Method and device for identifying electricity usage violations of electric vehicle charging facilities

Abstract

The present invention discloses a method and device for identifying a default electricity user of an electric vehicle charging facility, comprising: analyzing the charging data of an electric vehicle to obtain a first electricity behavior characteristic variable; analyzing the daily load curve distribution of the electric vehicle to determine a second electricity behavior characteristic variable; combining the first electricity behavior characteristic variable and the second electricity behavior characteristic variable to obtain a target electricity behavior characteristic variable; clustering charging piles according to the target electricity behavior characteristic variable to determine the cluster center; calculating the relative distance from the target object to the cluster center, and identifying whether the target object is a default electricity user based on the relative distance. In the present invention, the identification process does not need to rely on manual investigation, which improves the efficiency and coverage of identifying default users of electric vehicle charging facilities.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and device for identifying electricity usage violations of electric vehicle charging facilities. Background Art

[0002] Electric vehicle charging facilities are facilities that provide energy replenishment for electric vehicles, mainly including: AC charging piles, DC charging piles and large-scale battery swap stations. Due to the different types of charging piles, the numerous electricity price categories, and the large differences in the electricity prices of different categories, some high-price customers use various means to implement lower electricity prices for all or part of their electricity prices, thereby reducing their own electricity costs through high-price low-connection. This kind of breach of contract electricity use by charging facilities endangers the safety of electricity supply and use and disrupts the normal order of electricity supply and use.

[0003] In order to reduce the losses of power supply companies, power supply companies mainly rely on manual regular inspections, random sampling, user reports and other methods to study and analyze a large amount of audit data. The workload is large and the marketing audit clues are unclear. Random inspections are adopted and judgments are made based on the experience of staff. In this case, the audit scope is small and potential abnormal users cannot be discovered.

[0004] It can be seen that the traditional identification of electric vehicle charging default electricity consumption mostly relies on manual investigation, which has problems such as high human resource consumption, low work efficiency and insufficient coverage. Summary of the invention

[0005] In view of the above problems, the present invention provides a method and device for identifying electricity usage violations in electric vehicle charging facilities, so as to solve the problems that traditional electric vehicle charging electricity usage violation identification relies mostly on manual investigation, resulting in large human resource consumption, low work efficiency and insufficient coverage.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] A method for identifying electricity usage violations of electric vehicle charging facilities, comprising:

[0008] Analyze the electric vehicle charging data to obtain the first electricity consumption behavior characteristic variable;

[0009] Analyze the daily load curve distribution of electric vehicles and determine the second power consumption behavior characteristic variable;

[0010] Combining the first electricity usage behavior characteristic variable and the second electricity usage behavior characteristic variable to obtain a target electricity usage behavior characteristic variable;

[0011] Clustering the charging piles according to the target electricity consumption behavior characteristic variables to determine the cluster center;

[0012] The relative distance from the target object to the cluster center is calculated, and based on the relative distance, it is identified whether the target object is a defaulting electricity user.

[0013] Optionally, the first power consumption behavior characteristic variable includes: the median, interval days, coefficient of variation and stationary difference order of the daily power consumption time series;

[0014] Among them, the median represents the average power consumption capacity, the interval days represent the number of days when the power consumption is lower than the power consumption threshold, the coefficient of variation represents the fluctuation of the data, and the stable difference order represents the degree of stability of the data.

[0015] Optionally, the second power consumption behavior characteristic variable includes: load intermittent duration and differential stable duration, wherein the load intermittent duration represents the non-charging time of the day, and the differential stable duration is the first-order difference of the extracted power curve, and the differential stable duration when the differential value is within a preset range.

[0016] Optionally, analyzing the daily load curve distribution of the electric vehicle to determine the second power consumption behavior characteristic variable includes:

[0017] Analyze the daily load curve distribution of electric vehicles to obtain the power curve of the target sampling point, where the target sampling point is a sampling point obtained by sampling every 15 minutes within 24 hours;

[0018] Based on the power curve of the target sampling point, a second power usage behavior characteristic variable is determined.

[0019] Optionally, clustering the charging piles according to the target power consumption behavior characteristic variable to determine the cluster center includes:

[0020] Clustering the charging piles according to the target power consumption behavior characteristic variable to obtain clustered groups, wherein the probability density function of each cluster is different;

[0021] Determine the cluster center corresponding to each group.

[0022] Optionally, calculating the relative distance from the target object to the cluster center, and identifying whether the target object is a defaulting electricity user based on the relative distance, includes:

[0023] Calculating the relative distance from the target object to the cluster center of the subgroup corresponding to the target object;

[0024] The target object whose relative distance is greater than the distance threshold is determined as a defaulting electricity user.

[0025] A device for identifying electric vehicle charging facility's breach of contract electricity usage, comprising:

[0026] A first analysis unit, used to analyze the electric vehicle charging data to obtain a first power consumption behavior characteristic variable;

[0027] A second analysis unit is used to analyze the daily load curve distribution of the electric vehicle and determine a second power consumption behavior characteristic variable;

[0028] a combining unit, configured to combine the first power usage behavior characteristic variable and the second power usage behavior characteristic variable to obtain a target power usage behavior characteristic variable;

[0029] A clustering unit, used to cluster the charging piles according to the target power consumption behavior characteristic variable and determine the cluster center;

[0030] The identification unit is used to calculate the relative distance between the target object and the cluster center, and to identify whether the target object is a defaulting electricity user based on the relative distance.

[0031] Optionally, the first power consumption behavior characteristic variable includes: the median, interval days, coefficient of variation and stationary difference order of the daily power consumption time series;

[0032] The median represents the average power consumption capacity, the interval days represent the number of days when the power consumption is lower than the power consumption threshold, the coefficient of variation represents the fluctuation of the data, and the stable difference order represents the stability of the data;

[0033] The second power consumption behavior characteristic variable includes: load intermittent time and differential stable time, wherein the load intermittent time represents the non-charging time of the day, and the differential stable time is the first-order difference of the extracted power curve, and the differential stable time when the differential value is within a preset range.

[0034] Optionally, the second analysis unit is specifically used for:

[0035] Analyze the daily load curve distribution of electric vehicles to obtain the power curve of the target sampling point, where the target sampling point is a sampling point obtained by sampling every 15 minutes within 24 hours;

[0036] Based on the power curve of the target sampling point, a second power usage behavior characteristic variable is determined.

[0037] Optionally, the clustering unit is specifically used for:

[0038] Clustering the charging piles according to the target power consumption behavior characteristic variable to obtain clustered groups, wherein the probability density function of each cluster is different;

[0039] Determine the cluster center corresponding to each group;

[0040] The identification unit is specifically used for:

[0041] Calculating the relative distance from the target object to the cluster center of the subgroup corresponding to the target object;

[0042] The target object whose relative distance is greater than the distance threshold is determined as a defaulting electricity user.

[0043] Compared with the prior art, the present invention provides a method and device for identifying the default electricity consumption of electric vehicle charging facilities, including: analyzing the charging data of electric vehicles to obtain a first electricity consumption behavior characteristic variable; analyzing the daily load curve distribution of electric vehicles to determine a second electricity consumption behavior characteristic variable; combining the first electricity consumption behavior characteristic variable and the second electricity consumption behavior characteristic variable to obtain a target electricity consumption behavior characteristic variable; clustering charging piles according to the target electricity consumption behavior characteristic variable to determine the cluster center; calculating the relative distance from the target object to the cluster center, and identifying whether the target object is a default electricity user based on the relative distance. In the present invention, the identification process does not need to rely on manual investigation, which improves the efficiency and coverage of identifying default users of electric vehicle charging facilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0045] Figure 1 A schematic diagram of a flow chart of a method for identifying a violation of electricity usage by an electric vehicle charging facility provided by an embodiment of the present invention;

[0046] Figure 2 A schematic diagram of a high power curve provided by an embodiment of the present invention;

[0047] Figure 3 A schematic diagram of a medium power curve provided by an embodiment of the present invention;

[0048] Figure 4 A schematic diagram of a low battery curve provided by an embodiment of the present invention;

[0049] Figure 5 A 96-point active power curve of a user on a certain day provided by an embodiment of the present invention;

[0050] Figure 6 A schematic diagram of a cluster of active power curves of 96 points per day for a certain user for one month provided by an embodiment of the present invention;

[0051] Figure 7 A schematic diagram of the median distribution of low-voltage electricity consumption provided by an embodiment of the present invention;

[0052] Figure 8 A low-voltage electricity fluctuation distribution diagram provided by an embodiment of the present invention;

[0053] Fig. 9 The probability density diagram of each attribute of cluster 1 provided in the embodiment of the present invention;

[0054] Fig.10 The probability density diagram of each attribute of cluster 2 provided in the embodiment of the present invention;

[0055] Fig.11 The probability density diagram of each attribute of cluster 3 provided in the embodiment of the present invention;

[0056] Fig.12 A schematic diagram of a discrete point detection distance error provided by an embodiment of the present invention;

[0057] Fig.13 A schematic diagram of the structure of a device for identifying electricity usage violations in electric vehicle charging facilities provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0058] 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.

[0059] The terms "first" and "second" and the like in the specification and claims of the present invention and the above drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but may include steps or units that are not listed.

[0060] In an embodiment of the present invention, a method for identifying the default in electricity consumption of electric vehicle charging facilities is provided. The default in electricity consumption of charging facilities is identified based on the analysis of the charging power time series and the daily load curve characteristics. By statistically analyzing the charging characteristics and travel patterns of electric vehicles, the mean, number of interruption days, coefficient of variation, stationary difference order of the daily power consumption time series, and the interval duration and differential stationary duration of the daily load curve are extracted as characteristic variables of electricity consumption behavior. The charging piles are clustered according to the extracted electricity consumption characteristics, the relative distance from the sample object to the cluster center is calculated, the degree to which the object belongs to the cluster cluster is evaluated, and then the suspected users of default in electricity consumption are identified based on a pre-set reasonable threshold.

[0061] See also Figure 1 , a method for identifying a violation of electricity usage by an electric vehicle charging facility provided by an embodiment of the present invention may include the following steps:

[0062] S101. Analyze the electric vehicle charging data to obtain a first electricity consumption behavior characteristic variable.

[0063] S102: Analyze the daily load curve distribution of the electric vehicle to determine the second power consumption behavior characteristic variable.

[0064] S103: Combine the first electricity usage behavior characteristic variable and the second electricity usage behavior characteristic variable to obtain a target electricity usage behavior characteristic variable.

[0065] In order to extract the complete characteristic variables of the charging power time series in the embodiment of the present invention, the first power consumption behavior characteristic variable and the second power consumption behavior characteristic variable are respectively obtained by analyzing the electric vehicle charging data and the electric vehicle daily load curve distribution, wherein the first power consumption behavior characteristic variable and the second power consumption behavior characteristic variable both include several characteristic variables. Specifically, the first power consumption behavior characteristic variable includes: the median, the interval days, the coefficient of variation and the order of stationary differences of the daily power consumption time series; wherein the median represents the average power consumption capacity, the interval days represents the number of days when the power consumption is lower than the power consumption threshold, the coefficient of variation represents the fluctuation of the data, and the order of stationary differences represents the degree of stability of the data.

[0066] The second power consumption behavior characteristic variable includes: load intermittent time and differential stable time, wherein the load intermittent time represents the non-charging time of the day, and the differential stable time is the first-order difference of the extracted power curve, and the differential stable time when the differential value is within a preset range.

[0067] Then, the first electricity usage behavior characteristic variable and the second electricity usage behavior characteristic variable are combined to obtain a target electricity usage behavior characteristic variable, that is, all behavior characteristic variables used for subsequent analysis.

[0068] S104: Cluster the charging piles according to the target electricity consumption behavior characteristic variables to determine the cluster center.

[0069] S105: Calculate the relative distance between the target object and the cluster center, and identify whether the target object is a defaulting electricity user based on the relative distance.

[0070] The charging piles are clustered according to the extracted target electricity consumption behavior characteristic variables, and the corresponding cluster center can be determined in the clustering process. The target object refers to the user who needs to be identified, and then the relative distance from the target object to the cluster center is calculated, and the relative distance is compared with the preset threshold to determine whether the target object belongs to the cluster cluster corresponding to the cluster center, thereby determining whether the target object is a suspected user of electricity consumption in violation of the contract.

[0071] In one embodiment of the present invention, the analysis of the daily load curve distribution of the electric vehicle to determine the second electricity consumption behavior characteristic variable includes: analyzing the daily load curve distribution of the electric vehicle to obtain the power curve of the target sampling point, the target sampling point is a sampling point obtained by sampling every 15 minutes within 24 hours; based on the power curve of the target sampling point, determining the second electricity consumption behavior characteristic variable.

[0072] The reason why the above-mentioned characteristic variables are selected as the final target electricity consumption behavior characteristic variables in the embodiment of the present invention is that by analyzing the daily electricity consumption curve and the 96-point load curve, it can be concluded that the above-mentioned electricity consumption behavior characteristic variables can reflect the user's actual electricity consumption behavior, thereby making accurate judgments.

[0073] When analyzing the daily electricity consumption curve rules, a set of daily electricity consumption curves within the vicinity of the average daily electricity consumption values ​​of three segments is randomly selected to observe the distribution and change rules. Figure 2-Figure 4 It can be seen that the curve fluctuates greatly and the trend is stable, with no obvious upward or downward trend. The high power consumption is not interrupted, the medium power consumption is interrupted, and the low power consumption is obvious. Figure 2 This is a schematic diagram of the high power curve. Figure 3 This is a schematic diagram of the medium power curve. Figure 4 This is a schematic diagram of the low battery curve.

[0074] See also Figure 5 and Figure 6 , Figure 5 This is the 96-point active power curve of a certain user on a certain day. Figure 6 This is a schematic diagram of a user's 96-point daily active power curve cluster for one month. 96 points means that the instantaneous power value is sampled every 15 minutes, so there are 96 power values ​​every day. Analyzing the regularity of the 96-point load curve, the power increases instantaneously and continues during charging. After charging, the power decreases instantaneously. When not charging, the power remains low and close to 0. That is, the daily load curve shows intermittent power consumption, and there are sudden rises and falls at the beginning and end of power consumption, which is similar to the superposition of multiple piecewise constant function curves. Putting the power curves of each period of time together, it is found that the charging period of each day is similar, indicating that the user has a regular travel pattern.

[0075] According to the characteristics of business data, the median, number of interruption days, coefficient of variation, and stationary difference order of the daily electricity consumption time series are selected as characteristic attributes of electricity consumption behavior.

[0076] Median: It can describe the average power consumption capacity and avoid the influence of extreme values ​​to a certain extent. From its distribution, it can be judged that the median is lower than the mean. See Figure 7 , which shows a schematic diagram of the median distribution of low-voltage electricity consumption provided by an embodiment of the present invention. It can be seen that the distribution of low-voltage electricity and low-power users is more dense.

[0077] Intermittent days: When the power consumption of a day is zero or lower than 0.2, it is presumed that no one uses it that day, which is defined as intermittent days. The number of intermittent days of charging piles is counted. It can be seen from Table 1 that 67% of low-voltage users have intermittent power use, and it is presumed that low-voltage intermittent power use is a normal power use behavior.

[0078] Table 1

[0079] Interval days statistics Low voltage users count 5233 mean 34.96 std 36.82 min 0 25％ 0 50％ 18 75％ 72 max 92 Number of users with 92 days of interruption 710 Number of users with 0 days of interruption 1726

[0080] Coefficient of variation: The coefficient of variation (standard deviation / mean) can describe the fluctuation of data to a certain extent. The larger the coefficient of variation, the greater the volatility of the sequence. Figure 8 It can be concluded from the low-voltage electricity consumption fluctuation distribution diagram that the normal electricity consumption of low-voltage charging piles fluctuates greatly.

[0081] Stationarity: The unit root check (ADF) method is used to check the stationarity of the data to determine whether the data sequence has a trend. When the data is stable, it is checked after successive differences until it is stable, and the final difference order is recorded. When the order is 0, it means that the data is stable, and other numbers represent non-stationary data, which is stable after n-order differences. It can be concluded from Table 2 that most low-voltage users are stable.

[0082] Table 2

[0083] Stationary after n-order differences Stability Number of low voltage users Proportion 0 smooth 4335 82.84％ 1 Non-stationary 862 16.47％ >＝2 Non-stationary 36 0.69％

[0084] When extracting the characteristic variables of the daily load curve, according to the characteristics of the business data, the interval duration and differential stable duration of the daily load curve are selected as the characteristic attributes of the electricity consumption behavior, that is, the second characteristic variable of the electricity consumption behavior.

[0085] Load intermittency: The power consumption of the charging pile during the interval between two charges is very low. When the active power is less than 2, it is estimated that it is not charged. The uncharged time of the day is counted and defined as the load intermittent time. A long intermittent time indicates normal power consumption, and a short intermittent time indicates long continuous power consumption, which indicates abnormal power consumption.

[0086] Load step-by-step nature: In a charging event, the charging power is relatively stable. When charging starts or ends, the power increases or decreases instantaneously. Therefore, the power curve of a single charging pile every day usually exhibits piecewise constant function characteristics, that is, step-by-step nature. Fit the power curve function and find the derivative function. The derivative function value fluctuates mostly around zero, with a few extreme values. Take the first-order difference of the power curve. When the difference value is between plus or minus 0.1, it is defined as differential stability. The statistical difference stability duration is used. If the time is long, it is inferred that the power consumption is normal, otherwise it is abnormal power consumption.

[0087] In a possible implementation of the present invention, clustering the charging piles according to the target electricity consumption behavior characteristic variables to determine the cluster center includes: clustering the charging piles according to the target electricity consumption behavior characteristic variables to obtain clustered groups, wherein the probability density function of each cluster is different; and determining the cluster center corresponding to each cluster.

[0088] Correspondingly, the calculation of the relative distance from the target object to the cluster center, and identification of whether the target object is a defaulting electricity user based on the relative distance, includes: calculating the relative distance from the target object to the cluster center of the group corresponding to the target object; determining the target object whose relative distance is greater than a distance threshold as a defaulting electricity user.

[0089] Specifically, the charging piles are clustered according to the extracted electricity consumption characteristic variables (also called electricity consumption characteristic attributes), and the distance (or relative distance) from the target object to the cluster center is calculated. The degree to which the object belongs to the cluster is evaluated based on this, and the defaulting electricity users are identified.

[0090] According to the sample data, the median, number of interruption days, coefficient of variation, order of stationary difference, interval duration of 96-point active power curve in the recent week, and differential stability duration of the daily electricity consumption time series in the recent month are selected as the characteristic attributes of electricity consumption behavior. Since ordinary electricity meters do not collect 96-point load data, the characteristic attributes of the daily electricity consumption time series in the recent month can be selected at this time. According to the existing sample data, the K-Means clustering algorithm is used to divide it into 3 categories. The probability density function of each cluster is as follows Figures 9 to 11 As shown, Fig. 9 The probability density diagram of each attribute of cluster 1 is shown. Fig.10 It is the probability density diagram of each attribute of cluster 2. Fig.11 This is the probability density diagram of each attribute of cluster 3.

[0091] The characteristics of cluster 1 are: the median daily electricity consumption is concentrated between 0 and 2, the coefficient of variation is concentrated between 1 and 4, and the number of intermittent days is concentrated between 30 and 90 days.

[0092] The characteristics of cluster 2 are: the median daily electricity consumption is concentrated between 0 and 80, the coefficient of variation is concentrated between 0 and 3, and the number of intermittent days is concentrated between 0 and 30 days.

[0093] The characteristics of cluster 3 are as follows: the median daily electricity consumption is concentrated between 0 and 120, the coefficient of variation is concentrated between 0 and 2, and the number of intermittent days is concentrated between 0 and 10 days.

[0094] It can be seen that the median daily electricity consumption of cluster 1 is small, the fluctuation is large, and the discontinuity is significant. At the same time, according to the unit root test (ADF), it is stable, which represents a small-capacity residential charging pile; the median daily electricity consumption, volatility, and discontinuity of cluster 2 are all at a medium level. At the same time, according to the unit root test (ADF), it is stable, representing a medium-capacity residential or commercial charging pile; the median daily electricity consumption of cluster 3 is large, the fluctuation is small, and the discontinuity is not significant. According to the unit root test (ADF), most of them are non-stationary, which are large-capacity commercial charging piles.

[0095] Calculate the relative distance between each sample (i.e., target object) and the center of each cluster, analyze the outlier samples, set the threshold to 3, and identify 111 defaulting electricity users from 5233 low-voltage users. Fig.12 , which is a distance error diagram. The horizontal axis is the number, the vertical axis is the relative distance, the light gray represents normal users, and the dark gray represents abnormal users.

[0096] The embodiment of the present invention provides a method for identifying the default electricity user of an electric vehicle charging facility, comprising: analyzing the charging data of the electric vehicle to obtain a first electricity behavior characteristic variable; analyzing the daily load curve distribution of the electric vehicle to determine a second electricity behavior characteristic variable; combining the first electricity behavior characteristic variable and the second electricity behavior characteristic variable to obtain a target electricity behavior characteristic variable; clustering the charging piles according to the target electricity behavior characteristic variable to determine the cluster center; calculating the relative distance from the target object to the cluster center, and identifying whether the target object is a default electricity user based on the relative distance. In the present invention, the identification process does not need to rely on manual investigation, which improves the efficiency and coverage of identifying default users of electric vehicle charging facilities.

[0097] Based on the above embodiments, the present invention also provides a device for identifying the default electricity usage of electric vehicle charging facilities, see Fig.13 ,include:

[0098] A first analysis unit 10 is used to analyze the electric vehicle charging data to obtain a first power consumption behavior characteristic variable;

[0099] A second analysis unit 20 is used to analyze the daily load curve distribution of the electric vehicle and determine a second power consumption behavior characteristic variable;

[0100] A combining unit 30, configured to combine the first power consumption behavior characteristic variable and the second power consumption behavior characteristic variable to obtain a target power consumption behavior characteristic variable;

[0101] A clustering unit 40, configured to cluster the charging piles according to the target power consumption behavior characteristic variable and determine a cluster center;

[0102] The identification unit 50 is used to calculate the relative distance between the target object and the cluster center, and identify whether the target object is a defaulting electricity user based on the relative distance.

[0103] Furthermore, the first power consumption behavior characteristic variable includes: the median, interval days, coefficient of variation and stationary difference order of the daily power consumption time series;

[0104] The median represents the average power consumption capacity, the interval days represent the number of days when the power consumption is lower than the power consumption threshold, the coefficient of variation represents the fluctuation of the data, and the stable difference order represents the stability of the data;

[0105] The second power consumption behavior characteristic variable includes: load intermittent time and differential stable time, wherein the load intermittent time represents the non-charging time of the day, and the differential stable time is the first-order difference of the extracted power curve, and the differential stable time when the differential value is within a preset range.

[0106] Furthermore, the second analysis unit is specifically used for:

[0107] Analyze the daily load curve distribution of electric vehicles to obtain the power curve of the target sampling point, where the target sampling point is a sampling point obtained by sampling every 15 minutes within 24 hours;

[0108] Based on the power curve of the target sampling point, a second power usage behavior characteristic variable is determined.

[0109] Furthermore, the clustering unit is specifically used for:

[0110] Clustering the charging piles according to the target power consumption behavior characteristic variable to obtain clustered groups, wherein the probability density function of each cluster is different;

[0111] Determine the cluster center corresponding to each group;

[0112] The identification unit is specifically used for:

[0113] Calculating the relative distance from the target object to the cluster center of the group corresponding to the target object;

[0114] The target object whose relative distance is greater than the distance threshold is determined as a defaulting electricity user.

[0115] The embodiment of the present invention provides a device for identifying the default electricity consumption of electric vehicle charging facilities, comprising: a first analysis unit analyzes the charging data of electric vehicles to obtain a first electricity consumption behavior characteristic variable; a second analysis unit analyzes the daily load curve distribution of electric vehicles to determine a second electricity consumption behavior characteristic variable; a combination unit combines the first electricity consumption behavior characteristic variable with the second electricity consumption behavior characteristic variable to obtain a target electricity consumption behavior characteristic variable; a clustering unit clusters charging piles according to the target electricity consumption behavior characteristic variable to determine the cluster center; an identification unit calculates the relative distance from the target object to the cluster center, and identifies whether the target object is a default electricity user based on the relative distance. In the present invention, the identification process does not need to rely on manual investigation, which improves the efficiency and coverage of identifying default users of electric vehicle charging facilities.

[0116] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0117] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0118] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for identifying electricity usage violations of electric vehicle charging facilities, characterized in that: include: Analyze the electric vehicle charging data to obtain the first electricity consumption behavior characteristic variable; Analyze the daily load curve distribution of electric vehicles and determine the second power consumption behavior characteristic variable; Combining the first electricity usage behavior characteristic variable and the second electricity usage behavior characteristic variable to obtain a target electricity usage behavior characteristic variable; Clustering the charging piles according to the target electricity consumption behavior characteristic variables to determine the cluster center; Calculating the relative distance between the target object and the cluster center, and identifying whether the target object is a defaulting electricity user based on the relative distance; The first power consumption behavior characteristic variable includes: the median, interval days, coefficient of variation and stable difference order of the daily power consumption time series; The median represents the average power consumption capacity, the interval days represent the number of days when the power consumption is lower than the power consumption threshold, the coefficient of variation represents the fluctuation of the data, and the stable difference order represents the stability of the data; Among them, the second power consumption behavior characteristic variable includes: load intermittent duration and differential stable duration, wherein the load intermittent duration represents the non-charging time of the day, and the differential stable duration is the first-order difference of the extracted power curve, and the differential stable duration when the differential value is within a preset range.

2. The method according to claim 1, characterized in that The analyzing the daily load curve distribution of the electric vehicle to determine the second power consumption behavior characteristic variable includes: Analyze the daily load curve distribution of electric vehicles to obtain the power curve of the target sampling point, where the target sampling point is a sampling point obtained by sampling every 15 minutes within 24 hours; Based on the power curve of the target sampling point, a second power usage behavior characteristic variable is determined.

3. The method according to claim 1, characterized in that The clustering of charging piles according to the target power consumption behavior characteristic variable to determine the cluster center includes: Clustering the charging piles according to the target power consumption behavior characteristic variable to obtain clustered groups, wherein the probability density function of each cluster is different; Determine the cluster center corresponding to each group.

4. The method according to claim 3, characterized in that The calculating the relative distance between the target object and the cluster center, and identifying whether the target object is a defaulting electricity user based on the relative distance, includes: Calculating the relative distance from the target object to the cluster center of the subgroup corresponding to the target object; The target object whose relative distance is greater than the distance threshold is determined as a defaulting electricity user.

5. A device for identifying the violation of electricity usage by electric vehicle charging facilities, characterized in that: include: A first analysis unit, used to analyze the electric vehicle charging data to obtain a first power consumption behavior characteristic variable; A second analysis unit is used to analyze the daily load curve distribution of the electric vehicle and determine a second power consumption behavior characteristic variable; a combining unit, configured to combine the first power usage behavior characteristic variable and the second power usage behavior characteristic variable to obtain a target power usage behavior characteristic variable; A clustering unit, used to cluster the charging piles according to the target power consumption behavior characteristic variable and determine the cluster center; an identification unit, used to calculate the relative distance between the target object and the cluster center, and to identify whether the target object is a defaulting electricity user based on the relative distance; The first power consumption behavior characteristic variable includes: the median, interval days, coefficient of variation and stable difference order of the daily power consumption time series; The median represents the average power consumption capacity, the interval days represent the number of days when the power consumption is lower than the power consumption threshold, the coefficient of variation represents the fluctuation of the data, and the stable difference order represents the stability of the data; The second power consumption behavior characteristic variable includes: load intermittent time and differential stable time, wherein the load intermittent time represents the non-charging time of the day, and the differential stable time is the first-order difference of the extracted power curve, and the differential stable time when the differential value is within a preset range.

6. The device according to claim 5, characterized in that The second analysis unit is specifically used for: Analyze the daily load curve distribution of electric vehicles to obtain the power curve of the target sampling point, where the target sampling point is a sampling point obtained by sampling every 15 minutes within 24 hours; Based on the power curve of the target sampling point, a second power usage behavior characteristic variable is determined.

7. The device according to claim 5, characterized in that The clustering unit is specifically used for: Clustering the charging piles according to the target power consumption behavior characteristic variable to obtain clustered groups, wherein the probability density function of each cluster is different; Determine the cluster center corresponding to each group; The identification unit is specifically used for: Calculating the relative distance from the target object to the cluster center of the subgroup corresponding to the target object; The target object whose relative distance is greater than the distance threshold is determined as a defaulting electricity user.

Images