered 0, and the initially selected gas station is numbered... , The road network traffic within the service area of ​​the gas station refers to the total mileage traveled by vehicles in the road network during peak hours; the first feature dataset and the second feature dataset are shown in Table 1 and Table 2.

[0066] Table 1 First Feature Dataset

[0067] Gas station (number) Gas station operating area (m2) Gas station oil gun number (branch) Gas station annual sales (ten thousand yuan) Gas station service range of motor vehicle parking space (ten thousand) Gas station service range of road network traffic (ten thousand car kilometers / hour) 0 1110 16 12997 1.5 3.2 1 954 13 8262 1.4 3.9 2 1456 14 8270 1.4 3.5 3 1060 14 10217 1.8 3 4 1057 9 11781 1.5 3.8 5 1010 12 9671 2 3.1 6 1438 10 10526 1.8 3.3 7 951 14 11301 1.8 3.7

[0068] Table 2 Second Feature Dataset

[0069]

[0070] Specifically, in this embodiment, the oil types include 92-octane gasoline, 95-octane gasoline, 98-octane gasoline, and 0-grade diesel, and the number of oil types is as follows: =4, the vehicle types include motorcycles, small passenger cars, large passenger cars, small trucks, medium-sized trucks, and large trucks, the number of vehicle types is as follows. =6. See Table 3 for gas station fuel type codes and Table 4 for vehicle model codes. These can be adjusted appropriately based on actual conditions during use.

[0071] Table 3. Gas Station Fuel Type - Code Reference Table

[0072] Table 4 Vehicle Model Number Comparison Table

[0073]

[0074] Specifically, in this embodiment, step S1, calculating the Mahalanobis distance between the initially selected gas stations and between the initially selected gas stations and the newly built gas stations, specifically involves:

[0075] The feature covariance matrix is ​​calculated based on the first feature dataset. The feature covariance matrix is ​​then regularized and inverted. The Mahalanobis distances between the initially selected gas stations and between the initially selected gas stations and the newly built gas stations are calculated and standardized.

[0076] Specifically, in this embodiment, the feature covariance matrix is:

[0077]

[0078] in, for * The characteristic covariance matrix, For the first gas station selection The first feature and the second The covariance of each feature;

[0079] The first selected gas station The first feature and the second The covariance of each feature is:

[0080]

[0081] in, This represents the total number of gas stations selected in the initial screening. For the first The first gas station selected in the preliminary round Each characteristic observation value For the first gas station selection The mean of the characteristic observations, For the first gas station selection The mean of the characteristic observations;

[0082] The first selected gas station The mean of the feature observations is:

[0083]

[0084] The regularized covariance matrix is:

[0085]

[0086] in, The regularized covariance matrix is... As a parameter, in this invention, This invention does not impose specific limitations; the specific details can be set according to needs. It is a 5x5 identity matrix.

[0087] Specifically, in this embodiment, the Mahalanobis distance between the initially selected gas stations is:

[0088]

[0089] in, For the first The Mahalanobis distance between the initial gas stations , , For the first Feature vectors of the initially selected gas stations For the first Feature vectors of the initially selected gas stations The regularization matrix is ​​the eigencovariance matrix. The inverse of the regularization matrix; , For the first The first gas station selected in the preliminary round Each characteristic observation value , For the first The first gas station selected in the preliminary round Each characteristic observation value;

[0090] The Mahalanobis distance between the initially selected gas station and the newly built gas station is:

[0091]

[0092] in, For the first The Mahalanobis distance between the initial selected gas stations and the newly built gas stations The feature vector of the newly built gas station. , For the newly built gas station Each characteristic observation value For the first Feature vectors of the initially selected gas stations , For the first The first gas station selected in the preliminary round Each characteristic observation value .

[0093] Table 5. Mahalanobis distances between each of the eight gas stations

[0094]

[0095] Specifically, in this embodiment, the Mahalanobis distance is standardized as follows:

[0096] Calculate the mean and standard deviation of the Mahalanobis distances between the initially selected gas stations; the mean is:

[0097]

[0098] in, for The mean of the Mahalanobis distances between the initial selected gas stations; the standard deviation is:

[0099]

[0100] in, for Standard deviation of Mahalanobis distance between the initial selected gas stations;

[0101] Based on the mean and standard deviation of the Mahalanobis distances between the initially selected gas stations, the standardized Mahalanobis distance between the initially selected gas stations and the newly built gas stations is calculated; the standardized Mahalanobis distance is:

[0102]

[0103] in, The standardized Mahalanobis distance between the initially selected gas stations and the newly built gas stations.

[0104] Table 6 Standardized Mahalanobis distance between the initially selected gas stations and the newly built gas stations

[0105]

[0106] Specifically, in this embodiment, step S1, which selects N analog gas stations based on Mahalanobis distance, specifically involves:

[0107] Set threshold ,pass Standardized Mahalanobis distance between the initially selected gas stations and the newly built gas stations Filtering is performed, i.e., those that meet the conditions are selected. ≤ The initial selection of gas stations is designated as analogous gas stations, and their corresponding number is denoted as M. If M ≥ 3, these M gas stations are determined as analogous gas stations. If M < 3, additional gas stations need to be selected, and the above steps are repeated until M ≥ 3. The final number of analogous gas stations is then determined to be N. After screening the initial gas stations, N analog gas stations were obtained. The Mahalanobis distance between these N analog gas stations and the newly built gas stations is: .

[0108] Specifically, in this embodiment, when =1, specifically meaning that the Mahalanobis distance between the analogous gas stations and the newly built gas stations, and the mean of the Mahalanobis distance between them and the initially selected gas stations, does not exceed 1 standard deviation. See Table 6 for the standardized Mahalanobis distances of the initially selected gas stations and the newly built gas stations numbered 1, 2, 3, 4, 5, 6, and 7. Therefore, these six initial gas stations were identified as analogous gas stations. For ease of subsequent description, the six analogous gas stations were renumbered sequentially. The Mahalanobis distance between it and the newly built gas station is denoted as . , .

[0109] Specifically, in this embodiment, the peak hour sales coefficients for each type of oil at the analog gas station are as follows:

[0110]

[0111] in, To draw an analogy with gas stations oil products Peak hour sales coefficient To draw an analogy with gas stations the oil products the peak hour sales volume, the analog gas station the oil products the annual average daily sales volume; =1, indicates No. 92 gasoline, =2, indicates No. 95 gasoline, =3, indicates No. 98 gasoline, =4, indicates No. 0 diesel oil; =1, indicates a motorcycle, =2, indicates a small passenger car, =3, indicates a large passenger car, =4, indicates a small truck, =5, indicates a medium truck, =6, indicates a large truck;

[0112] The peak hour vehicle generation rate of each oil product is:

[0113]

[0114] wherein, the analog gas station the oil products the vehicle type the vehicle generation rate, the peak hour oil product the vehicle type the vehicle generation volume of the analog gas station; The peak hour sales coefficient of each oil product of the analog gas station and the peak hour vehicle generation rate of each oil product are shown in Tables 7 and 8.

[0115] Table 7 Peak hour sales coefficient of each oil product of the analog gas station

[0116]

[0117] Table 8 Peak hour vehicle generation rate of each oil product

[0118] Specifically, in the present embodiment, the step S2 predicts the peak hour vehicle generation volume of the new gas station based on the peak hour sales coefficient of each oil product of the analog gas station and the peak hour vehicle generation rate of each oil product, and includes:

[0119]

[0120] ​Based on the peak hour sales coefficients and peak hour traffic generation rates of each fuel type at comparable gas stations, combined with the Mahalanobis distance between comparable gas stations and newly built gas stations, the peak hour sales coefficients and peak hour traffic generation rates of each fuel type at newly built gas stations are predicted.

[0121] The peak-hour sales coefficients for each type of fuel at the newly built gas station are as follows:

[0122]

[0123] in, For the new gas station fuel Peak hour sales coefficient To draw an analogy with gas stations Mahal distance from the newly built gas station;

[0124] The peak hour traffic generation rates for each fuel type and vehicle type are as follows:

[0125]

[0126] in, For peak hour oil products Model Traffic generation rate.

[0127] Table 9. Peak-hour sales coefficients and peak-hour traffic generation rates of various fuel types for newly built gas stations.

[0128]

[0129] Specifically, in this embodiment, step S2 predicts the peak-hour traffic generation volume of the newly built gas station based on the peak-hour sales coefficient of each oil product at the gas station and the peak-hour traffic generation rate of each oil product by vehicle type, including:

[0130] Based on the peak-hour sales coefficients of each fuel type at the newly built gas station and the peak-hour traffic generation rates of each fuel type by vehicle type, the peak-hour traffic generation volume of the newly built gas station by vehicle type is predicted as follows:

[0131]

[0132] in, For peak-hour vehicle types at newly built gas stations Traffic generation volume, For the peak hour fuel quality of newly built gas stations Model Traffic generation volume;

[0133] The newly built gas station's peak-hour fuel quality Model Traffic generation volume is:

[0134]

[0135] in, For the new gas station fuel Annual sales volume This refers to the number of operating days per year for newly built gas stations.

[0136] Specifically, in this embodiment, based on the projected annual sales volume of each oil product in the feasibility study report of the new gas station (the annual sales volumes of 92-octane gasoline, 95-octane gasoline, 98-octane gasoline and 0# diesel are 4457.89, 3984.21, 3615.79 and 3042.11 tons respectively) and the number of operating days per year (365 days), the traffic generation volume of each oil product by vehicle type during the peak hours of the new gas station is obtained.

[0137] Table 10 Traffic Generation by Fuel Type and Vehicle Type During Peak Hours at Newly Built Refueling Stations

[0138]

[0139] The above are merely preferred embodiments of the present invention and do not limit the patent scope of the present invention. All equivalent structural transformations made using the contents of the present invention's specification and drawings under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A method for forecasting traffic demand for a new gas station project, characterized in that, The method comprises the following steps: S1, obtaining initial selected gas stations in a target area according to traffic site conditions, calculating Mahalanobis distances between the initial selected gas stations and between the initial selected gas stations and a new gas station based on multi-dimensional feature data of the initial selected gas stations, and selecting N analog gas stations according to the Mahalanobis distances; The step S1 selects the N analog gas stations according to the Mahalanobis distances, specifically as follows: Set threshold ,pass Standardized Mahalanobis distance between the initially selected gas stations and the newly built gas stations Filtering is performed, i.e., those that meet the conditions are selected. ≤ The initial selection of gas stations is designated as analogous gas stations, and their corresponding number is denoted as M. If M ≥ 3, these M gas stations are determined as analogous gas stations. If M < 3, additional gas stations need to be selected, and the above steps are repeated until M ≥ 3. The final number of analogous gas stations is then determined to be N. After screening the initial gas stations, N analog gas stations were obtained. The Mahalanobis distance between these N analog gas stations and the newly built gas station is: ; S2, obtaining annual average daily sales of each oil product, peak hour sales of each oil product and peak hour traffic generation of each oil product by vehicle type of the N analog gas stations, calculating peak hour sales coefficients of each oil product and peak hour traffic generation rates of each oil product by vehicle type of the analog gas stations, and predicting peak hour traffic generation of each oil product by vehicle type of the new gas station based on the peak hour sales coefficients of each oil product and the peak hour traffic generation rates of each oil product by vehicle type of the analog gas stations; The peak hour traffic generation of each oil product by vehicle type of the new gas station is predicted based on the peak hour sales coefficients of each oil product and the peak hour traffic generation rates of each oil product by vehicle type of the analog gas stations, specifically as follows: According to the peak hour sales coefficients of each oil product and the peak hour traffic generation rates of each oil product by vehicle type of the analog gas stations, the peak hour sales coefficients of each oil product and the peak hour traffic generation rates of each oil product by vehicle type of the new gas station are predicted in combination with the Mahalanobis distances between the analog gas stations and the new gas station. The peak hour traffic generation of each oil product by vehicle type of the new gas station is predicted according to the peak hour sales coefficients of each oil product and the peak hour traffic generation rates of each oil product by vehicle type of the new gas station.

2. The method of claim 1, wherein, The step S1 obtains the initial selected gas stations in the target area according to the traffic site conditions, specifically as follows: According to the traffic location condition, the target region is acquired The initial selected gas stations selling the same oil products are acquired, and multi-dimensional feature data of the initial selected gas stations are acquired to obtain a first feature data set and a second feature data set.

3. The method of claim 2, wherein, The first feature data set includes operating land area of the gas station, number of oil guns, annual sales, number of motor vehicle parking spaces in the service range of the gas station and road network flow in the service range. The second feature data set includes annual average daily sales of each oil product, peak hour sales of each oil product and peak hour traffic generation of each oil product by vehicle type.

4. The method of claim 2, wherein, The step S1 calculates the Mahalanobis distances between the initial selected gas stations and between the initial selected gas stations and the new gas station, specifically as follows: The feature covariance matrix is calculated according to the first feature data set, and the feature covariance matrix is subjected to regularization processing and inversion to calculate the Mahalanobis distances between the initial selected gas stations and between the initial selected gas stations and the new gas station, and the Mahalanobis distances are standardized.

5. The method of claim 4, wherein, The feature covariance matrix is as follows: ； wherein is * a feature covariance matrix, is a covariance between the first feature and the second feature; The first selected gas station has a first feature The covariance of the first feature and the second feature is: The covariance of the first feature and the second feature is: ； wherein, is the total number of primary gas stations, is the first feature observation of the th primary gas station, is the first feature observation of the th primary gas station, is the mean of the first feature observation of the th primary gas station, is the mean of the first feature observation of the th primary gas station.

6. The method of claim 5, wherein, The Mahalanobis distance between the initial selected gas stations is as follows: ； in, For the first The Mahalanobis distance between the initial gas stations , , For the first Feature vectors of the initially selected gas stations For the first Feature vectors of the initially selected gas stations The regularization matrix is ​​the eigencovariance matrix. The inverse of the regularization matrix; , For the first The first gas station selected in the preliminary round Each characteristic observation value , For the first The first gas station selected in the preliminary round Each characteristic observation value; The Mahalanobis distance between the initial selected gas stations and the new gas station is as follows: ； wherein, is the Mahalanobis distance between the jth preliminary gas station and the new gas station, is the feature vector of the new gas station, , is the jth feature observation of the new gas station, is the jth feature vector of the ith preliminary gas station, , is the jth feature observation of the ith preliminary gas station, , , is the jth feature observation of the ith preliminary gas station, , , .

7. The method of claim 6, wherein, The Mahalanobis distances are standardized, specifically as follows: The mean and the standard deviation of the Mahalanobis distances between the initial selected gas stations are calculated respectively; the mean is as follows: ； wherein is the mean of the Mahalanobis distances between the initial selected gas stations; the standard deviation is: ； wherein, is the standard deviation of the Mahalanobis distance between the individual primary filling stations; The standardized Mahalanobis distance between the initial selected gas stations and the new gas station is calculated according to the mean and the standard deviation of the Mahalanobis distances between the initial selected gas stations; the standardized Mahalanobis distance is as follows: ； wherein, is the standardized Mahalanobis distance between the initial selected gas station and the newly built gas station.

8. The method for forecasting traffic demand of a new gas station project according to any one of claims 1-7, characterized in that, The peak hour sales coefficient of each oil product of the analog gas station is as follows: ； wherein, a peak hour sales factor for the oil products of the analogous gas station, a peak hour sales factor for the oil products of the analogous gas station, a peak hour sales factor for the oil products of the analogous gas station, a peak hour sales factor for the oil products of the analogous gas station, a peak hour sales factor for the oil products of the analogous gas station, a peak hour sales factor for the oil products of the analogous gas station, a peak hour sales factor for the oil products of the analogous gas station, a peak hour sales factor for the oil products of the analogous gas station, a peak hour sales factor for the oil products of the analogous gas station, The peak hour traffic generation rate of each oil product by vehicle type is as follows: ； wherein, the traffic generation rate for a vehicle type at an analog gas station, the volume of fuel for a vehicle type at an analog gas station, the traffic generation rate for a vehicle type at an analog gas station, the volume of fuel for a vehicle type at an analog gas station, the traffic generation rate for a vehicle type at an analog gas station, the volume of fuel for a vehicle type at an analog gas station, the traffic generation rate for a vehicle type at an analog gas station, the volume of fuel for a vehicle type at an analog gas station.

9. The method of claim 8, wherein, The peak hour sales coefficient of each oil product of the new gas station is as follows: ； wherein, for new gas station oil products peak hour sales factor, for analog gas stations Mahalanobis distance from new gas stations; The peak hour traffic generation rate of each oil product by vehicle type is as follows: ； wherein, peak hour oil product vehicle type traffic generation rate.

10. The method of claim 9, wherein, The peak hour traffic generation of each oil product by vehicle type of the new gas station is as follows: ； wherein, is the traffic generation for the new gas station for the peak hour vehicle type, is the traffic generation for the new gas station for the peak hour fuel type; is the number of fuel types.​​​ The new gas station peak hour oil products Vehicle type Traffic generation is: ； wherein, the annual sales of oil products at the new gas station, the annual sales of oil products at the new gas station, the annual sales of oil products at the new gas station,

Images