A calibration method for AC charging pile based on big data remote communication

Abstract

The present invention discloses an AC charging pile calibration method based on big data remote communication, which belongs to the technical field of charging pile calibration. The AC charging pile calibration method based on big data remote communication collects multiple working parameter values ​​during the vehicle charging process, and records the working parameter values ​​based on the time sequence to generate a charging data set; the data set is analyzed by an analysis module to obtain abnormal charging information; corresponding control operations are performed on the charging piles with abnormal charging information; the calibration system of the AC charging pile calibration method based on big data remote communication can monitor the operating status of the charging pile in real time through big data analysis and an isolated forest algorithm, and accurately detect abnormal conditions during the charging process. Through the calculation of the abnormal score, the system can quickly identify the working abnormality of the charging pile, and then automatically calibrate the abnormal charging pile through the remote control module, adjust parameters such as current, voltage or power, and ensure the continuous and stable operation of the system.

Description

Technical Field

[0001] The present invention relates to the technical field of charging pile calibration, and in particular to an AC charging pile calibration method based on big data remote communication. Background Art

[0002] With the popularity of electric vehicles (EVs), charging piles have developed rapidly as infrastructure around the world. However, charging piles may face a variety of technical problems in long-term operation, such as unstable current and voltage, power fluctuations, etc. These problems not only affect the charging efficiency, but may also cause equipment failure or safety hazards. Traditional charging pile maintenance methods are mostly regular inspections or manual adjustments, which have the problems of slow response speed and high maintenance costs. At the same time, the development of big data technology and remote communication technology has provided new solutions for the intelligent management and maintenance of charging piles. By performing real-time analysis of big data during the charging process, abnormal behaviors can be quickly identified, and the working parameters of the charging piles can be remotely calibrated, thereby improving the reliability and efficiency of the system. Therefore, the charging pile calibration system based on big data and remote communication has become an important technical means to improve the intelligence and automation level of charging infrastructure.

[0003] After consulting the relevant public technologies, the technical solution with publication number CN113406557A proposes a propagation-type remote calibration method for charging piles, which uses a distributed network to propagate standard quantity values ​​in the charging pile network so that multiple charging piles in the network can be calibrated with each other; the technical solution with publication number CN109870649B proposes a calibration system and method for an AC charging pile calibration device, which uses an independent AC quantity detection device to calibrate the charging electrical parameters and pricing of the charging pile to ensure the accuracy of the statistical pricing of the charging pile; the technical solution with publication number US20220227250A1 proposes a charging pile calibration system, which uses a certified, external calibration module to connect to the existing charging pile, and then identifies the calibration module through a standard certification process to perform the calibration operation of the charging pile.

[0004] The above technical solutions all propose technical solutions for calibrating charging piles. However, in view of the current application scale of configuring a large number of charging piles, the implementation efficiency of detection and calibration needs to be further improved. Therefore, more efficient application solutions need to be proposed. Summary of the invention

[0005] In view of the above problems existing in the prior art, the present invention is proposed.

[0006] To achieve the above object, the present invention provides the following technical solution: an AC charging pile calibration method based on big data remote communication, comprising:

[0007] Collect multiple operating parameter values ​​during the vehicle charging process, and record the operating parameter values ​​based on time sequence to generate a charging data set;

[0008] Analyze the data set through the analysis module to obtain abnormal charging information;

[0009] Perform corresponding control operations on charging piles that have abnormal charging information.

[0010] As a further solution of the present invention: multiple operating parameter values ​​during the vehicle charging process are acquired through a data acquisition module.

[0011] As a further solution of the present invention: the step of analyzing the data set by the analysis module to obtain abnormal charging information includes:

[0012] Extracting a charging data set, and performing data preprocessing on the big data in the charging data set;

[0013] Establish an anomaly analysis algorithm tree based on multiple isolation forest algorithms;

[0014] Calculate the expected path length of the policy tree for each data point;

[0015] Calculate the anomaly score for each data point in the charging dataset;

[0016] Determine the charging station to which the abnormal data point belongs.

[0017] As a further solution of the present invention: the charging data set includes generating a Class A data point by using the working parameter value of a charging pile in a complete charging process;

[0018] Each class A data point is split, and two or more class B data points are obtained after the split, and the two or more class B data points are added to the charging data set;

[0019] Class A data points include charging current, charging voltage, charging time, charging start and end time, battery BMS status, vehicle information, charging pile information, and environmental information when the charging pile is working;

[0020] Among them, vehicle information includes vehicle model, battery model used in the vehicle, battery production date and battery batch;

[0021] Among them, the charging pile information includes the charging pile number and geographical location;

[0022] Among them, the environmental information when the charging pile is working includes ambient temperature, humidity, air pressure and sunlight intensity.

[0023] As a further solution of the present invention: extracting a charging data set, and performing data preprocessing on the big data in the charging data set includes:

[0024] Extract the original charging data set of the charging pile to form an original data set;

[0025] Perform data cleaning on the original data set;

[0026] Normalize the data set and standardize each feature data;

[0027] Perform feature engineering on the standardized data set to extract features;

[0028] Each feature is taken as a separate dimension, and the data of each complete charging process is constructed into a multi-dimensional feature vector.

[0029] As a further solution of the present invention: the step of establishing an abnormal analysis algorithm tree based on multiple isolation forest algorithms includes:

[0030] Randomly select one of the features from the multiple feature elements contained in the feature vector;

[0031] Find the minimum and maximum values ​​of the selected feature in the data set, and randomly select a split point between the two to obtain two subsets;

[0032] Repeat the above steps until the subset is segmented into only one data point or the depth of the segmentation tree reaches the preset maximum depth.

[0033] As a further solution of the present invention: the formula for the expected value of the path length of the strategy tree for each data point is:

[0034]

[0035] Where E[h(m)] represents the expected path length of data point m, h(m) represents the path length in each individual strategy tree, and n represents the number of data points in the dataset X.

[0036] As a further solution of the present invention: the anomaly score formula for each data point in the charging data set is:

[0037]

[0038] Among them, S(m,n) represents the anomaly score, μ(n) represents the normalization constant, and h(m) represents the average path length when the number of samples n is given, which is used to standardize the path length of sample m;

[0039] Among them, the normalization constant μ(n) formula is:

[0040]

[0041] Here, H() represents the harmonic number function, and it is approximately considered that H(n)≈ln(n)+r, where r is the Euler constant.

[0042] As a further solution of the present invention: the data points are classified as normal or abnormal according to their abnormal scores, that is:

[0043]

[0044] In the above formula, ψ1 and ψ2 are abnormal thresholds.

[0045] As a further solution of the present invention: a remote high-precision current detection system for an AC / DC charging pile, comprising a current detection module, an error calibration module, a safety protection module and a user management module of a use method;

[0046] The current detection module is connected to the safety protection module through the error calibration module;

[0047] The security protection module is connected with the user management module.

[0048] Compared with the prior art, the beneficial effect of the present invention is that the calibration method of AC charging pile based on big data remote communication, the calibration system can monitor the operating status of the charging pile in real time through big data analysis and isolation forest algorithm, and accurately detect abnormal conditions in the charging process. Through the calculation of abnormal scores, the system can quickly identify the working abnormality of the charging pile, and then automatically calibrate the abnormal charging pile through the remote control module, adjust the current, voltage or power and other parameters, and ensure the continuous and stable operation of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0050] Figure 1 It is a schematic diagram of the overall process in the embodiment provided by the present invention.

[0051] Figure 2 A schematic flow chart of the steps of analyzing a data set by an analysis module to obtain abnormal charging information in an embodiment provided by the present invention.

[0052] Figure 3 The present invention provides a flowchart of extracting a charging data set and performing a data preprocessing step on the big data in the charging data set in an embodiment.

[0053] Figure 4A schematic diagram of a flow chart of establishing an anomaly analysis algorithm tree based on multiple isolation forest algorithms in an embodiment provided by the present invention.

[0054] Figure 5 This is a comparison diagram of the output voltage error rate after the calibration system is used in the embodiments provided by the present invention.

[0055] Figure 6 A schematic diagram of the architecture of a computer system used in an embodiment of the present invention. DETAILED DESCRIPTION

[0056] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings.

[0057] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0058] Secondly, the present invention is described in detail with reference to the schematic diagram. When describing the embodiments of the present invention in detail, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0059] Furthermore, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0060] Example 1

[0061] like Figure 1 and 2 As shown, the present invention provides a technical solution: an AC charging pile calibration method based on big data remote communication, comprising the steps of:

[0062] S1: Collect multiple working parameter values ​​during the vehicle charging process, and record the working parameter values ​​based on time sequence to generate a charging data set;

[0063] Among them, the values ​​of multiple working parameters during the vehicle charging process are acquired through the data acquisition module. For real-time data such as voltage, current, etc., the data acquisition module is configured in the charging circuit inside and / or outside each charging pile to collect the working parameter values ​​of the charging pile in real time. The specific data acquisition module is integrated into the circuit of the charging pile and can be accessed through on-board access or through a universal data port.

[0064] S2: Analyze the data set through the analysis module to obtain abnormal charging information;

[0065] It should be noted that the analysis module is configured to perform big data analysis on charging data sets, detect abnormal charging records, and trace the charging piles with abnormal charging records. The analysis module is deployed in a remote control system or in a charging station.

[0066] S3: Perform corresponding control operations on the charging pile that has abnormal charging information.

[0067] It should be noted that the control module is configured to perform abnormal response control operations on charging piles with abnormal charging records.

[0068] Further, the step of analyzing the data set by the analysis module to obtain abnormal charging information includes:

[0069] S21: extracting a charging data set, and performing data preprocessing on the big data in the charging data set;

[0070] S22: Establishing anomaly analysis algorithm tree based on multiple isolation forest algorithms;

[0071] S23: Calculate the expected value of the path length of the strategy tree for each data point;

[0072] S24: Calculate the anomaly score of each data point in the charging data set;

[0073] S25: Determine the charging pile to which the abnormal data point belongs.

[0074] Specifically, the charging data set includes generating a Class A data point from the working parameter values ​​of a charging pile in a complete charging process; each Class A data point is segmented to obtain two or more Class B data points after segmentation, and the two or more Class B data points are added to the charging data set; the Class A data points include charging current, charging voltage, charging time, charging start and end time, battery BMS status, vehicle information, charging pile information and environmental information when the charging pile is working; wherein, the vehicle information includes the vehicle model, the battery model used in the vehicle, the battery production date and the battery batch; wherein, the charging pile information includes the charging pile number and the geographical location; wherein, the environmental information when the charging pile is working includes the ambient temperature, humidity, air pressure and sunshine intensity.

[0075] It should be noted that the A-level data points can be segmented by time period or charging stage (such as initial stage, stable stage, and end stage) to more finely analyze the characteristic behaviors of different stages; this step further refines the data and can more accurately detect local anomalies. The A-level data points are the average values ​​in a single charging process, and after further segmenting the data points, the characteristic performance of the charging working parameters in each time period during the charging stage can be more accurately understood, reflecting more detection details.

[0076] It should be noted that the control module includes counting the number of abnormal data points accumulated by each charging pile within a preset statistical period. When the number of abnormal data points accumulated by a charging pile exceeds a preset threshold in the preset statistical period, an abnormal response control operation is performed on the charging pile.

[0077] Preferably, the control module includes counting the abnormality score accumulated by each charging pile within a preset statistical period; when the accumulated abnormality score of a charging pile exceeds a preset threshold within the preset statistical period, executing an abnormal response control operation on the charging pile.

[0078] Preferably, the abnormal response control operation includes two stages:

[0079] Phase 1: Intervention to adjust the charging pile working parameters to observe whether the abnormal situation is improved or eliminated;

[0080] The second stage: stop the charging pile work.

[0081] It should be noted that the operating parameters can be changed according to the charging logic of the charging pile, such as reducing the voltage when the voltage is too high.

[0082] It should be noted that if the safety threshold is obviously exceeded, the charging pile will be stopped; the safety threshold can be set by the charging pile manufacturer based on its products.

[0083] In a preferred embodiment, the data acquisition module may include a measurement circuit and a plurality of sensors; the data acquisition module may be independently configured for each charging pile; or, in a large-scale charging station, a data acquisition module may be configured in the main power transmission device of the charging station and data may be collected from a plurality of charging piles allocated to the main power transmission device.

[0084] It should be noted that the measurement circuit is an integrated circuit composed of a single-chip circuit or a microprocessor circuit. Existing commercial chips can be directly used. Through the program written into the firmware, the sensor is periodically accessed to read the current value of the sensor. This is a basic circuit application and is not a restrictive requirement of this technical solution. Chip models such as Renergy RN8209D, HLW8032, and ATT7022B are not described here.

[0085] In a preferred embodiment, the measurement circuit can be used to collect multiple electrical parameter data of the charging pile during operation, including: charging current, input voltage, output voltage, charging power, power factor, accumulated charging power, grid frequency, etc.

[0086] In a preferred embodiment, the sensors configured by the data acquisition module include various sensors for collecting ambient temperature, charging pile circuit temperature, ambient humidity, charging pile geographical location, etc.

[0087] In a preferred embodiment, the analysis module and the control module can be remotely deployed in a geographical location far away from the charging pile or the charging pile group, and communicate with the data acquisition module in a wired or wireless connection manner.

[0088] In a preferred embodiment, an analysis module and / or a control module may be configured for each charging pile group.

[0089] In some preferred embodiments, a common analysis module and / or control module may be configured for multiple charging pile groups to facilitate centralized calibration and large-scale management of the charging piles.

[0090] Further, in a preferred embodiment, when intervention is implemented to adjust the operating parameters of the charging pile, one or more of the following adjustment operations may be performed to attempt to calibrate the charging pile and to perform continuous observation of subsequent charging data on the charging pile: charging current adjustment, charging voltage adjustment, and charging pile total output power adjustment;

[0091] And further, it may include changing the working state of the heat dissipation device to adjust the working temperature of the charging pile;

[0092] And further, the power factor can be adjusted to change the charging efficiency.

[0093] The calibration system of this technical solution can monitor the operating status of the charging pile in real time through big data analysis and the isolation forest algorithm, and accurately detect abnormalities in the charging process. By calculating the abnormality score, the system can quickly identify the abnormal operation of the charging pile, and then automatically calibrate the abnormal charging pile through the remote control module, adjust the current, voltage or power and other parameters, and ensure the continuous and stable operation of the system.

[0094] The calibration system of this technical solution can automatically count and process the historical data of charging piles, make intelligent judgments on the accumulated abnormal points or abnormal scores of charging piles within a preset statistical period, and automatically perform corresponding response control operations. This greatly reduces the need for manual maintenance, reduces operating costs, and improves maintenance efficiency and system reliability.

[0095] The calibration system of this technical solution can process multi-dimensional data from multiple charging piles, such as charging current, voltage, duration, power, etc., and supports data segmentation, real-time preprocessing and phased anomaly detection. This solution can be flexibly expanded according to the data requirements of different scenarios and different models, and is suitable for charging pile networks of different sizes, with good scalability.

[0096] Example 2

[0097] Combined with Figure 3 This embodiment is different from the previous embodiment in that it further discloses extracting a charging data set, performing data preprocessing on the big data in the charging data set, and after obtaining the charging data set, preprocessing the data in the data set to ensure the accuracy of subsequent analysis and anomaly detection; the goal of data preprocessing is to standardize and clean a large amount of data in the charging process so as to better identify anomalies in the charging behavior. The steps of extracting a charging data set and performing data preprocessing on the big data in the charging data set include:

[0098] S211: extracting the original charging data set of the charging pile to form an original data set;

[0099] S212: performing data cleaning processing on the original data set;

[0100] S213: performing normalization processing on the data set and standardizing processing on each feature data;

[0101] S214: Perform feature engineering processing on the standardized data set to extract features;

[0102] S215: Taking each feature as a separate dimension, the data of each complete charging process is constructed into a multi-dimensional feature vector

[0103] Specifically, extract the original charging data set of a charging pile to form the original data set X = {x1, x2, ... x i}, where x i A multidimensional feature vector representing a charging process numbered i; each multidimensional feature vector x integrates multiple operating parameters, and preferably includes vehicle information and charging pile information, with each feature being a dimension in x.

[0104] Specifically, the original data set X is subjected to data cleaning processing, preferably including processing missing values, that is, processing missing data in the charging data set, which can be processed by interpolation, mean filling or discarding the record; further, outliers in the data set are removed, that is, obviously unreasonable data points (such as charging current that is suddenly too high or too low) are deleted, and these obvious noise points are identified and removed through preliminary statistical analysis; further, it also includes time synchronization of the data, aligning the data based on the time tag to ensure that all charging parameters are in the correct position on the time axis.

[0105] Specifically, the data set X is normalized and each feature data is standardized. Preferably, the standardization method uses Z-score standardization to adjust the data value of each feature to a normal distribution with a mean of 0 and a standard deviation of 1, so as to ensure that features of different dimensions have the same weight in the analysis.

[0106] Specifically, feature engineering is performed on the standardized data set to extract meaningful features of the charging behavior; preferably, the features that can be extracted may include:

[0107] Voltage change rate V′(t), that is: V′=ΔV / Δt;

[0108] Current change rate I′(t), that is: I′=ΔI / Δt;

[0109] Charging efficiency, that is, charging efficiency = charged power / charging time;

[0110] Peak voltage V max 、Current I max , which is the maximum voltage or current value reached during the charging process, used to reflect abnormal fluctuations in the charging process;

[0111] The charging power curve P(t) is used to calculate the change of charging power over time and understand whether the power fluctuates abnormally during the charging process.

[0112] Specifically, the above-mentioned features are taken as separate dimensions, and the data of each complete charging process is constructed into a multidimensional feature vector; illustratively, the multidimensional feature vector x of the charging process can be expressed as:

[0113] x=[V′(t),I′(t),charging efficiency,V max ,I max ,P(t), vehicle model, battery model];

[0114] For each charging process, the above characteristic vector can be formed; optionally, the elements in the characteristic vector can be selected according to the actual situation. This embodiment only selects several related elements by way of example, and is not intended to be a limiting implementation method of the present technical solution;

[0115] It should be noted that, in the following description, each multidimensional feature vector x can be regarded as a multidimensional data point for analysis.

[0116] Optionally, the charging data can be further segmented, exemplarily by time periods or charging stages (such as initial stage, stable stage, and final stage) to more finely analyze the characteristic behaviors of different stages; this step further refines the data and can more accurately detect local anomalies.

[0117] When the original multidimensional feature vector x is segmented to form multiple sub-feature vectors, for example, when segmented according to the charging stage, the multidimensional feature vector x in a complete charging journey can be divided into three feature vectors x a 、x b and x c ; Then each multidimensional feature vector x is segmented a 、x b and x c can be considered as one data point.

[0118] The calibration method of AC charging pile based on big data remote communication, the calibration system can monitor the operation status of the charging pile in real time through big data analysis and isolation forest algorithm, and accurately detect abnormal conditions in the charging process. Through the calculation of abnormal scores, the system can quickly identify the abnormal operation of the charging pile, and then automatically calibrate the abnormal charging pile through the remote control module, adjust the current, voltage or power and other parameters, and ensure the continuous and stable operation of the system.

[0119] Example 3

[0120] Combined with Figure 4 This embodiment is different from the previous embodiment in that it further discloses establishing an algorithm tree based on an isolation forest algorithm, recursively segmenting a charging data set by randomly selecting features and random thresholds; specifically, the steps of establishing an abnormal analysis algorithm tree based on multiple isolation forest algorithms include the following steps:

[0121] S221: Randomly select feature vector x iSelect one feature f from the multiple feature elements contained;

[0122] S222: Select the value of feature f and find its minimum value f in the data set i-min and the maximum value f i-max , and randomly select a split point d between the two, and we can get two subsets U up and U low .

[0123] Among them, the split point d is used as a partition condition, and the data set X is divided into two subsets U based on the feature f and the corresponding split point d. up and U low ,Right now,

[0124] U up (f)≥d,

[0125] U ulow (f) <d

[0126] Repeat the above steps until the subset is segmented into only one data point or the depth of the segmentation tree reaches the preset maximum depth; that is, until any of the following conditions is met:

[0127] (1) The subset is divided into only one data point, that is, reaching the leaf node;

[0128] (2) The depth of the partition tree reaches the preset maximum depth.

[0129] The calibration method of AC charging pile based on big data remote communication, the calibration system can monitor the operation status of the charging pile in real time through big data analysis and isolation forest algorithm, and accurately detect abnormal conditions in the charging process. Through the calculation of abnormal scores, the system can quickly identify the abnormal operation of the charging pile, and then automatically calibrate the abnormal charging pile through the remote control module, adjust the current, voltage or power and other parameters, and ensure the continuous and stable operation of the system.

[0130] Example 4

[0131] This embodiment is different from the previous embodiment in that it further discloses that in step S23, the expected value of the path length of the strategy tree for each data point is calculated; wherein a path length function h(m) is defined, and the path length function h(m) indicates that starting from the root node, a data point m can reach a leaf node after being segmented h(m) times; if the data point m is very easy to segment, that is, it can be isolated with a relatively small number of segmentations, then it will reach the leaf node earlier, and the path length function h(m) will be shorter, and such a point is more likely to be an abnormal point; conversely, if the data point m needs to be isolated after many segmentations, its path length function h(m) will be longer, and such a data point is usually a "normal point" because it has a high similarity with other data points.

[0132] The formula for the expected path length of the strategy tree for each data point is:

[0133]

[0134] Where E[h(m)] represents the expected path length of data point m, h(m) represents the path length in each individual strategy tree, and n represents the number of data points in the dataset X.

[0135] Furthermore, in step S24, the anomaly score formula for each data point in the electrical data set is:

[0136]

[0137] Among them, S(m,n) represents the anomaly score, μ(n) represents the normalization constant, and h(m) represents the average path length when the number of samples n is given, which is used to standardize the path length of sample m;

[0138] Among them, the normalization constant μ(n) formula is:

[0139]

[0140] Here, H() represents the harmonic number function, and it is approximately considered that H(n)≈ln(n)+r, where r is the Euler constant.

[0141] It should be noted that data points are classified as normal or abnormal according to their anomaly scores, namely:

[0142]

[0143] In the above formula, ψ1 and ψ2 are abnormal thresholds; preferably, ψ1=0.75, ψ2=0.25, or are specifically set according to the actual abnormal performance of the charging system.

[0144] Example 5

[0145] This embodiment is different from the previous embodiment in that this embodiment further discloses that the corresponding control operation is processed by a control module; wherein the control module includes counting the number of abnormal data points accumulated by each charging pile within a preset statistical period, and when the number of abnormal data points accumulated by a charging pile exceeds a preset threshold in the preset statistical period, executing an abnormal response control operation on the charging pile.

[0146] In a preferred embodiment, the following program code may be used to perform the above analysis steps:

[0147] import numpy as np

[0148] import pandas as pd

[0149] from sklearn.ensemble import IsolationForest

[0150] #Set the charging data set data.csv, which contains the charging parameters of multiple charging piles

[0151] df=pd.read_csv('charging_data.csv')

[0152] #Remove missing values

[0153] df = df.dropna()

[0154] # Normalize the data

[0155] from sklearn.preprocessingimport StandardScaler

[0156] scaler = StandardScaler()

[0157] df[['charging_current','charging_voltage','charging_duration','charging_power']]=scaler.fit_transform(

[0158] df[['charging_current','charging_voltage','charging_duration','charging_power']] )

[0160] #Set n_estimators to 100 to use 100 isolated trees

[0161] isolation_forest=IsolationForest(n_estimators=100,contamination=0.1,random_state=42)

[0162] #Train model

[0163] isolation_forest.fit(df[['charging_current','charging_voltage','charging_duration','charging_power']])

[0164] #Get the path length:

[0165] path_lengths=isolation_forest.decision_function(df[['charging_current','charging_voltage','charging_duration','charging_power']])

[0166] # Calculate the anomaly score for each data point

[0167] scaler=MinMaxScaler(feature_range=(0,1))

[0168] anomaly_scores=scaler.fit_transform(anomaly_scores_raw.reshape(-1,1))df['anomaly_score']=1-anomaly_scores

[0169] #Anomaly score calculation, the lower the score, the more likely it is an anomaly

[0170] #Determine the charging pile to which the abnormal data point belongs, and set the data set column 'charging_pile_id' to represent the charging pile number

[0171] #Filter out abnormal data points and determine the charging pile to which they belong

[0172] anomalies=df[df['anomaly']==-1]

[0173] anomaly_charging_piles=anomalies['charging_pile_id'].unique()

[0174] #Output abnormal charging piles and related data

[0175] print("Anomaly charging piles found:", anomaly_charging_piles)

[0176] print("Abnormal data record:\n",anomalies)

[0177] Furthermore, in some preferred embodiments, a tendency detection may be further performed on a specific vehicle model to determine whether the abnormality rate increases due to compatibility issues between the charging pile and the specified vehicle model when the charging operation is performed.

[0178] Example 6

[0179] As attached Figure 5 As shown, in order to compare the actual voltage and target voltage deviation rates of multiple groups of charging piles using the calibration system described in the technical solution and those not using the calibration system, by collecting the average value of the voltage deviation rate of 50 charging piles during 1000 hours of operation, it can be seen that the performance of the charging piles after using the calibration system is more stable as a whole.

[0180] In a preferred embodiment, the anomaly score of a specific vehicle model is compared with the anomaly score of the global charging data to observe whether the behavior of the specific vehicle model deviates from the overall charging data; specifically, the method includes executing:

[0181] (1) Direct anomaly score comparison: For a dataset X of a specific model model , using the same isolation forest model as the global data, calculate the anomaly score S for each charging record of this model model (m); then S model (m) and the global dataset X global The anomaly scores are compared with

[0182] (2) Divergence calculation: This method can more quantitatively measure the difference between the charging behavior distribution of a specific vehicle model and the global charging behavior distribution. Divergence calculation can be used to compare the difference between two probability distributions. Preferably, the following calculation formula can be used to calculate the divergence:

[0183]

[0184] In the above formula, P(i) is the abnormal score distribution of a specific vehicle model, and Q(i) is the abnormal score distribution of the global charging data. By calculating the divergence, if the D(P‖Q) value is large, it means that the charging behavior of a specific vehicle model is significantly different from the overall charging behavior, and there is potential abnormal behavior.

[0185] In some more embodiments, algorithms such as Jensen-Shannon divergence or Wasserstein distance may be used to perform abnormal analysis on specific vehicle models.

[0186] This embodiment should be understood to include at least all the features of any of the above embodiments, and further improve upon them;

[0187] For example, as shown in the attached Figure 6 As shown, the implementation of the computer system 500 used by the control module 30 in the calibration system is described; the computer system 500 can be used to identify and judge the data storage, calculation and result output process of each working module in the system;

[0188] Illustratively, computer system 500 includes a bus 502 or other communication mechanism for communicating information, one or more processors 504 coupled to bus 502 for processing information; processor 504 may be, for example, one or more general-purpose microprocessors;

[0189] The computer system 500 also includes a main memory 506, such as a random access memory (RAM), a cache, and / or other dynamic storage device, coupled to the bus 502 for storing information and instructions to be executed by the processor 504; the main memory 506 may also be used to store temporary variables or other intermediate information during the execution of instructions to be executed by the processor 504; these instructions, when stored in a storage medium accessible to the processor 504, present the computer system 500 as a special-purpose machine customized to perform the operations specified in the instructions;

[0190] The computer system 500 may also include a read-only memory (ROM) 508 or other static storage device coupled to the bus 502 for storing static information and instructions for the processor 504; a storage device 510 such as a disk, an optical disk, or a USB drive (flash drive) will be coupled to the bus 502 for storing information and instructions;

[0191] And further, coupled to the bus 502 may also include a display 122 for displaying various information, data, media, etc., an input device 514 for allowing a user of the computer system 500 to control, manipulate, and / or interact with the computer system 500;

[0192] A preferred way to interact with the management system may be through a cursor control device 516, such as a computer mouse or similar control / navigation mechanism;

[0193] Furthermore, the computer system 500 may also include a network device 518 coupled to the bus 502; wherein the network device 518 may include, for example, a wired network card, a wireless network card, a switching chip, a router, a switch, and other components;

[0194] In general, the terms "engine", "component", "system", "database", etc., as used herein, may refer to logic embodied in hardware or firmware, or to a collection of software instructions, possibly with entry and exit points, written in a programming language such as Java, C, or C++; software components may be compiled and linked into executable programs, installed in a dynamic link library, or may be written in an interpreted programming language (e.g., BASIC, Perl, or Python); it should be understood that software components may be called from other components or from themselves, and / or may be called in response to detected events or interrupts;

[0195] Software components configured to execute on a computing device may be provided on a computer-readable medium, such as a compact disc, digital video disc, flash drive, diskette, or any other tangible medium, or as a digital download (and may be initially stored in a compressed or installable format that requires installation, decompression, or decryption prior to execution); such software code may be stored in part or in whole on a memory device of the executing computing device for execution by the computing device; software instructions may be embedded in firmware, such as an EPROM; it is also understood that hardware components may be composed of connected logic units (such as gates and flip-flops), and / or may be composed of programmable units (such as programmable gate arrays or processors);

[0196] Computer system 500 includes custom hardwired logic, one or more ASICs or FPGAs, firmware, and / or program logic that can be used to implement the techniques described herein, which in combination with the computer system makes computer system 500 a special-purpose computing device;

[0197] According to one or more embodiments, the techniques herein may be performed by computer system 500 in response to processor 504 executing one or more sequences of one or more instructions contained in main memory 506; such instructions may be read into main memory 506 from another storage medium, such as storage device 510; execution of the sequences of instructions contained in main memory 506 causes processor 504 to perform the process steps described herein; in alternative embodiments, hard-wired circuitry may be used in place of or in combination with software instructions;

[0198] As used herein, the term "non-transitory media" and similar terms refer to any media that store data and / or instructions that cause a machine to operate in a specific manner; such non-transitory media may include non-volatile media and / or volatile media; non-volatile media include, for example, optical or magnetic disks, such as storage device 510; volatile media include dynamic memory, such as main memory 506;

[0199] Among them, common forms of non-transitory media include, for example, floppy disks, diskettes, hard disks, solid-state drives, magnetic tapes or any other magnetic data storage medium, CD-ROMs, any other optical data storage medium, any physical medium having a pattern of holes, RAM, PROM and EPROM, FLASH-EPROM, NVRAM, any other memory chip or cartridge, and network versions thereof;

[0200] Non-transient media are distinct from transmission media but may be used in conjunction with transmission media; transmission media participate in the transmission of information between non-transient media; for example, transmission media include coaxial cables, copper wires, and optical fibers, including the wires that make up bus 502; transmission media may also take the form of sound waves or light waves, such as radio waves and infrared data communications.

[0201] Importantly, it should be noted that the construction and arrangement of the present application shown in a number of different exemplary embodiments are only exemplary. Although only a few embodiments are described in detail in this disclosure, it should be readily understood by those who refer to this disclosure that many modifications are possible, for example, the size, scale, structure, shape and proportion of various elements, and parameter values ​​such as temperature, pressure, etc., mounting arrangements, use of materials, color, directional changes, etc., without substantially departing from the novel teachings and advantages of the subject matter described in the application. For example, the element shown as integrally formed can be composed of multiple parts or elements, the position of the element can be inverted or otherwise changed, and the nature or number or position of the discrete element can be changed or changed. Therefore, all such modifications are intended to be included in the scope of the present invention. The order or sequence of any process or method steps can be changed or reordered according to alternative embodiments. In the claims, any "device plus function" clause is intended to cover the structure of the execution function described herein, and is not only structurally equivalent but also equivalent structure. Without departing from the scope of the present invention, other substitutions, modifications, changes and omissions can be made in the design, operating conditions and arrangement of the exemplary embodiments. Therefore, the invention is not limited to a specific embodiment, but extends to numerous modifications still falling within the scope of the appended claims.

[0202] Furthermore, in order to provide a concise description of exemplary embodiments, all features of an actual embodiment may not be described, i.e., those features that are not relevant to the best mode presently contemplated for carrying out the invention or those features that are not relevant to implementing the invention.

[0203] It should be understood that in the development of any actual implementation, as in any engineering or design project, numerous implementation-specific decisions may be made. Such a development effort may be complex and time-consuming, but for those of ordinary skill having the benefit of this disclosure, the development effort will be a routine task of design, fabrication, and production without undue experimentation.

[0204] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for calibrating an AC charging pile based on big data remote communication, characterized in that: include, Collect multiple operating parameter values ​​during the vehicle charging process, and record the operating parameter values ​​based on time sequence to generate a charging data set; Analyze the data set through the analysis module to obtain abnormal charging information; Perform corresponding control operations on charging piles that have abnormal charging information.

2. The AC charging pile calibration method based on big data remote communication according to claim 1, characterized in that: The values ​​of multiple operating parameters during the vehicle charging process are obtained through the data acquisition module.

3. The AC charging pile calibration method based on big data remote communication according to claim 1 or 2, characterized in that: The steps of analyzing the data set through the analysis module and obtaining abnormal charging information include: Extracting a charging data set, and performing data preprocessing on the big data in the charging data set; Establish an anomaly analysis algorithm tree based on multiple isolation forest algorithms; Calculate the expected path length of the policy tree for each data point; Calculate the anomaly score for each data point in the charging dataset; Determine the charging station to which the abnormal data point belongs.

4. The AC charging pile calibration method based on big data remote communication according to claim 3, characterized in that: The charging data set includes generating a Class A data point by taking the working parameter values ​​of a charging pile in a complete charging process; Each class A data point is split, and two or more class B data points are obtained after the split, and the two or more class B data points are added to the charging data set; Class A data points include charging current, charging voltage, charging time, charging start and end time, battery BMS status, vehicle information, charging pile information, and environmental information when the charging pile is working; Among them, vehicle information includes vehicle model, battery model used in the vehicle, battery production date and battery batch; Among them, the charging pile information includes the charging pile number and geographical location; Among them, the environmental information when the charging pile is working includes ambient temperature, humidity, air pressure and sunlight intensity.

5. The AC charging pile calibration method based on big data remote communication according to claim 4, characterized in that: Extracting the charging data set, the steps of performing data preprocessing on the big data in the charging data set include: Extract the original charging data set of the charging pile to form an original data set; Perform data cleaning on the original data set; Normalize the data set and standardize each feature data; Perform feature engineering on the standardized data set to extract features; Each feature is taken as a separate dimension, and the data of each complete charging process is constructed into a multi-dimensional feature vector.

6. The AC charging pile calibration method based on big data remote communication according to claim 5, characterized in that: The steps to build an anomaly analysis algorithm tree based on multiple isolation forest algorithms include: Randomly select one of the features from the multiple feature elements contained in the feature vector; Find the minimum and maximum values ​​of the selected feature in the data set, and randomly select a split point between the two to obtain two subsets; Repeat the above steps until the subset is segmented into only one data point or the depth of the segmentation tree reaches the preset maximum depth.

7. The AC charging pile calibration method based on big data remote communication according to claim 6, characterized in that: The formula for the expected path length of the strategy tree for each data point is: Where E[h(m)] represents the expected path length of data point m, h(m) represents the path length in each individual strategy tree, and n represents the number of data points in the dataset X.

8. The AC charging pile calibration method based on big data remote communication according to claim 7, characterized in that: The anomaly score formula for each data point in the charging dataset is: Among them, S(m,n) represents the anomaly score, μ(n) represents the normalization constant, and h(m) represents the average path length when the number of samples n is given, which is used to standardize the path length of sample m; Among them, the normalization constant μ(n) formula is: Here, H() represents the harmonic number function, and it is approximately considered that H(n)≈ln(n)+r, where r is the Euler constant.

9. The AC charging pile calibration method based on big data remote communication according to claim 8, characterized in that: The data points are classified as normal or abnormal according to their anomaly scores, namely: In the above formula, ψ1 and ψ2 are abnormal thresholds.

10. The AC charging pile calibration method based on big data remote communication according to claim 9, characterized in that: The corresponding control operations are processed by the control module; Among them, the control module includes counting the number of abnormal data points accumulated by each charging pile within a preset statistical period. When the number of abnormal data points accumulated by a charging pile exceeds a preset threshold in the preset statistical period, an abnormal response control operation is performed on the charging pile.

Images