High-precision charging pile measurement data processing system and method based on artificial intelligence

Abstract

The present invention discloses a high-precision charging pile measurement data processing system and method based on artificial intelligence, which relates to the technical field of intelligent charging management, including analyzing the degree of equipment matching between equipment operating parameters in the marked equipment operating data and the charging pile; obtaining the historical environmental monitoring records of the charging pile, obtaining the abnormal equipment operating parameters from the abnormal equipment operating data, and analyzing the data correlation between the environmental monitoring indicators in the historical environmental monitoring records and the abnormal equipment operating parameters in the charging pile; obtaining the historical data measurement records of the charging pile, establishing a charging fault prediction model for the charging pile, obtaining the historical environmental monitoring records, obtaining the associated environmental data, and optimizing the charging fault prediction model; monitoring the charging status of the charging pile in the current cycle, and using the charging fault prediction model to predict faults for the charging pile, and manage the charging of the charging pile.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent charging management, and in particular to a high-precision charging pile measurement data processing system and method based on artificial intelligence. Background Art

[0002] Charging piles will generate a large amount of data during the operation of the equipment, including charging time, charging power, charging amount and other information. The use of artificial intelligence can quickly analyze and process a large amount of data and identify valuable information and patterns in the data, which far exceeds the traditional data processing capabilities. In addition, with the help of artificial intelligence, the charging data of charging piles can be deeply analyzed, which helps to optimize the charging strategy and maximize energy efficiency, thereby reducing user waiting time and improving user experience and satisfaction.

[0003] At present, most of the methods for analyzing the equipment status of charging piles mainly adopt the method of analyzing the equipment status of the charging piles based on the data corresponding to the equipment operating parameters of the charging piles collected, and judging whether the charging piles have the risk of equipment failure. However, there are many disadvantages. First, because the data accuracy of the measurement data of the charging piles is not enough, the measurement data of the equipment cannot be accurately obtained. Secondly, in the actual process, the environment in which the charging piles are used and the equipment conditions of the charging piles are different. These will have an impact on the charging piles, and the analysis results of the equipment status of the charging piles will be far from the actual situation of the charging piles themselves, which will not only affect the charging efficiency of users, but may even cause damage to the charging piles and the charging vehicles. Summary of the invention

[0004] The purpose of the present invention is to provide a high-precision charging pile measurement data processing system and method based on artificial intelligence to solve the problems raised in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solution: a high-precision charging pile measurement data processing method based on artificial intelligence, the method comprising:

[0006] Step S100: acquiring the fault time data of the charging pile, acquiring the historical data measurement record of the charging pile, acquiring the marked device operation data of the charging pile, analyzing the device operation parameters in the marked device operation data and the device matching degree between the charging pile, and obtaining the abnormal device operation data;

[0007] Step S200: Obtain historical environmental monitoring records of the charging pile, obtain abnormal equipment operation parameters from abnormal equipment operation data, analyze the data correlation between environmental monitoring indicators in the historical environmental monitoring records and abnormal equipment operation parameters in the charging pile, and obtain correlated environmental data;

[0008] Step S300: establishing a charging fault prediction model for a charging pile, acquiring associated environmental data, and optimizing the charging fault prediction model;

[0009] Step S400: monitor the charging status of the charging pile in the current cycle, use the charging fault prediction model to predict the fault of the charging pile, and manage the charging of the charging pile.

[0010] Furthermore, step S100 includes:

[0011] Step S101: acquiring the fault time data of the charging pile, acquiring several historical time periods when the charging pile fails from the fault time data, monitoring the device status of the charging pile in several historical time periods respectively, and obtaining several historical data measurement records;

[0012] Step S102: obtaining the average values ​​of various equipment operating parameters of the measured charging pile from the historical data measurement records;

[0013] Step S103: obtaining preset thresholds of various equipment operating parameters of the charging pile from the marked equipment operating data;

[0014] Step S104: Analyze the matching degree between each device operation parameter in the marked device operation data and the charging pile. The specific analysis process is as follows:

[0015] From a number of historical data measurement records, obtain device operating parameters whose average values ​​are all less than a threshold value, and record them as marked device operating parameters of the charging pile;

[0016] Obtain several marking equipment operating parameters of the charging pile, and calculate the equipment matching value between the charging pile and several marking equipment operating parameters, among which the equipment matching value B of the charging pile and the equipment operating parameter of the ath marking device is a :

[0017] ,

[0018] Among them, C a The threshold value of the equipment operating parameter marked for item a; D a,i is the average value of the a-th item marking equipment operating parameter in the i-th historical data measurement record among several historical data measurement records; j is the total number of several historical data measurement records;

[0019] Step S105: Setting the device matching threshold B´. a ≤B´, it is determined that there is no device match between the a-th item of the marked device operation parameter and the charging pile in the marked device operation data, and the a-th item of the marked device operation parameter is recorded as the abnormal device operation parameter of the charging pile;

[0020] Step S106: Acquire several abnormal equipment operation parameters of the charging pile, and collect them to obtain abnormal equipment operation data of the charging pile.

[0021] Further, step S200 includes:

[0022] Step S201: Obtain each historical environmental monitoring record of the charging pile, and obtain the average value of each environmental monitoring index in the charging pile from the historical environmental monitoring record;

[0023] Step S202: Calculate the mark value E=(F-μ) / σ of the environmental monitoring index in the charging pile in the historical environmental monitoring records, where F is the average value of the environmental monitoring index in the historical environmental monitoring records, μ is the mean of the average values ​​of the environmental monitoring index in each historical environmental monitoring record, and σ is the standard deviation of the environmental monitoring index;

[0024] Step S203: Establish the environment matrix G of the charging pile:

[0025] ,

[0026] Where n is the total number of historical environmental monitoring records; m is the total number of environmental monitoring indicators; E 11 E is the mark value of the first environmental monitoring indicator in the first historical environmental monitoring record of the charging pile; n1 is the mark value of the first environmental monitoring indicator in the nth historical environmental monitoring record of the charging pile; E 1m is the mark value of the mth environmental monitoring indicator in the first historical environmental monitoring record of the charging pile; E nm is the mark value of the mth environmental monitoring indicator in the nth historical environmental monitoring record of the charging pile;

[0027] Step S204: Calculate the covariance matrix H of the environment matrix G:

[0028] ,

[0029] Among them, G T is the transpose of the environment matrix;

[0030] The covariance matrix H is decomposed into eigenvalue λ and eigenvector v. The specific decomposition formula is: H·v=λ·v. The eigenvalues ​​corresponding to the environmental monitoring indicators of the charging pile are obtained through the decomposition formula. According to the numerical values ​​of the eigenvalues, the environmental monitoring indicators corresponding to the preset z eigenvalues ​​are selected and marked.

[0031] The projection matrix Q that constitutes the environment matrix G = [v 1 、v 2 ,...,v z ], where v1 、v 2 ,...,v z They are represented as the characteristic vectors of the labeled 1st, 2nd, ..., zth environmental monitoring indicators respectively;

[0032] Step S205: Obtain the main environment matrix Y=Q·G of the charging pile. Based on the main environment matrix, analyze the data correlation between each environmental monitoring indicator in the historical environmental monitoring record and the abnormal equipment operation parameters in the charging pile. The data correlation between each environmental monitoring indicator and the αth abnormal equipment operation parameter in the charging pile is analyzed. The specific process is as follows:

[0033] Obtain the mark value of the αth abnormal equipment operation parameter in each historical data measurement record of the charging pile, and aggregate them to obtain the matrix S α ;

[0034] Construct the correlation regression model S of the αth abnormal equipment operation parameter α :

[0035] ,

[0036] Among them, X 1 , X 2 , ..., X z Respectively, they are the marked 1st, 2nd, ..., zth environmental monitoring indicators; β 1 , β 2 , ..., β z They are respectively expressed as the regression coefficients of the 1st, 2nd, ..., zth environmental monitoring indicators; β 0 Expressed as the correlation regression model S α The intercept term of ; ε represents the error term of the associated regression model;

[0037] Get the correlation regression model S α The design matrix X = [v 0 、v 1 、v 2 ,...,v z ], where v 0 All n elements in are 1;

[0038] Get the correlation regression model S α The regression coefficient matrix β in is:

[0039] ,

[0040] Where β = [β 1 , β 2 , ..., β z ];

[0041] Using the association regression model, the predicted value of the αth abnormal equipment operating parameter in each historical data measurement record is obtained, and the difference between the marked value and the predicted value of the αth abnormal equipment operating parameter in each historical data measurement record is calculated to obtain the association regression model S α The error term ε is:

[0042] Step S206: Obtain the constructed correlation regression model S α , when the regression coefficient of a marked environmental monitoring indicator is greater than the preset first threshold β´ 1 , it is determined that a certain environmental monitoring indicator has a positive data correlation with the αth abnormal equipment operation parameter. When the regression coefficient of a certain environmental monitoring indicator is less than the preset second threshold β´ 2 , it is determined that a certain environmental monitoring indicator has a negative data correlation with the αth abnormal equipment operation parameter, where β´ 1 >0>β´ 2 ;

[0043] Step S207: Acquire several marked environmental monitoring indicators and aggregate them to obtain a correlation environment set of the αth abnormal equipment operation parameter, wherein the several environmental monitoring indicators have data correlation with the αth abnormal equipment operation parameter;

[0044] Obtain and aggregate the associated environment sets of several abnormal equipment operating parameters in the charging pile to obtain the associated environment data of the charging pile;

[0045] In the above steps, the main environment matrix is ​​used to analyze the data correlation between various environmental monitoring indicators in the historical environmental monitoring records and the abnormal equipment operating parameters in the charging pile. This is because in high-dimensional space, the data will become sparse, thereby increasing the complexity and computational cost of the associated regression model. The main environment matrix is ​​used to reduce the number of variables and avoid the problems caused by the "dimensionality disaster" when performing regression analysis. It not only reduces the consumption of computing resources and improves the computing speed, but also enhances the stability of the associated regression model.

[0046] Furthermore, step S300 includes:

[0047] Step S301: taking the average values ​​of various equipment operating parameters in the historical data measurement records as input data of the charging fault prediction model, taking the equipment failure probability of the charging pile as output data, and establishing the charging fault prediction model of the charging pile;

[0048] Step S302: Optimize the charging fault prediction model based on the average values ​​of various environmental monitoring indicators in the historical environmental monitoring records of the charging pile and in combination with the associated environmental data. The specific optimization process is as follows:

[0049] Obtain a test set of the charging fault prediction model, set a fault probability threshold, and determine that the charging pile is faulty when the equipment failure probability of the charging pile is greater than the failure probability threshold;

[0050] Using the test set, calculate the model accuracy R=(U △ +U) / (U △ +U+U´+U´ △ ), where U △ represents the number of charging piles correctly predicted by the charging fault prediction model, which is the total number of faulty samples, U represents the number of charging piles correctly predicted by the charging fault prediction model, which is the total number of normal samples, U´ △ represents the total number of samples where the charging fault prediction model incorrectly predicts the charging pile to be faulty; U´ represents the total number of samples where the charging fault prediction model incorrectly predicts the charging pile to be normal;

[0051] The various model parameters in the charging fault prediction model are optimized and adjusted until the model accuracy R is greater than the preset model accuracy threshold, and it is determined that the optimization of the charging fault prediction model is completed.

[0052] Furthermore, step S400 includes:

[0053] Step S401: obtaining measurement data and environmental monitoring data of the charging pile in the current cycle, wherein the measurement data includes: data corresponding to various equipment operating parameters of the charging pile, and the environmental monitoring data includes: data corresponding to various environmental monitoring indicators in the environment where the charging pile is located;

[0054] Step S402: Based on the measurement data and environmental monitoring data, a charging fault prediction model is used to predict charging faults of the charging pile, and intelligent management of charging of the charging pile is performed.

[0055] In order to better implement the above method, a high-precision charging pile measurement data processing system based on artificial intelligence is also proposed. The system includes a device matching analysis module, a correlation analysis module, a model optimization module, and an intelligent charging module;

[0056] The device matching analysis module is used to analyze the device operating parameters in the marked device operating data and the device matching degree between the charging piles to obtain abnormal device operating data;

[0057] The correlation analysis module is used to analyze the data correlation between the environmental monitoring indicators in the historical environmental monitoring records and the abnormal equipment operating parameters in the charging pile to obtain the correlated environmental data;

[0058] Model optimization module, used to obtain historical environmental monitoring records, obtain associated environmental data, and optimize the charging fault prediction model;

[0059] The intelligent charging module is used to monitor the charging status of the charging pile, and use the charging fault prediction model to predict the fault of the charging pile and manage the charging of the charging pile.

[0060] Further, the device matching analysis module includes a data acquisition unit and an abnormal device operation data unit;

[0061] A data acquisition unit, used to acquire the fault time data of the charging pile, acquire the historical data measurement record of the charging pile, and acquire the operation data of the marking device;

[0062] The abnormal device operation data unit is used to analyze various device operation parameters in the marked device operation data and the degree of device matching between the charging piles to obtain abnormal device operation data.

[0063] Further, the association analysis module includes a matrix construction unit and an association analysis unit;

[0064] A matrix construction unit, used to obtain the tag values ​​of the environmental monitoring indicators in the charging pile in the historical environmental monitoring records, and construct the main environmental matrix of the charging pile;

[0065] The correlation analysis unit is used to analyze the data correlation between various environmental monitoring indicators in the historical environmental monitoring records and the abnormal equipment operating parameters in the charging pile, so as to obtain the associated environmental data of the charging pile.

[0066] Furthermore, the model optimization module includes a model building unit and a model optimization unit;

[0067] A model building unit, used to build a charging failure prediction model for a charging pile;

[0068] The model optimization unit is used to optimize the charging fault prediction model based on the historical environmental monitoring records and associated environmental data of the charging pile.

[0069] Further, the intelligent charging module includes an intelligent charging unit;

[0070] The intelligent charging unit is used to predict charging failures of the charging piles based on the measurement data of the charging piles and the environmental monitoring data, using a charging failure prediction model, and to intelligently manage the charging of the charging piles.

[0071] Compared with the prior art, the beneficial effects of the present invention are: the present invention realizes intelligent processing of high-precision charging pile measurement data, analyzes abnormal equipment operating parameters according to the equipment conditions of the charging pile itself, and analyzes the correlation between environmental monitoring indicators and equipment operating parameters by constructing an associated regression model, thereby optimizing the constructed charging fault prediction model, which not only greatly improves the accuracy of charging pile fault prediction, but also reduces the probability of charging accidents, and further ensures the charging rate and safety of charging piles. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 It is a method flow chart of a high-precision charging pile measurement data processing method based on artificial intelligence of the present invention;

[0073] Figure 2 It is a module schematic diagram of the high-precision charging pile measurement data processing system based on artificial intelligence of the present invention. DETAILED DESCRIPTION

[0074] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0075] Example: Figure 1-Figure 2 As shown, the present invention provides a technical solution, a high-precision charging pile measurement data processing method based on artificial intelligence, the method comprising:

[0076] Step S100: acquiring the fault time data of the charging pile, acquiring the historical data measurement record of the charging pile, acquiring the marked device operation data of the charging pile, analyzing the device operation parameters in the marked device operation data and the device matching degree between the charging pile, and obtaining the abnormal device operation data;

[0077] Wherein, step S100 includes:

[0078] Step S101: acquiring the fault time data of the charging pile, acquiring several historical time periods when the charging pile fails from the fault time data, monitoring the device status of the charging pile in several historical time periods respectively, and obtaining several historical data measurement records;

[0079] Step S102: obtaining the average values ​​of various equipment operating parameters of the measured charging pile from the historical data measurement records;

[0080] For example, various equipment operating parameters include equipment voltage, equipment current, etc.;

[0081] Step S103: obtaining preset thresholds of various equipment operating parameters of the charging pile from the marked equipment operating data;

[0082] Step S104: Analyze the matching degree between each device operation parameter in the marked device operation data and the charging pile. The specific analysis process is as follows:

[0083] From a number of historical data measurement records, obtain device operating parameters whose average values ​​are all less than a threshold value, and record them as marked device operating parameters of the charging pile;

[0084] Obtain several marking equipment operating parameters of the charging pile, and calculate the equipment matching value between the charging pile and several marking equipment operating parameters, among which the equipment matching value B of the charging pile and the equipment operating parameter of the ath marking device is a :

[0085] ,

[0086] Among them, C a The threshold value of the equipment operating parameter marked for item a; D a,i is the average value of the a-th item marking equipment operating parameter in the i-th historical data measurement record among several historical data measurement records; j is the total number of several historical data measurement records;

[0087] For example, the total number of historical data measurement records j is 3; the first item marks the threshold value C of the equipment operating parameter 1 =10; the average value D of the first item of the equipment operating parameter in the first historical data measurement record 1,1 =7; the average value D of the first item of the equipment operating parameter in the second historical data measurement record 1,2 =8; the average value D of the first item of the equipment operating parameter in the third historical data measurement record 1,3 is 6;

[0088] Calculate the device matching value B of the device operating parameters marked in item 1 1 :

[0089] ,

[0090] Step S105: Setting the device matching threshold B´. a ≤B´, it is determined that there is no device match between the a-th item of the marked device operation parameter and the charging pile in the marked device operation data, and the a-th item of the marked device operation parameter is recorded as the abnormal device operation parameter of the charging pile;

[0091] Step S106: Acquire several abnormal equipment operation parameters of the charging pile, and collect them to obtain abnormal equipment operation data of the charging pile;

[0092] Step S200: Obtain historical environmental monitoring records of the charging pile, obtain abnormal equipment operation parameters from abnormal equipment operation data, analyze the data correlation between environmental monitoring indicators in the historical environmental monitoring records and abnormal equipment operation parameters in the charging pile, and obtain correlated environmental data;

[0093] Wherein, step S200 includes:

[0094] Step S201: Obtain each historical environmental monitoring record of the charging pile, and obtain the average value of each environmental monitoring index in the charging pile from the historical environmental monitoring record;

[0095] For example, various environmental monitoring indicators include ambient temperature, ambient humidity, etc.;

[0096] Step S202: Calculate the mark value E=(F-μ) / σ of the environmental monitoring index in the charging pile in the historical environmental monitoring records, where F is the average value of the environmental monitoring index in the historical environmental monitoring records, μ is the mean of the average values ​​of the environmental monitoring index in each historical environmental monitoring record, and σ is the standard deviation of the environmental monitoring index;

[0097] Step S203: Establish the environment matrix G of the charging pile:

[0098] ,

[0099] Where n is the total number of historical environmental monitoring records; m is the total number of environmental monitoring indicators; E 11 E is the mark value of the first environmental monitoring indicator in the first historical environmental monitoring record of the charging pile; n1 is the mark value of the first environmental monitoring indicator in the nth historical environmental monitoring record of the charging pile; E 1m is the mark value of the mth environmental monitoring indicator in the first historical environmental monitoring record of the charging pile; E nm is the mark value of the mth environmental monitoring indicator in the nth historical environmental monitoring record of the charging pile;

[0100] Step S204: Calculate the covariance matrix H of the environment matrix G:

[0101] ,

[0102] Among them, G T is the transpose of the environment matrix;

[0103] The covariance matrix H is decomposed into eigenvalue λ and eigenvector v. The specific decomposition formula is: H·v=λ·v. The eigenvalues ​​corresponding to the environmental monitoring indicators of the charging pile are obtained through the decomposition formula. According to the numerical values ​​of the eigenvalues, the environmental monitoring indicators corresponding to the preset z eigenvalues ​​are selected and marked.

[0104] The projection matrix Q that constitutes the environment matrix G = [v 1 、v 2 ,...,v z ], where v 1 、v 2 ,...,v z They are represented as the characteristic vectors of the labeled 1st, 2nd, ..., zth environmental monitoring indicators respectively;

[0105] Step S205: Obtain the main environment matrix Y=Q·G of the charging pile. Based on the main environment matrix, analyze the data correlation between each environmental monitoring indicator in the historical environmental monitoring record and the abnormal equipment operation parameters in the charging pile. The data correlation between each environmental monitoring indicator and the αth abnormal equipment operation parameter in the charging pile is analyzed. The specific process is as follows:

[0106] Obtain the mark value of the αth abnormal equipment operation parameter in each historical data measurement record of the charging pile, and aggregate them to obtain the matrix S α ;

[0107] Construct the correlation regression model S of the αth abnormal equipment operation parameter α :

[0108] ,

[0109] Among them, X 1 , X 2 , ..., X z Respectively, they are the marked 1st, 2nd, ..., zth environmental monitoring indicators; β 1 , β 2 , ..., β z They are respectively expressed as the regression coefficients of the 1st, 2nd, ..., zth environmental monitoring indicators; β 0 Expressed as the correlation regression model S α The intercept term of ; ε represents the error term of the associated regression model;

[0110] Get the correlation regression model S α The design matrix X = [v 0 、v 1 、v 2 ,...,v z ], where v 0 All n elements in are 1;

[0111] Get the correlation regression model S α The regression coefficient matrix β in is:

[0112] ,

[0113] Where β = [β 1 , β 2 , ..., β z ];

[0114] Using the association regression model, the predicted value of the αth abnormal equipment operating parameter in each historical data measurement record is obtained, and the difference between the marked value and the predicted value of the αth abnormal equipment operating parameter in each historical data measurement record is calculated to obtain the association regression model S α The error term ε is:

[0115] Step S206: Obtain the constructed correlation regression model S α , when the regression coefficient of a marked environmental monitoring indicator is greater than the preset first threshold β´ 1 , it is determined that a certain environmental monitoring indicator has a positive data correlation with the αth abnormal equipment operation parameter. When the regression coefficient of a certain environmental monitoring indicator is less than the preset second threshold β´ 2 , it is determined that a certain environmental monitoring indicator has a negative data correlation with the αth abnormal equipment operation parameter, where β´ 1 >0>β´ 2 ;

[0116] Step S207: Acquire several marked environmental monitoring indicators and aggregate them to obtain a correlation environment set of the αth abnormal equipment operation parameter, wherein the several environmental monitoring indicators have data correlation with the αth abnormal equipment operation parameter;

[0117] Obtain and aggregate the associated environment sets of several abnormal equipment operating parameters in the charging pile to obtain the associated environment data of the charging pile;

[0118] Step S300: establishing a charging fault prediction model for a charging pile, acquiring associated environmental data, and optimizing the charging fault prediction model;

[0119] Wherein, step S300 includes:

[0120] Step S301: taking the average values ​​of various equipment operating parameters in the historical data measurement records as input data of the charging fault prediction model, taking the equipment failure probability of the charging pile as output data, and establishing the charging fault prediction model of the charging pile;

[0121] Step S302: Optimize the charging fault prediction model based on the average values ​​of various environmental monitoring indicators in the historical environmental monitoring records of the charging pile and in combination with the associated environmental data. The specific optimization process is as follows:

[0122] Obtain a test set of the charging fault prediction model, set a fault probability threshold, and determine that the charging pile is faulty when the equipment failure probability of the charging pile is greater than the failure probability threshold;

[0123] Using the test set, calculate the model accuracy R=(U △ +U) / (U △ +U+U´+U´ △ ), where U △ represents the number of charging piles correctly predicted by the charging fault prediction model, which is the total number of faulty samples, U represents the number of charging piles correctly predicted by the charging fault prediction model, which is the total number of normal samples, U´ △ represents the total number of samples where the charging fault prediction model incorrectly predicts the charging pile to be faulty; U´ represents the total number of samples where the charging fault prediction model incorrectly predicts the charging pile to be normal;

[0124] Optimize and adjust various model parameters in the charging fault prediction model until the model accuracy R is greater than the preset model accuracy threshold, and determine that the charging fault prediction model optimization is completed;

[0125] Step S400: monitor the charging status of the charging pile in the current cycle, and use the charging fault prediction model to predict the fault of the charging pile and manage the charging of the charging pile;

[0126] Wherein, step S400 includes:

[0127] Step S401: obtaining measurement data and environmental monitoring data of the charging pile in the current cycle, wherein the measurement data includes: data corresponding to various equipment operating parameters of the charging pile, and the environmental monitoring data includes: data corresponding to various environmental monitoring indicators in the environment where the charging pile is located;

[0128] Step S402: Based on the measurement data and the environmental monitoring data, a charging fault prediction model is used to predict charging faults of the charging pile, and intelligent management of charging of the charging pile is performed;

[0129] In order to better implement the above method, a high-precision charging pile measurement data processing system based on artificial intelligence is also proposed. The system includes a device matching analysis module, a correlation analysis module, a model optimization module, and an intelligent charging module;

[0130] The device matching analysis module is used to analyze the device operating parameters in the marked device operating data and the device matching degree between the charging piles to obtain abnormal device operating data;

[0131] The correlation analysis module is used to analyze the data correlation between the environmental monitoring indicators in the historical environmental monitoring records and the abnormal equipment operating parameters in the charging pile to obtain the correlated environmental data;

[0132] Model optimization module, used to obtain historical environmental monitoring records, obtain associated environmental data, and optimize the charging fault prediction model;

[0133] Intelligent charging module, which is used to monitor the charging status of the charging pile, and use the charging fault prediction model to predict the fault of the charging pile and manage the charging of the charging pile;

[0134] Among them, the equipment matching analysis module includes a data acquisition unit and an abnormal equipment operation data unit;

[0135] A data acquisition unit, used to acquire the fault time data of the charging pile, acquire the historical data measurement record of the charging pile, and acquire the operation data of the marking device;

[0136] The abnormal equipment operation data unit is used to analyze the equipment operation parameters in the marked equipment operation data and the equipment matching degree between the charging piles to obtain the abnormal equipment operation data;

[0137] Wherein, the correlation analysis module includes a matrix construction unit and a correlation analysis unit;

[0138] A matrix construction unit, used to obtain the tag values ​​of the environmental monitoring indicators in the charging pile in the historical environmental monitoring records, and construct the main environmental matrix of the charging pile;

[0139] A correlation analysis unit is used to analyze the data correlation between various environmental monitoring indicators in the historical environmental monitoring records and the abnormal equipment operation parameters in the charging pile to obtain the associated environmental data of the charging pile;

[0140] Among them, the model optimization module includes a model building unit and a model optimization unit;

[0141] A model building unit, used to build a charging failure prediction model for a charging pile;

[0142] A model optimization unit, used to optimize the charging fault prediction model based on the historical environmental monitoring records and associated environmental data of the charging pile;

[0143] Wherein, the intelligent charging module includes an intelligent charging unit;

[0144] The intelligent charging unit is used to predict charging failures of the charging piles based on the measurement data of the charging piles and the environmental monitoring data, using a charging failure prediction model, and to intelligently manage the charging of the charging piles.

[0145] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A high-precision charging pile measurement data processing method based on artificial intelligence, characterized in that: The method comprises: Step S100: acquiring the fault time data of the charging pile, acquiring the historical data measurement record of the charging pile, acquiring the marked device operation data of the charging pile, analyzing the device operation parameters in the marked device operation data and the device matching degree between the charging pile, and obtaining abnormal device operation data; Step S200: obtaining the historical environmental monitoring records of the charging pile, obtaining abnormal equipment operation parameters from the abnormal equipment operation data, analyzing the data correlation between the environmental monitoring indicators in the historical environmental monitoring records and the abnormal equipment operation parameters in the charging pile, and obtaining the correlated environmental data; Step S300: establishing a charging fault prediction model for the charging pile, acquiring the associated environmental data, and optimizing the charging fault prediction model; Step S400: monitoring the charging status of the charging pile in the current cycle, and using the charging fault prediction model to predict faults of the charging pile, and managing the charging of the charging pile; The step S100 includes: Step S101: acquiring the fault time data of the charging pile, acquiring several historical time periods when the charging pile fails from the fault time data, and monitoring the device status of the charging pile in the several historical time periods respectively to obtain several historical data measurement records; Step S102: obtaining the average values ​​of various equipment operating parameters of the charging pile measured from the historical data measurement records; Step S103: obtaining preset thresholds of various equipment operating parameters of the charging pile from the marking equipment operating data; Step S104: Analyze the matching degree between each device operation parameter in the marking device operation data and the device between the charging pile. The specific analysis process is as follows: From the plurality of historical data measurement records, obtain device operating parameters whose average values ​​are all less than a threshold value, and record them as marked device operating parameters of the charging pile; Obtain several marking device operating parameters of the charging pile, and calculate the device matching value between the charging pile and the several marking device operating parameters, wherein the device matching value B of the charging pile and the ath marking device operating parameter is a : ， Among them, C a The threshold value of the operating parameter of the device marked in item a; D a,i is the average value of the a-th marking device operating parameter in the ith historical data measurement record among the plurality of historical data measurement records; j is the total number of the plurality of historical data measurement records; Step S105: Setting the device matching threshold B´. a ≤B´, it is determined that in the marking device operation data, there is no device match between the a-th marking device operation parameter and the charging pile, and the a-th marking device operation parameter is recorded as the abnormal device operation parameter of the charging pile; Step S106: Acquire several abnormal equipment operation parameters of the charging pile, and collect them to obtain abnormal equipment operation data of the charging pile.

2. The high-precision charging pile measurement data processing method based on artificial intelligence according to claim 1 is characterized in that: The step S200 includes: Step S201: Obtain each historical environmental monitoring record of the charging pile, and obtain the average value of each environmental monitoring index in the charging pile from the historical environmental monitoring record; Step S202: Calculate the mark value E=(F-μ) / σ of the environmental monitoring index in the charging pile in the historical environmental monitoring record, where F is the average value of the environmental monitoring index in the historical environmental monitoring record, μ is the mean of the average values ​​of the environmental monitoring index in each historical environmental monitoring record, and σ is the standard deviation of the environmental monitoring index; Step S203: Establishing the environment matrix G of the charging pile: ， Where n is the total number of the historical environmental monitoring records; m is the total number of the environmental monitoring indicators; E 11 is the mark value of the first environmental monitoring indicator in the first historical environmental monitoring record of the charging pile; E n1 is the mark value of the first environmental monitoring indicator in the nth historical environmental monitoring record of the charging pile; E 1m is the mark value of the mth environmental monitoring indicator in the first historical environmental monitoring record of the charging pile; E nm is the mark value of the mth environmental monitoring indicator in the nth historical environmental monitoring record of the charging pile; Step S204: Calculate the covariance matrix H of the environment matrix G: ， Among them, G T is the transpose of the environment matrix; Decomposing the covariance matrix H by eigenvalue λ and eigenvector v, the specific decomposition formula is: H·v=λ·v, obtaining the eigenvalues ​​corresponding to the environmental monitoring indicators of the charging pile through the decomposition formula, and selecting the environmental monitoring indicators corresponding to the preset z eigenvalues ​​according to the numerical values ​​of the eigenvalues, and marking them; The projection matrix Q = [v1, v2, ..., v z ], where v1, v2, ..., v z They are represented as the characteristic vectors of the labeled 1st, 2nd, ..., zth environmental monitoring indicators respectively; Step S205: Obtain the main environment matrix Y=Q·G of the charging pile, and analyze the data correlation between each environmental monitoring indicator in the historical environmental monitoring record and the abnormal equipment operation parameter in the charging pile based on the main environment matrix, wherein the data correlation between each environmental monitoring indicator and the αth abnormal equipment operation parameter in the charging pile is analyzed, and the specific process is as follows: Obtain the label value of the αth abnormal equipment operating parameter in each historical data measurement record of the charging pile, and collect them to obtain the matrix S α ; Construct the correlation regression model S of the αth abnormal equipment operation parameter α : ， Among them, X1, X2, ..., X z They are respectively represented by the marked 1st, 2nd, ..., zth environmental monitoring indicators; β1, β2, ..., β z They are respectively represented by the regression coefficients of the first, second, ..., zth environmental monitoring indicators; β0 is represented by the correlation regression model S α The intercept term of ; ε represents the error term of the association regression model; Get the associated regression model S α The design matrix X = [v0, v1, v2, ..., v z ], wherein the n elements in v0 are all 1; Get the associated regression model S α The regression coefficient matrix β in is: ， Where, β=[β1、β2、...、β z ]; The association regression model is used to obtain the predicted value of the αth abnormal equipment operating parameter in each historical data measurement record, and the difference between the marked value and the predicted value of the αth abnormal equipment operating parameter in each historical data measurement record is calculated to obtain the association regression model S α The error term ε is: Step S206: Obtain the constructed correlation regression model S α When the regression coefficient of a marked environmental monitoring indicator is greater than the preset first threshold β´1, it is determined that the environmental monitoring indicator has a positive data association with the αth abnormal equipment operation parameter; when the regression coefficient of the environmental monitoring indicator is less than the preset second threshold β´2, it is determined that the environmental monitoring indicator has a negative data association with the αth abnormal equipment operation parameter, wherein β´1>0>β´2; Step S207: Acquire several marked environmental monitoring indicators and aggregate them to obtain a correlation environment set of the αth abnormal equipment operation parameter, wherein the several environmental monitoring indicators have data correlation with the αth abnormal equipment operation parameter; A set of associated environments of several abnormal equipment operating parameters in the charging pile is obtained and aggregated to obtain associated environment data of the charging pile.

3. The high-precision charging pile measurement data processing method based on artificial intelligence according to claim 2 is characterized in that: The step S300 includes: Step S301: taking the average value of each device operating parameter in the historical data measurement record as input data of the charging fault prediction model, taking the device failure probability of the charging pile as output data, and establishing the charging fault prediction model of the charging pile; Step S302: Optimizing the charging fault prediction model according to the average values ​​of various environmental monitoring indicators in the historical environmental monitoring records of the charging pile and in combination with the associated environmental data. The specific optimization process is as follows: Obtaining a test set of the charging fault prediction model, setting a fault probability threshold, and determining that the charging pile is faulty when the equipment failure probability of the charging pile is greater than the failure probability threshold; Using the test set, calculate the model accuracy R of the charging fault prediction model = (U △ +U) / (U △ +U+U´+U´ △ ), where U △ represents the total number of samples where the charging fault prediction model correctly predicts the charging pile to be faulty, U represents the total number of samples where the charging fault prediction model correctly predicts the charging pile to be normal, U´ △ represents the total number of samples in which the charging fault prediction model incorrectly predicts that the charging pile is faulty; U´ represents the total number of samples in which the charging fault prediction model incorrectly predicts that the charging pile is normal; The various model parameters in the charging fault prediction model are optimized and adjusted until the model accuracy R is greater than a preset model accuracy threshold, and it is determined that the optimization of the charging fault prediction model is completed.

4. The high-precision charging pile measurement data processing method based on artificial intelligence according to claim 3 is characterized in that: The step S400 includes: Step S401: obtaining measurement data and environmental monitoring data of the charging pile in the current cycle, wherein the measurement data includes: data corresponding to various equipment operating parameters of the charging pile, and the environmental monitoring data includes: data corresponding to various environmental monitoring indicators in the environment where the charging pile is located; Step S402: Based on the measurement data and the environmental monitoring data, use the charging fault prediction model to predict charging faults for the charging pile, and intelligently manage charging of the charging pile.

5. A high-precision charging pile measurement data processing system based on artificial intelligence, used to execute the high-precision charging pile measurement data processing method based on artificial intelligence according to any one of claims 1 to 4, characterized in that: The system includes a device matching analysis module, a correlation analysis module, a model optimization module, and a smart charging module; The device matching analysis module is used to analyze the device operating parameters in the marked device operating data and the device matching degree between the charging piles to obtain abnormal device operating data; The correlation analysis module is used to analyze the data correlation between the environmental monitoring indicators in the historical environmental monitoring records and the abnormal equipment operating parameters in the charging pile to obtain the associated environmental data; The model optimization module is used to obtain the historical environmental monitoring records, obtain the associated environmental data, and optimize the charging fault prediction model; The intelligent charging module is used to monitor the charging status of the charging pile, and use the charging fault prediction model to predict the fault of the charging pile and manage the charging of the charging pile.

6. The high-precision charging pile measurement data processing system based on artificial intelligence according to claim 5 is characterized in that: The device matching analysis module includes a data acquisition unit and an abnormal device operation data unit; The data acquisition unit is used to acquire the fault time data of the charging pile, acquire the historical data measurement record of the charging pile, and acquire the operation data of the marking device; The abnormal device operation data unit is used to analyze various device operation parameters in the marked device operation data and the degree of device matching between the charging piles to obtain abnormal device operation data.

7. The high-precision charging pile measurement data processing system based on artificial intelligence according to claim 5 is characterized in that: The association analysis module includes a matrix construction unit and an association analysis unit; The matrix construction unit is used to obtain the mark value of the environmental monitoring index in the charging pile in the historical environmental monitoring record to construct the main environmental matrix of the charging pile; The correlation analysis unit is used to analyze the data correlation between the various environmental monitoring indicators in the historical environmental monitoring records and the abnormal equipment operating parameters in the charging pile to obtain the associated environmental data of the charging pile.

8. The high-precision charging pile measurement data processing system based on artificial intelligence according to claim 5 is characterized in that: The model optimization module includes a model building unit and a model optimization unit; The model building unit is used to build a charging fault prediction model for the charging pile; The model optimization unit is used to optimize the charging fault prediction model according to the historical environmental monitoring records of the charging pile and the associated environmental data.

9. The high-precision charging pile measurement data processing system based on artificial intelligence according to claim 5 is characterized in that: The intelligent charging module includes an intelligent charging unit; The intelligent charging unit is used to predict charging faults of the charging pile based on the measurement data and environmental monitoring data of the charging pile and use the charging fault prediction model to intelligently manage the charging of the charging pile.

Images