A charging pile performance evaluation method and device based on cloud platform

Abstract

The invention discloses a charging pile performance evaluation method and device based on a cloud platform. The method comprises: acquiring charging pile maintenance data through a cloud platform; calculating the maintenance rate under various indicators according to the maintenance data; establishing a gray cloud model to calculate the membership of various indicators at different levels; calculating the weight matrix of various indicators based on the AHP hierarchy analysis method; obtaining a comprehensive clustering coefficient according to the membership of various indicators at different levels and the weight matrix; calculating a relative indicator state value according to the comprehensive clustering coefficient; determining a comprehensive correction rate according to the relative indicator state value by using the TOPSIS method; and evaluating the charging pile performance based on the comprehensive correction rate. The method can effectively improve the credibility of the evaluation.

Description

Technical Field

[0001] The present invention relates to the technical field of charging piles, and in particular to a charging pile performance evaluation method and device based on a cloud platform. Background Art

[0002] As the output of electric vehicles increases year by year, the scale of charging piles is also getting larger and larger. As one of the infrastructures of electric vehicles, the safety and reliability of charging piles are particularly important. At present, charging piles mainly transmit data through wired networks, wireless networks, etc., and use cloud platforms to store and process data. The cloud platform can not only meet the storage needs of huge amounts of data, but also analyze the data. For example, patent document CN113285989A discloses an intelligent management method and system for electric vehicle charging stations based on the Internet of Things. The data acquisition layer collects charging equipment, charging field environment and behavior data, manages and dispatches charging fields distributed in different areas, traces information and analyzes data, and realizes intelligent management of charging fields. Through the analysis of cloud platform big data, the performance of charging piles can be evaluated. The performance of existing charging piles is often evaluated only through simple maintenance conditions. However, my country has a vast territory, and the climate and environment in various places are quite different. Climate factors and environmental factors have a certain impact on charging piles. Therefore, the existing charging pile performance evaluation method cannot comprehensively and effectively evaluate the performance of charging piles, with large errors and low evaluation credibility. Summary of the invention

[0003] The present invention provides a charging pile performance evaluation method and device based on a cloud platform, which can effectively improve the credibility of the charging pile performance evaluation.

[0004] A charging pile performance evaluation method based on a cloud platform, comprising:

[0005] Obtain charging pile maintenance data through the cloud platform;

[0006] Calculate the maintenance rate under each indicator according to the maintenance data;

[0007] Establish the gray cloud model and calculate the membership degree of each index at different levels;

[0008] Calculate the weight matrix of each index based on AHP;

[0009] According to the different levels of membership of each indicator and the weight matrix, a comprehensive clustering coefficient is obtained;

[0010] Calculating a relative indicator status value according to the comprehensive clustering coefficient;

[0011] According to the relative indicator status value, the comprehensive correction rate is determined by using the TOPSIS method;

[0012] The charging pile performance is evaluated based on the comprehensive correction rate.

[0013] Furthermore, the maintenance rate under each indicator is calculated by the following formula:

[0014]

[0015] Among them, x i is the charging pile maintenance rate under indicator i, Apl ij is the number of maintenance of type j charging piles under index i, ∑Apl ij The number of repairs for all charging stations.

[0016] Furthermore, the indicators include charging pile brand, charging pile type, type of area where the charging pile is located, average service life of the charging pile, average daily utilization rate of the charging pile, spring temperature range, summer temperature range, autumn temperature range and winter temperature range.

[0017] Furthermore, a gray cloud model is established to calculate the membership of different levels of various indicators, including:

[0018] Establish maintenance rate levels;

[0019] The maintenance rate level is used as a gray class of a gray cloud model, and the left boundary and the right boundary of each gray class are determined according to the interval value of the maintenance rate level;

[0020] Generate a normal random number according to the left boundary and the right boundary;

[0021] Determine a white weighting function for each gray class according to the normal random number and the maintenance rate under the index;

[0022] Calculating a white weighted value according to the white weighted function;

[0023] The degree of membership is calculated according to the white weighted value.

[0024] Furthermore, the normal random number En* has En as the expectation and He as the standard deviation:

[0025]

[0026]

[0027] Rx is the right boundary, Lx is the left boundary, En is the expectation, l and b are coefficients;

[0028] The white weighted function of the maintenance rate under the indicator with respect to the moderate measure of the gray class k is as follows:

[0029]

[0030] The white weighted function of the upper limit measure of the maintenance rate under the indicator with respect to the gray class k is as follows:

[0031]

[0032] The normal gray cloud model white weighted function of the maintenance rate under the indicator with respect to the lower limit measure of gray class k is as follows:

[0033]

[0034] Among them, xi is the maintenance rate under index i, Ex is the expected value, E x =0.5*(R x +L x ), Rx is the right boundary, Lx is the left boundary, is a normal random number about index i, is the white weighting function;

[0035] The white weighted value under each gray class is the average value of the calculation results of multiple white weighted functions;

[0036] The membership degree is calculated by the following formula:

[0037]

[0038] in, is the membership degree of index i to gray class k, is the white weighted value of index i with respect to gray class k, and n is the number of gray classes.

[0039] Furthermore, the weight matrix of each indicator is calculated based on the AHP hierarchical analysis method, including:

[0040] Set the comprehensive maintenance rate as the target layer, and the maintenance rate of each indicator as the criterion layer, and establish a multi-level analysis model;

[0041] Establishing a judgment matrix, by constructing the judgment matrix to compare the relative importance between two elements in the multi-level analysis model;

[0042] The maximum eigenvalue and eigenvector of the judgment matrix are calculated to obtain a weight matrix.

[0043] Furthermore, the comprehensive clustering coefficient is calculated by the following formula:

[0044]

[0045] Among them, σ k is the comprehensive clustering coefficient, m is the number of indicators, W i is the weight matrix, is the membership degree of index i to gray class k.

[0046] Furthermore, the relative indicator status value is calculated by the following formula:

[0047]

[0048] Among them, f is the relative indicator state value, k represents the gray class, σ k is the comprehensive clustering coefficient, Ex k Represents the expectation of gray class k.

[0049] Furthermore, the TOPSIS method is used to determine the comprehensive correction rate, including:

[0050] The TOPSIS method is used to determine the positive ideal solution and the negative ideal solution;

[0051] Calculate the positive and negative distances of each index from the positive ideal solution and the negative ideal solution:

[0052]

[0053] Among them, f + is a positive ideal solution, f - is a negative ideal solution, S + is the forward distance, S - is the negative distance, and f is the relative indicator state value;

[0054] According to the positive distance and the negative distance, the comprehensive correction rate is calculated:

[0055]

[0056] Among them, η is the comprehensive correction rate.

[0057] Furthermore, the charging pile performance evaluation is performed based on the comprehensive correction rate, including:

[0058] Obtain local assessment scores through the cloud platform;

[0059] A comprehensive score is calculated based on the evaluation score and the comprehensive correction rate.

[0060] A charging pile performance evaluation device based on a cloud platform, comprising:

[0061] A data acquisition module is used to obtain charging pile maintenance data through the cloud platform;

[0062] A maintenance rate calculation module, used to calculate the maintenance rate under multiple indicators according to the maintenance data;

[0063] Model building module, used to build the ash cloud model and calculate the membership of different levels of various indicators;

[0064] The weight calculation module is used to calculate the weight matrix of each indicator based on the AHP hierarchical analysis method;

[0065] A coefficient calculation module, used to obtain a comprehensive clustering coefficient according to the different levels of membership of each indicator and the weight matrix;

[0066] A state calculation module, used to calculate the relative indicator state value according to the comprehensive clustering coefficient;

[0067] A comprehensive calculation module, used for determining a comprehensive correction rate by using a TOPSIS method according to the relative indicator state value;

[0068] An evaluation module is used to evaluate the performance of the charging pile based on the comprehensive correction rate.

[0069] The charging pile performance evaluation method and device based on the cloud platform provided by the present invention have at least the following beneficial effects:

[0070] It can reduce the accidental errors caused by detection data, and at the same time take into account the maintenance conditions of all detected charging piles with the same charging pile brand, charging pile type, charging pile area type, average service life of the charging pile, average daily utilization rate of the charging pile, spring temperature range, summer temperature range, autumn temperature range and winter temperature range characteristics as the evaluation object, so the evaluation credibility is higher. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 A step diagram of an embodiment of a charging pile performance evaluation method based on a cloud platform provided by the present invention.

[0072] Figure 2 A schematic structural diagram of an embodiment of a multi-level analysis model in a charging pile performance evaluation method based on a cloud platform provided by the present invention.

[0073] Figure 3 A schematic structural diagram of an embodiment of a charging pile performance evaluation device based on a cloud platform provided by the present invention. DETAILED DESCRIPTION

[0074] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0075] refer to Figure 1 In some embodiments, a charging pile performance evaluation method based on a cloud platform is provided, comprising:

[0076] S1. Obtain charging pile maintenance data through the cloud platform;

[0077] S2. Calculating maintenance rates under multiple indicators based on the maintenance data;

[0078] S3, establish a gray cloud model and calculate the membership of different levels of each indicator;

[0079] S4, calculate the weight matrix of each indicator based on AHP;

[0080] S5. Obtaining a comprehensive clustering coefficient according to the different levels of membership of each indicator and the weight value;

[0081] S6. Calculating the relative indicator status value according to the comprehensive clustering coefficient;

[0082] S7. Determine the comprehensive correction rate using the TOPSIS method according to the relative indicator status value;

[0083] S8. Evaluate the performance of the charging pile based on the comprehensive maintenance rate.

[0084] Specifically, in step S1, various data of the charging pile are collected through the detection device, including DC current output error, DC voltage output error, charging voltage regulation accuracy, charging current regulation accuracy, ripple factor, voltage limiting characteristics, current limiting characteristics, current imbalance, mean time between failures, charger life, charging cost, charging efficiency, power factor, harmonic pollution, insulation performance, protection level, etc. These data are stored in the cloud platform, which also stores data on all maintenance charging piles of power grid staff.

[0085] Furthermore, in step S2, the maintenance rate under each indicator is calculated by the following formula:

[0086]

[0087] Among them, x i is the charging pile maintenance rate under indicator i, Apl i is the number of charging pile maintenance under index i, ∑Apl ij The number of repairs for all charging stations.

[0088] Furthermore, the indicators include charging pile brand, charging pile type, type of area where the charging pile is located, average service life of the charging pile, average daily utilization rate of the charging pile, spring temperature range, summer temperature range, autumn temperature range and winter temperature range.

[0089] Specifically, for charging pile brands, several major brands with high market share are mainly considered. Four brands are given as examples, but not limited to them. The charging pile brands are as follows: Jiteladian is A1, State Grid is A2, Xingxing Charging is A3, and others are A4. As of now, the total number of repairs of the corresponding brand of charging piles in the country divided by the historical number of repairs of all charging piles is defined as the supplier repair rate corresponding to the charging pile x1;

[0090] According to the type of charging pile, AC charging pile is recorded as B1, DC charging pile is recorded as B2, and AC / DC charging pile is recorded as B3. The total number of repairs of the corresponding model of the charging pile divided by the historical number of repairs of all charging piles is defined as the repair rate of the model corresponding to the charging pile x2;

[0091] The type of region where the charging pile is located is recorded as C1 for arid areas, C2 for semi-arid areas, C3 for humid areas, and C4 for semi-humid areas. The total number of repairs in the region corresponding to the charging pile divided by the historical number of repairs of all charging piles is defined as the maintenance rate of the region corresponding to the charging pile x3;

[0092] Considering that the average service life of charging piles is 5 to 15 years, the service life of charging piles is recorded as 0 to 3 years as D1, 3 to 8 years as D2, 8 to 12 years as D3, and 12 to 15 years as D4. The total number of repairs in the country within the corresponding service life of charging piles divided by the historical number of repairs of all charging piles is defined as the corresponding service time maintenance rate of the charging pile x4;

[0093] The daily average utilization rate of charging piles is recorded as 0-25% as E1, 25%-50% as E2, 50%-75% as E3, and 75%-100% as E4. The number of all repairs in the country within the range of the daily average utilization rate of charging piles up to now divided by the number of historical repairs of all charging piles is defined as the daily average utilization rate maintenance rate corresponding to the charging pile x5;

[0094] Considering the actual temperature across the country in spring, the spring average temperature is recorded as -10℃~0℃ as F1, 0℃~10℃ as F2, 10℃~20℃ as F3, and 20℃~30℃ as F4. Up to now, the number of all repairs of charging piles in the country within the spring average temperature range divided by the historical repair number of all charging piles is defined as the spring average temperature maintenance rate corresponding to the charging pile x6;

[0095] Considering the actual temperature across the country in summer, the summer average temperature is recorded as 10℃~20℃ as G1, 20℃~30℃ as G2, and 30℃~40℃ as G3. As of now, the number of all repairs in the country within the summer average temperature range of the charging piles divided by the historical number of repairs of all charging piles is defined as the summer average temperature repair rate corresponding to the charging pile x7;

[0096] Considering the actual temperature across the country in autumn, the autumn average temperature is recorded as -10℃~0℃ as H1, 0℃~10℃ as H2, 10℃~20℃ as H3, and 20℃~30℃ as H4. As of now, the number of all repairs of charging piles in the country within the autumn average temperature range divided by the historical repair number of all charging piles is defined as the autumn average temperature maintenance rate corresponding to the charging pile x8;

[0097] Considering the actual temperature across the country in winter, the average winter temperature is -25℃ ~ -15℃ as I1, -15℃ ~ -5℃ as I2, -5℃ ~ 5℃ as I3, 5℃ ~ 15℃ as I4, and 15℃ ~ 25℃ as I5. As of now, the number of all repairs in the country within the winter average temperature range corresponding to the charging pile divided by the historical number of repairs of all charging piles is defined as the winter average temperature maintenance rate corresponding to the charging pile x9.

[0098] Furthermore, in step S3, a gray cloud model is established to calculate the membership of different levels of various indicators, including:

[0099] S31. Establish maintenance rate levels;

[0100] Taking the above nine indicators as an example, the grade table is shown in Table 1:

[0101]

[0102]

[0103] Table 1

[0104] S32, using the maintenance rate level as a gray class of a gray cloud model, and determining a left boundary and a right boundary of each gray class according to an interval value of the maintenance rate level;

[0105] In this embodiment, the maintenance rate levels are as shown in Table 1, including four levels: good, normal, caution, and dangerous, that is, four gray classes, and the range of the value interval of each level is the left boundary and the right boundary.

[0106] S33, generating a normal random number according to the left boundary and the right boundary;

[0107] Specifically, the normal random number En* takes En as the expectation and He as the standard deviation, and the specific calculation formula is as follows:

[0108]

[0109]

[0110] Rx is the right boundary, Lx is the left boundary, En is the expectation, l and b are coefficients, l is 3 in the white weighted function of the upper limit measure and the lower limit measure, l is 6 in the white weighted function of the moderate measure, and b takes the empirical value of 10.

[0111] S34, determining a white weighting function for each gray class according to the normal random number and the maintenance rate under the index;

[0112] The white weighted function of the maintenance rate under the indicator with respect to the moderate measure of the gray class k is as follows:

[0113]

[0114] The white weighted function of the upper limit measure of the maintenance rate under the indicator with respect to the gray class k is as follows:

[0115]

[0116] The normal gray cloud model white weighted function of the maintenance rate under the indicator with respect to the lower limit measure of gray class k is as follows:

[0117]

[0118] Among them, xi is the maintenance rate under index i, Ex is the expected value, E x =0.5*(R x +L x ), Rx is the right boundary, Lx is the left boundary, is a normal random number about index i, is the white weighting function;

[0119] Taking four gray categories as an example, k=1 is good, k=2 is normal, k=3 is caution, k=4 is dangerous, moderate measurement, then k=2 and 3, upper limit measurement, then k=4, lower limit measurement, then k=1.

[0120] S35, calculating a white weighted value according to the white weighted function;

[0121] In the above embodiment, each maintenance rate xi needs to be calculated by the white weighting function four times, which are k=1, k=2, k=3, and k=4. Since the white weighting function is a function of normal random numbers, the white weighting value calculated by the white weighting function each time is different. In order to improve the randomness, the average value of the calculation results of multiple white weighting functions is used as the white weighting value of each gray class. For example, the calculation is performed 100 times and the average value is taken:

[0122]

[0123] in, is the white weighted value of index i with respect to gray class k, is the white weighting function.

[0124] S36, calculating the degree of membership according to the white weighted value;

[0125] The membership degree is calculated by the following formula:

[0126]

[0127] in, is the membership degree of index i to gray class k, is the white weighted value of index i with respect to gray class k, and n is the number of gray classes.

[0128] A maintenance rate xi is converted into four white weighted values ​​through the white weighting function, which respectively represent the weights of the indicator good, normal, caution, and dangerous, and the degree of membership is the relative weight.

[0129] Furthermore, in step S4, the weight matrix of each indicator is calculated based on the AHP hierarchical analysis method, including:

[0130] S41. Set the comprehensive maintenance rate as the target layer, the maintenance rate of each indicator as the criterion layer, and establish a multi-level analysis model;

[0131] Taking the above 9 indicators as an example, the established multi-level analysis model is as follows: Figure 2 shown.

[0132] S42, establishing a judgment matrix, and comparing the relative importance between two elements in the multi-level analysis model by constructing the judgment matrix;

[0133] Taking the above nine indicators as an example, the relative importance between the two elements is shown in Table 2, and their meanings are shown in Table 3:

[0134]

[0135] Table 2

[0136]

[0137] Table 3

[0138] S43, calculating the maximum eigenvalue and eigenvector of the judgment matrix to obtain a weight matrix.

[0139] Judgment matrix M = (x mn ) b×b , x mn Indicates the relative importance of one factor compared to another factor. The judgment matrix M has a vector W and a number λ such that M*W=λ*W. λ is called the eigenvalue of the matrix, and W is called the eigenvector corresponding to the eigenvalue λ. The number of eigenvalues ​​of a matrix is ​​the same as the value of the matrix, λ max is the largest of all eigenvalues, the weight matrix W between indicators i is max The eigenvalues ​​for the eigenvectors of .

[0140] In addition, it is necessary to perform consistency check on the judgment matrix to verify whether the judgment matrix meets the consistency requirements, so as to ensure the objectivity of the results. The consistency index value is calculated according to the following formula;

[0141]

[0142] Where CI is the consistency index, λmax is the maximum eigenvalue of the judgment matrix, and b is the order of the judgment matrix.

[0143] According to the consistency index CI, the consistency rate CR is calculated by combining the following formula:

[0144]

[0145] Where CR is the consistency rate; CI is the consistency index; RI is the degree of freedom index, and the value of RI is shown in Table 2. Generally speaking, only when CR < 0.1, the judgment matrix is ​​considered to pass the consistency test.

[0146] Furthermore, in step S5, the comprehensive clustering coefficient is calculated by the following formula:

[0147]

[0148] Among them, σ k is the comprehensive clustering coefficient, m is the number of indicators, W i is the weight matrix, is the membership degree of index i to gray class k.

[0149] Furthermore, in step S6, the relative indicator status value is calculated by the following formula:

[0150]

[0151] Among them, f is the relative indicator state value, k represents the gray class, σ k is the comprehensive clustering coefficient, and Exk represents the expectation of gray class k.

[0152] Furthermore, in step S7, the comprehensive correction rate is determined using the TOPSIS method, including:

[0153] S71. Use TOPSIS method to determine positive ideal solution and negative ideal solution;

[0154] S72. Calculate the positive distance and negative distance of each index from the positive ideal solution and the negative ideal solution:

[0155]

[0156] Among them, f + is a positive ideal solution, f - is a negative ideal solution, S + is the forward distance, S - is the negative distance, and f is the relative indicator state value;

[0157] f + =[0,0,0,0,0,0,0,0,0],f -=[9,9,9,9,9,9,9,9,9], positive ideal solution f + Represents all evaluation indicators x i All of them belong to the relative index state value when k=1 state is completely good, and the positive ideal solution f - Represents all evaluation indicators x i They are all relative indicator status values ​​when they are in a completely dangerous state.

[0158] S73. Calculate the comprehensive correction rate according to the positive distance and the negative distance:

[0159]

[0160] Among them, η is the comprehensive correction rate.

[0161] The comprehensive correction rate η indicates how close the evaluation object is to the worst state value. The smaller η is, the better the charging pile state is.

[0162] In addition, a comprehensive evaluation can be performed based on the evaluation score obtained by the charging pile evaluation method using local detection data and the comprehensive correction rate. The comprehensive evaluation formula is (1-η)*Y, where Y refers to the evaluation score obtained by the traditional charging pile evaluation method using local detection data, and Y= 100*(1-y i )*W i ,y i Indicates the relative degradation of the index, W i represents the weight between them, y it represents the detection value of indicator i, y iw Indicates the worst value of the charging pile indicator i, y ib Indicates the optimal value of charging pile index i, 100 means converted to percentage. i Including the insulation capacity and opening and closing capacity of the charging pile.

[0163] The method provided by the above embodiment can reduce the accidental errors generated by the detection data, and at the same time take into account the maintenance conditions of all detected charging piles with the same charging pile brand, charging pile type, type of area where the charging pile is located, average service life of the charging pile, average daily utilization rate of the charging pile, spring temperature range, summer temperature range, autumn temperature range and winter temperature range characteristics as the evaluation object, so that the evaluation credibility is higher.

[0164] In some embodiments, reference Figure 3 , a charging pile performance evaluation device based on a cloud platform is provided, comprising:

[0165] An acquisition module 201 is used to acquire charging pile maintenance data through a cloud platform;

[0166] A maintenance rate calculation module 202, used to calculate maintenance rates under multiple indicators based on the maintenance data;

[0167] The model building module 203 is used to build a gray cloud model and calculate the membership of different levels of various indicators;

[0168] The weight calculation module 204 is used to calculate the weight matrix of each indicator based on the AHP hierarchy analysis method;

[0169] The coefficient calculation module 205 is used to obtain the comprehensive clustering coefficient according to the membership degree of different levels of each indicator and the weight value;

[0170] A state calculation module 206, used to calculate a relative indicator state value according to the comprehensive clustering coefficient;

[0171] A comprehensive calculation module 207, used to determine the comprehensive maintenance rate using the TOPSIS method according to the relative indicator status value;

[0172] The evaluation module 208 is used to evaluate the performance of the charging pile based on the comprehensive maintenance rate.

[0173] Specifically, the maintenance rate calculation module 202 calculates the maintenance rate under each indicator by using formula (1).

[0174] The indicators include charging pile brand, charging pile type, type of area where the charging pile is located, average service life of the charging pile, average daily utilization rate of the charging pile, spring temperature range, summer temperature range, autumn temperature range and winter temperature range.

[0175] Furthermore, the model building module 203 is used to:

[0176] Establish maintenance rate levels;

[0177] The maintenance rate level is used as a gray class of a gray cloud model, and the left boundary and the right boundary of each gray class are determined according to the interval value of the maintenance rate level;

[0178] Generate a normal random number according to the left boundary and the right boundary;

[0179] Determine a white weighting function for each gray class according to the normal random number and the maintenance rate under the index;

[0180] Calculating a white weighted value according to the white weighted function;

[0181] The degree of membership is calculated according to the white weighted value.

[0182] The normal random number En* takes En as the expectation and He as the standard deviation, and its calculation formula is formula (2) and formula (3).

[0183] The white weighted function of the maintenance rate under the indicator with respect to the moderate measure of the gray class k is formula (4).

[0184] The white weighted function of the upper limit measure of the maintenance rate under the indicator with respect to the gray class k is shown in formula (5).

[0185] The white weighted function of the normal grey cloud model of the maintenance rate under the indicator with respect to the lower limit measure of the grey class k is shown in formula (6).

[0186] The white weighted value under each gray class is the average value of the calculation results of multiple white weighted functions;

[0187] The membership degree is calculated by formula (8).

[0188] The weight calculation module 204 is also used for:

[0189] Set the comprehensive maintenance rate as the target layer, and the maintenance rate of each indicator as the criterion layer, and establish a multi-level analysis model;

[0190] Establishing a judgment matrix, by constructing the judgment matrix to compare the relative importance between two elements in the multi-level analysis model;

[0191] The maximum eigenvalue and eigenvector of the judgment matrix are calculated to obtain a weight matrix.

[0192] The comprehensive clustering coefficient is calculated by formula (11).

[0193] The relative indicator status value is calculated by formula (12).

[0194] The comprehensive calculation module 207 is also used for:

[0195] The TOPSIS method is used to determine the positive ideal solution and the negative ideal solution;

[0196] Calculate the positive and negative distances of each index from the positive ideal solution and the negative ideal solution:

[0197] The comprehensive correction rate is calculated according to the positive distance and the negative distance.

[0198] Although preferred embodiments of the present invention have been described, additional changes and modifications may be made to these embodiments by those skilled in the art once the basic inventive concepts are known. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention. Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A charging pile performance evaluation method based on a cloud platform, characterized in that: include: Obtain charging pile maintenance data through the cloud platform; Calculate the maintenance rate under each indicator according to the maintenance data; Establish the gray cloud model and calculate the membership of different levels of each indicator; Calculate the weight matrix of each index based on AHP; According to the different levels of membership of each indicator and the weight matrix, a comprehensive clustering coefficient is obtained; Calculating a relative indicator status value according to the comprehensive clustering coefficient; According to the relative indicator status value, the comprehensive correction rate is determined by using the TOPSIS method; Performing charging pile performance evaluation based on the comprehensive correction rate; The maintenance rate under each indicator is calculated by the following formula: Among them, x i is the charging pile maintenance rate under indicator i, Apl ij is the number of maintenance of type j charging piles under index i, ∑Apl ij The number of repairs for all charging stations; The indicators include charging pile brand, charging pile type, type of area where the charging pile is located, average service life of the charging pile, average daily utilization rate of the charging pile, spring temperature range, summer temperature range, autumn temperature range and winter temperature range.

2. The method according to claim 1, characterized in that Establish the ash cloud model and calculate the membership of different levels of various indicators, including: Establish maintenance rate levels; The maintenance rate level is used as a gray class of a gray cloud model, and the left boundary and the right boundary of each gray class are determined according to the interval value of the maintenance rate level; Generate a normal random number according to the left boundary and the right boundary; Determine a white weighting function for each gray class according to the normal random number and the maintenance rate under the index; Calculating a white weighted value according to the white weighted function; The degree of membership is calculated according to the white weighted value.

3. The method according to claim 2, characterized in that The normal random number En* takes En as the expectation and He as the standard deviation: Rx is the right boundary, Lx is the left boundary, En is the expectation, l and b are coefficients; The white weighted function of the maintenance rate under the indicator with respect to the moderate measure of the gray class k is as follows: The white weighted function of the upper limit measure of the maintenance rate under the indicator with respect to the gray class k is as follows: The normal gray cloud model white weighted function of the maintenance rate under the indicator with respect to the lower limit measure of gray class k is as follows: Among them, xi is the maintenance rate under index i, Ex is the expected value, E x =0.5*(R x +L x ), Rx is the right boundary, Lx is the left boundary, is a normal random number about index i, is the white weighting function; The white weighted value under each gray class is the average value of the calculation results of multiple white weighted functions; The membership degree is calculated by the following formula: in, is the membership degree of index i to gray class k, is the white weighted value of index i with respect to gray class k, and n is the number of gray classes.

4. The method according to claim 3, characterized in that The weight matrix of each indicator is calculated based on the AHP hierarchical analysis method, including: Set the comprehensive maintenance rate as the target layer, and the maintenance rate of each indicator as the criterion layer, and establish a multi-level analysis model; Establishing a judgment matrix, by constructing the judgment matrix to compare the relative importance between two elements in the multi-level analysis model; The maximum eigenvalue and eigenvector of the judgment matrix are calculated to obtain a weight matrix.

5. The method according to claim 4, characterized in that The comprehensive clustering coefficient is calculated by the following formula: Among them, σ k is the comprehensive clustering coefficient, m is the number of indicators, W i is the weight matrix, is the membership degree of index i to gray class k.

6. The method according to claim 5, characterized in that The relative indicator status value is calculated by the following formula: Among them, f is the relative indicator state value, k represents the gray class, σ k is the comprehensive clustering coefficient, Ex k Represents the expectation of gray class k.

7. The method according to claim 6, characterized in that The comprehensive correction rate is determined using the TOPSIS method, including: The TOPSIS method is used to determine the positive ideal solution and the negative ideal solution; Calculate the positive and negative distances of each index from the positive ideal solution and the negative ideal solution: Among them, f + is a positive ideal solution, f - is a negative ideal solution, S + is the forward distance, S - is the negative distance, and f is the relative indicator state value; According to the positive distance and the negative distance, the comprehensive correction rate is calculated: Among them, η is the comprehensive correction rate.

8. A charging pile performance evaluation device based on a cloud platform, characterized in that: include: A data acquisition module is used to obtain charging pile maintenance data through the cloud platform; A maintenance rate calculation module, used to calculate the maintenance rate under multiple indicators according to the maintenance data; Model building module, used to build the ash cloud model and calculate the membership of different levels of various indicators; The weight calculation module is used to calculate the weight matrix of each indicator based on the AHP hierarchical analysis method; A coefficient calculation module, used to obtain a comprehensive clustering coefficient according to the different levels of membership of each indicator and the weight matrix; A state calculation module, used to calculate the relative indicator state value according to the comprehensive clustering coefficient; A comprehensive calculation module, used for determining a comprehensive correction rate by using a TOPSIS method according to the relative indicator state value; An evaluation module, used for evaluating the performance of the charging pile based on the comprehensive correction rate; The maintenance rate under each indicator is calculated by the following formula: Among them, x i is the charging pile maintenance rate under indicator i, Apl ij is the number of maintenance of type j charging piles under index i, ∑Apl ij The number of repairs for all charging stations; The indicators include charging pile brand, charging pile type, type of area where the charging pile is located, average service life of the charging pile, average daily utilization rate of the charging pile, spring temperature range, summer temperature range, autumn temperature range and winter temperature range.

Images