A method for calculating the charging brand preference of electric vehicle users

Abstract

The present invention provides a method for calculating the charging brand preference of electric vehicle users, comprising the following steps: A. obtaining charging facilities and electric vehicle related information; B. calculating the electric vehicle user charging feature matrix according to the charging facilities and electric vehicle related information obtained in step A; C. identifying the electric vehicle user charging time period and brand according to the electric vehicle user charging feature matrix calculated in step B; D. calculating the electric vehicle user charging brand preference according to the electric vehicle user charging time period and brand obtained in step C. The electric vehicle user charging brand preference identification method proposed in the present invention overcomes the limitations of the questionnaire method.

Description

Technical Field

[0001] The invention relates to the technical field of electric vehicle data analysis and application, and in particular to a method for calculating the charging brand preference of electric vehicle users. Background Art

[0002] Since electric vehicles are efficient, clean, low-carbon and environmentally friendly, they can improve energy strategic security and reduce greenhouse gas emissions. The country has proposed a new infrastructure strategy to vigorously develop electric vehicle charging stations.

[0003] There are many brands of electric vehicle charging facilities, including Teladian, Xingxing, State Grid, etc. The preferences of electric vehicle users for charging facilities are generally obtained through questionnaire surveys, and the results obtained through questionnaire surveys are often limited. By calculating the user's preference for charging brands, data can be provided for the planning and operation of charging facilities.

[0004] The Chinese invention patent application number 201710110119.8, "Method and device for predicting user brand preference behavior based on tags", obtains the user's online behavior data and e-commerce browsing data to obtain the user's preference for brand information and calculate the user's brand preference. The Chinese invention patent application number 201310286786.3, "Method, device and system for obtaining user preference for multimedia data", calculates the user preference for each multimedia requested to be played according to the number of multimedia requested to be played and the number of requested playbacks.

[0005] The existing methods for calculating brand preference involve traditional questionnaires, which can only obtain the brand preference of some people and have limitations. Patent application number 201710110119.8 is applicable to the calculation of online shopping brand preference, and patent application number 201310286786.3 is applicable to the calculation of multimedia user preference, but not to the calculation of electric vehicle user charging brand preference. Summary of the invention

[0006] In view of the above-mentioned deficiencies in the prior art, the present invention provides a method for calculating the charging brand preference of electric vehicle users.

[0007] The technical solution adopted by the present invention is:

[0008] A method for calculating the charging brand preference of electric vehicle users comprises the following steps:

[0009] A. Obtain information about charging facilities and electric vehicles;

[0010] B. Calculate the electric vehicle user charging feature matrix based on the charging facilities and electric vehicle related information obtained in step A;

[0011] C. Identify the charging time period and brand of the electric vehicle user according to the electric vehicle user charging feature matrix calculated in step B;

[0012] D. Calculate the electric vehicle user's charging brand preference based on the electric vehicle user's charging time period and brand obtained in step C.

[0013] Furthermore, step A obtains information related to charging facilities and electric vehicle users, including: charging station name, charging station operating brand name, charging station latitude and longitude, electric vehicle user number, time, electric vehicle charging status, and electric vehicle latitude and longitude.

[0014] Furthermore, in step B, based on the charging facilities and electric vehicle related information obtained in step A, the electric vehicle user charging feature matrix is ​​calculated, which specifically includes:

[0015] If there are n charging stations within the analysis range, the characteristic matrix of the charging stations is as follows:

[0016]

[0017] where NA i represents the name of the i-th charging station, PP i represents the brand of the i-th charging station, LonS i represents the longitude of the i-th charging station, LatS1 represents the latitude of the i-th charging station, i=1,2…,n

[0018] If the time range of the analysis is (1, T), and there are m electric vehicle users, then for electric vehicle user j, its user feature matrix is ​​expressed as:

[0019]

[0020] Among them CS t,j represents the charging status of electric vehicle user j at time t. When parking for charging, CS t,j =1, otherwise CS t,j =0; LonC t,j represents the longitude of the electric car user j at time t; LatC t,j represents the dimension of the electric car of electric car user j at time t. Where 0≤t≤T, 1≤j≤m;

[0021] Calculate the distance between electric vehicle user j and charging station i based on their latitude and longitude coordinates:

[0022] C=sin(LatS i ·π / 180)·sin(LatC j π / 180)+

[0023] cos(LatS i ·π / 180)·cos(LatC j ·π / 180)·cos((LonS i -LonC j )·π / 180)

[0024] D i,j = R·Arccos(C)·π / 180

[0025] Where D i,j is the distance between electric vehicle user j and charging station i;

[0026] The electric vehicle user-related information is expanded according to the calculated distance to form an electric vehicle user feature matrix:

[0027]

[0028] PS t,j is the characteristic value of the distance between electric vehicle user j and the electric vehicle charging station at time t, and the calculation rule is as follows:

[0029] If at time t, PS t,j =0;

[0030] If at time t, And the distance between electric vehicle user j and charging station i is the shortest, then PS t,j =PP i .

[0031] Furthermore, according to the electric vehicle user charging feature matrix calculated in step B, the charging time period and brand of the electric vehicle user are identified, specifically:

[0032] The expanded electric vehicle user feature matrix is:

[0033]

[0034] First, identify the time period when electric vehicle users charge:

[0035] For any time t1, if CS t1,j =1, and CS t1-1,j =0, then time t1 is the charging start period;

[0036] For any time t2, if CS t2,j =1, and CS t2+1,j =0, then time t1 is the end of charging period;

[0037] Then the time period [t1, t2] is the user's charging time period. In the time range [1, T], the charging time period set of electric vehicle user j is recorded as {CD 1,j ,CD 2,j ,…,CD K,j}, where CD k,j represents the kth charging period of electric vehicle user j, CD k,j =[t 1,k ,t 2,k ], calculate the user's charging brand set {CDP according to the user feature matrix CSP 1,j ,CDP 2,j ,…,CDP K,j};

[0038] The charging period CD k,j In [t 1,k ,t 2,k ] The corresponding distance feature quantity in the period is

[0039] like This indicates that the user did not charge at the charging station, and the charging brand of the corresponding period is recorded as CDP k,j =0;

[0040] like Then count the charging station brand that appears most frequently during this period. If it is charging station i, it is recorded as CDP k,j =PP i .

[0041] Furthermore, in step D, the electric vehicle user's charging brand preference is calculated based on the electric vehicle user's charging time period and brand obtained in step C, specifically:

[0042] For user j, charging brand PP i , according to the charging brand set {CDP 1,j ,CDP 2,j ,…,CDP K,j}, statistics charging brand PP i Number of occurrences in the set PPN i , statistics charging brand collection {CDP 1,j ,CDP 2,j ,…,CDP K,j}Number of times PPT is not 0, calculate the charging brand PP i Preference

[0043] The present invention provides a method for calculating the charging brand preference of electric vehicle users. The method mines relevant data of electric vehicle users and charging facilities through big data analysis to calculate the charging brand preference of electric vehicle users, thus overcoming the limitations of the questionnaire method. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a flow chart of one embodiment of a method for calculating the charging brand preference of electric vehicle users of the present invention. DETAILED DESCRIPTION

[0045] The technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the present invention.

[0046] See also Figure 1 , is a flow chart of one embodiment of a method for calculating a charging brand preference of an electric vehicle user according to the present invention, the method comprising the following steps:

[0047] A. Obtain information about charging facilities and electric vehicles, including: charging station name, charging station operating brand name, charging station latitude and longitude, electric vehicle user number, time, electric vehicle charging status, and electric vehicle latitude and longitude;

[0048] B. Based on the information obtained in step A, calculate the electric vehicle user charging feature matrix, including:

[0049] If there are n charging stations within the analysis range, the characteristic matrix of the charging stations is as follows:

[0050]

[0051] where NA i represents the name of the i-th charging station, PP i represents the brand of the i-th charging station, LonS i represents the longitude of the i-th charging station, LatS1 represents the latitude of the i-th charging station, i=1,2…,n

[0052] If the time range of the analysis is (1, T), and there are m electric vehicle users, then for electric vehicle user j, its user feature matrix can be expressed as:

[0053]

[0054] Among them CS t,j represents the charging status of electric vehicle user j at time t. When parking for charging, CS t,j =1, otherwise CS t,j =0; LonC t,j represents the longitude of the electric car user j at time t; LatCt,j It represents the dimension of the electric car of electric car user j at time t. Where 0≤t≤T, 1≤j≤m.

[0055] The distance between the electric vehicle user and the charging station is calculated based on their latitude and longitude coordinates, taking user j and the i-th charging station as an example:

[0056] C=sin(LatS i ·π / 180)·sin(LatC j π / 180)+

[0057] cos(LatS i ·π / 180)·cos(LatC j ·π / 180)·cos((LonS i -LonC j )·π / 180)

[0058] D i,j = R·Arccos(C)·π / 180

[0059] Where D i,j is the distance between electric vehicle user j and charging station i.

[0060] According to the calculated distance, the relevant information of electric vehicle users is expanded to form a user feature matrix. Taking user j as an example:

[0061]

[0062] PS t,j is the characteristic value of the distance between electric vehicle user j and the electric vehicle charging station at time t, and the calculation rule is as follows:

[0063] If at time t, PS t,j =0;

[0064] If at time t, And the distance between electric vehicle user j and charging station i is the shortest, then PS t,j =PP i ;

[0065] C. According to the electric vehicle user charging feature matrix calculated in step B, identify the charging time period and brand of the electric vehicle user, specifically:

[0066] The expanded electric vehicle user feature matrix is:

[0067]

[0068] First, identify the charging period of electric vehicle users, taking electric vehicle user j as an example:

[0069] For any time t1, if CS t1,j =1, and CS t1-1,j =0, then time t1 is the charging start period;

[0070] For any time t2, if CS t2,j =1, and CS t2+1,j =0, then time t1 is the end of charging period;

[0071] Then the time period [t1, t2] is the user's charging time period. In the time range [1, T], the charging time period set of electric vehicle users is recorded as {CD 1,j ,CD 2,j ,…,CD K,j}, where CD k,j represents the kth charging period of electric vehicle user j, CD k,j =[t 1,k ,t 2,k ] Calculate the user's charging brand set {CDP according to the user feature matrix CSP 1,j ,CDP 2,j ,…,CDP K,j}.

[0072] CD with charging period k,j For example, in [t 1,k ,t 2,k ] The corresponding distance feature quantity in the period is

[0073] like This indicates that the user did not charge at the charging station, and the charging brand for the corresponding period is recorded as CDP k,j =0;

[0074] like The brand of the charging station that appears most frequently during this period is counted. If it is charging station i, it is recorded as CDP k,j =PP i .

[0075] D. Calculate the charging brand preference of electric vehicle users based on the calculation results of step C, specifically:

[0076] Based on user j and charging brand PP i For example, according to the charging brand set {CDP 1,j ,CDP 2,j ,…,CDP K,j}, statistics charging brand PP i Number of occurrences in the set PPN i , statistics charging brand collection {CDP 1,j ,CDP2,j ,…,CDP K,j}Number of times PPT is not 0, calculate the charging brand PP i Preference

[0077] The technical solution and effects of the present invention are described in detail below with a specific embodiment:

[0078] Step A, obtain information about charging facilities and electric vehicle users, as shown in the following table (sample data). Charging station name, charging station operating brand name, charging station latitude and longitude, electric vehicle user number, time, electric vehicle charging status, electric vehicle latitude and longitude

[0079] Table 1 Charging station information

[0080]

[0081] Table 2 Electric vehicle user information

[0082]

[0083] Step B, based on the information obtained in step A, calculate the electric vehicle user charging feature matrix, taking user j as an example:

[0084] Table 3 Electric vehicle user charging characteristic matrix

[0085]

[0086]

[0087] Step C: The electric vehicle user charging feature matrix calculated in step B identifies the charging time period and brand of the electric vehicle user, and obtains the user's charging brand set as {CDP 1,j ,CDP 2,j ,…,CDP K,j}={Telaidian, Telaidian, ..., Xingxing}.

[0088] Step D, based on the calculation result of step C, calculate the charging brand preference of electric vehicle users, and obtain that user j's preference for Teladian is 80%, and the preference for Xingxing is 20%.

[0089] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by technicians in this technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for calculating the charging brand preference of electric vehicle users, characterized in that The following steps are involved: A. Obtain information about charging facilities and electric vehicles; B. Calculate the electric vehicle user charging feature matrix based on the charging facilities and electric vehicle related information obtained in step A; C. Identify the charging time period and brand of the electric vehicle user according to the electric vehicle user charging feature matrix calculated in step B; D. Calculate the electric vehicle user charging brand preference based on the charging period and brand of the electric vehicle user obtained in step C. In step B, calculate the electric vehicle user charging feature matrix based on the charging facilities and electric vehicle related information obtained in step A, specifically including: If there are n charging stations within the analysis range, the characteristic matrix of the charging stations is as follows: where NA i represents the name of the i-th charging station, PP i represents the brand of the i-th charging station, LonS i represents the longitude of the i-th charging station, LatS1 represents the latitude of the i-th charging station, i=1,2…,n If the time range of the analysis is (1, T), and there are m electric vehicle users, then for electric vehicle user j, its user feature matrix is ​​expressed as: Among them CS t,j represents the charging status of electric vehicle user j at time t. When parking for charging, CS t,j =1, otherwise CS t,j =0; LonC t,j represents the longitude of the electric car user j at time t; LatC t,j represents the dimension of the electric vehicle of electric vehicle user j at time t, where 0≤t≤T, 1≤j≤m; Calculate the distance between electric vehicle user j and charging station i based on their latitude and longitude coordinates: C=sin(LatS i ·π / 180)·sin(LatC j ·π / 180)+ cos(LatS i ·π / 180)·cos(LatC j ·π / 180)·cos((LonS i -LonC j )·π / 180) D i,j =R·Arccos(C)·π / 180 Where D i,j is the distance between electric vehicle user j and charging station i; The electric vehicle user-related information is expanded according to the calculated distance to form an electric vehicle user feature matrix: PS t,j is the characteristic value of the distance between electric vehicle user j and the electric vehicle charging station at time t, and the calculation rule is as follows: If at time t, PS t,j =0; If at time t, And the distance between electric vehicle user j and charging station i is the shortest, then PS t,j =PP i ; In step C, the charging time period and brand of the electric vehicle user are identified according to the electric vehicle user charging feature matrix calculated in step B, specifically: The expanded electric vehicle user feature matrix is: First, identify the time period when electric vehicle users charge: For any time t1, if CS t1,j =1, and CS t1-1,j =0, then time t1 is the charging start period; For any time t2, if CS t2,j =1, and CS t2+1,j =0, then time t1 is the end of charging period; Then the time period [t1, t2] is the user's charging time period. In the time range [1, T], the charging time period set of electric vehicle user j is recorded as {CD 1,j ,CD 2,j ,…,CD k,j }, where CD k,j represents the kth charging period of electric vehicle user j, CD k,j =[t 1,k ,t 2,k ], calculate the user's charging brand set {CDP according to the user feature matrix CSP 1,j ,CDP 2,j ,…,CDP k,j }; The charging period CD k,j In [t 1,k ,t 2,k ] The corresponding distance feature quantity in the period is like This indicates that the user did not charge at the charging station, and the charging brand of the corresponding period is recorded as CDP k,j =0; like The brand of the charging station that appears most frequently during this period is counted. If it is charging station i, it is recorded as CDP k,j =PP i .

2. A method for calculating the charging brand preference of electric vehicle users as claimed in claim 1, characterized in that: Step A obtains information related to charging facilities and electric vehicle users, including: charging station name, charging station operating brand name, charging station latitude and longitude, electric vehicle user number, time, electric vehicle charging status, and electric vehicle latitude and longitude.

3. The method for calculating the charging brand preference of electric vehicle users according to claim 1, characterized in that: In step D, the electric vehicle user's charging brand preference is calculated based on the electric vehicle user's charging time period and brand obtained in step C, specifically: For user j, charging brand PP i , according to the charging brand set {CDP 1,j ,CDP 2,j ,…,CDP k,j }, statistics charging brand PP i Number of occurrences in the set PPN i , statistics charging brand collection {CDP 1,j ,CDP 2,j ,…,CDP k,j }Number of times PPT is not 0, calculate the charging brand PP i Preference

Images