Multi-dimensional monitoring system and method for preventing electricity stealing

Abstract

The invention discloses a multi-dimensional monitoring system and method for preventing electricity larceny. The system comprises a data processing module, and a data acquisition module and an electricity larceny early warning module which are connected with the data processing module, relates to the technical field of electricity larceny monitoring, and solves the technical problem of low accuracy of anti-electricity larceny monitoring in the prior art. According to the invention, the electricity stealing suspicion coefficient is obtained through analysis based on the sample electricity consumption data; acquiring an electricity stealing suspicion coefficient prediction value of the user in the prediction period; comparing the predicted value of the electricity larceny suspicion coefficient with a preset suspicion range threshold value to obtain an electricity larceny suspicion level; and generating electricity larceny early warning information corresponding to each user based on the basic information and the electricity larceny suspicion level. According to the method, sample power consumption data of a plurality of continuous periods are fitted into a power consumption curve, and power consumption change characteristics are obtained through derivation; the electricity larceny suspicion coefficient of each user is obtained by performing formulation calculation on the electricity utilization change characteristics, so that the accuracy of anti-electricity larceny monitoring on the user is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of electricity theft monitoring, and in particular relates to a multi-dimensional monitoring system and method for anti-electricity theft. Background Art

[0002] In the marketing business of power companies, the quality of functional management such as electricity consumption, electricity charges, electricity metering, and electricity consumption inspection depends on whether the power supply and use are reasonable and whether the electricity charges are normal. Electricity charges are the main source of income for power companies and are greatly affected by the social electricity consumption order. Electricity theft makes it impossible for power companies to achieve normal electricity billing, resulting in huge losses. Therefore, it is necessary to ensure the real-time and effectiveness of anti-electricity theft work in order to timely detect electricity theft, avoid risks, and reduce economic losses of power companies.

[0003] The existing technology continuously monitors the electricity consumption data of users in the power supply area and obtains the users' electricity theft information by analyzing the electricity consumption data; however, the users' electricity theft behavior often occurs in a specific time period. The existing anti-electricity theft monitoring method is difficult to obtain the users' electricity theft probability in different time periods, resulting in low accuracy and efficiency of anti-electricity theft monitoring.

[0004] The present invention proposes a multi-dimensional monitoring system and method for anti-electricity theft to solve the above technical problems. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a multi-dimensional monitoring system and method for anti-electricity theft, which is used to solve the technical problem that the anti-electricity theft monitoring method in the prior art is difficult to obtain the probability of electricity theft by users in different time periods, resulting in low accuracy and efficiency of anti-electricity theft monitoring.

[0006] To achieve the above-mentioned object, a first aspect of the present invention provides a multi-dimensional monitoring system for anti-electricity theft, comprising: a data processing module, and a data acquisition module and an electricity theft warning module connected thereto;

[0007] The data acquisition module is used to obtain basic information of several users in the power supply area and the original power consumption data of each user in several consecutive periods; pre-process the original power consumption data to obtain sample power consumption data; wherein the basic information includes the user's address, name and payment record, and the original power consumption data and sample power consumption data both include the acquisition time, voltage data, current data and power consumption;

[0008] The data processing module is used to analyze the sample electricity consumption data to obtain the electricity theft suspicion coefficient; obtain the electricity theft suspicion coefficient prediction value of the user in the prediction period; compare the electricity theft suspicion coefficient prediction value with the preset suspicion range threshold to obtain the electricity theft suspicion level; wherein the electricity theft suspicion level includes no electricity theft suspicion, slight suspicion and severe suspicion;

[0009] The electricity theft warning module generates electricity theft warning information corresponding to each user based on basic information and the suspected level of electricity theft; wherein the electricity theft warning information includes basic information and the suspected level of electricity theft.

[0010] Preferably, the preprocessing of the original electricity consumption data includes:

[0011] A1: Extract raw electricity consumption data of several consecutive cycles;

[0012] A2: Eliminate abnormal data from the original electricity consumption data;

[0013] A3: Fill in the missing data in the original electricity consumption data;

[0014] A4: Integrate the pre-processed original electricity consumption data into sample electricity consumption data, and mark the serial number of the sample electricity consumption data of each user as i; wherein i=1, 2, ..., n, and n is the total number of user groups in the power supply area.

[0015] Preferably, the analysis based on the sample electricity usage data to obtain the electricity theft suspicion coefficient includes:

[0016] B1: Extract sample electricity consumption data of several consecutive cycles;

[0017] B2: Fitting the sample power consumption data of several consecutive cycles into a power consumption curve; wherein the power consumption curve includes a voltage curve, a current curve and a power consumption curve;

[0018] B3: Based on the power consumption curve, the power consumption change characteristics are obtained through data analysis; wherein the power consumption change characteristics include the maximum voltage fluctuation, the maximum current fluctuation and the maximum power consumption fluctuation;

[0019] B4: Calculate the electricity theft suspicion coefficient QDXi of user i by the formula QDXi=a×DYi+b×DLi+c×YDLi; where DYi is the voltage derivative of user i, DLi is the current derivative of user i, YDLi is the electricity consumption derivative of user i, and a, b, and c are proportional coefficients greater than 0.

[0020] The present invention obtains electricity consumption variation characteristics by fitting sample electricity consumption data of several continuous cycles into electricity consumption curves and calculating the first-order derivative functions of each electricity consumption curve; the electricity theft suspicion coefficient of each user is obtained by formulating and calculating the electricity consumption variation characteristics, so that the electricity theft suspicion coefficient can quantify the correlation between the user's electricity consumption data and the probability of electricity theft, thereby facilitating improving the accuracy of anti-electricity theft monitoring of users.

[0021] Preferably, fitting the sample power consumption data of several consecutive cycles into a power consumption curve comprises:

[0022] Extract sample electricity consumption data of several consecutive periods; establish an electricity consumption curve with time as the independent variable and sample electricity consumption data as the dependent variable.

[0023] Preferably, the power consumption change characteristics are obtained by data analysis based on the power consumption curve, including:

[0024] Extracting the power consumption curve; calculating the first-order derivative functions of the voltage curve, the current curve and the power consumption curve respectively, and obtaining the first derivative function of the voltage, the first derivative function of the current and the first derivative function of the power consumption;

[0025] Take the absolute values ​​of the first derivative of voltage, the first derivative of current and the first derivative of power consumption to obtain the second derivative of voltage, the second derivative of current and the second derivative of power consumption;

[0026] The maximum value of voltage derivative, current derivative and power consumption derivative are obtained from the second derivative of voltage, the second derivative of current and the second derivative of power consumption respectively, and the maximum value of voltage derivative is marked as the maximum value of voltage fluctuation, the maximum value of current derivative is marked as the maximum value of current fluctuation, and the maximum value of power consumption derivative is marked as the maximum value of power consumption fluctuation.

[0027] Preferably, the step of obtaining a predicted value of a user's electricity theft suspicion coefficient during a prediction period includes:

[0028] The electricity theft suspicion coefficient of the user in recent consecutive cycles is extracted and marked as a standard data set, and the standard data set is input into the electricity theft prediction model to obtain the predicted value of the electricity theft suspicion coefficient of the user in the prediction period; wherein, the electricity theft prediction model is obtained through artificial intelligence model training.

[0029] Preferably, the electricity theft prediction model is obtained by training an artificial intelligence model, including:

[0030] Extract the electricity theft suspicion coefficients of several consecutive cycles and integrate them into several groups of training data and test data; use the training data to train the artificial intelligence model; use the test data to test the trained artificial intelligence model, and adjust the artificial intelligence model according to the test results; finally obtain an electricity theft prediction model whose input is the electricity theft suspicion coefficients of several recent consecutive cycles and whose output is the electricity theft suspicion coefficient of the prediction cycle; wherein the artificial intelligence model includes a BP neural network model or an RBF neural network model.

[0031] The present invention extracts the predicted value of the electricity theft suspicion coefficient for training the artificial intelligence model. After the training is completed, the electricity theft prediction model is obtained. The electricity theft suspicion coefficient of a prediction period is obtained by inputting the electricity theft suspicion coefficient of several consecutive periods into the electricity theft prediction model. The electricity theft suspicion coefficient can be estimated in advance for the user in the corresponding time period, so that the power company can quickly and accurately verify the users with serious suspicion of electricity theft, which is conducive to improving the efficiency and accuracy of anti-electricity theft monitoring of users.

[0032] Preferably, the comparison between the predicted value of the electricity theft suspicion coefficient and a preset suspicion range threshold value includes:

[0033] C1: Extract the predicted value of the electricity theft suspicion coefficient;

[0034] C2: Set the suspect range threshold [XY1, XY2]; where XY1 <XY2;

[0035] C3: Determine whether the predicted value of the electricity theft suspicion coefficient is within the suspicion range threshold; if yes, mark the electricity theft suspicion level as mild suspicion; if no, jump to C4;

[0036] C4: Determine whether the predicted value of the electricity theft suspicion coefficient is greater than the maximum value XY2 of the suspicion range threshold; if yes, mark the electricity theft suspicion level as severe suspicion; if no, mark the electricity theft suspicion level as no suspicion.

[0037] Preferably, the generating of electricity theft warning information corresponding to each user based on basic information and electricity theft suspicion level includes:

[0038] Extract the basic information and the suspected level of electricity theft of each user; determine whether the suspected level of electricity theft is no suspicion of electricity theft; if yes, no electricity theft warning information will be generated; if no, the basic information and the suspected level of electricity theft will be integrated into the electricity theft warning information.

[0039] A second aspect of the present invention provides a multi-dimensional monitoring method for anti-electricity theft, comprising:

[0040] S1: Obtain basic information of several users in the power supply area and the original power consumption data of each user in several consecutive periods;

[0041] S2: preprocessing the original electricity consumption data to obtain sample electricity consumption data;

[0042] S3: Analyze the sample electricity consumption data to obtain the electricity theft suspicion coefficient;

[0043] S4: Obtain the predicted value of the user's electricity theft suspicion coefficient during the prediction period;

[0044] S5: Compare the electricity theft suspicion coefficient prediction value with the preset suspicion range threshold to obtain the electricity theft suspicion level;

[0045] S6: Generate electricity theft warning information corresponding to each user based on the basic information and the electricity theft suspicion level.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] 1. The present invention obtains electricity consumption variation characteristics by fitting sample electricity consumption data of several continuous periods into electricity consumption curves and calculating the first-order derivative functions of each electricity consumption curve; the electricity theft suspicion coefficient of each user is obtained by formulating and calculating the electricity consumption variation characteristics, so that the electricity theft suspicion coefficient can quantify the correlation between the user's electricity consumption data and the probability of electricity theft, thereby facilitating improving the accuracy of anti-electricity theft monitoring of users.

[0048] 2. The present invention extracts the predicted value of the electricity theft suspicion coefficient for training the artificial intelligence model. After the training is completed, the electricity theft prediction model is obtained. The electricity theft suspicion coefficient of several consecutive periods is input into the electricity theft prediction model to obtain the electricity theft suspicion coefficient of the prediction period. This can estimate in advance the probability of electricity theft by users in the corresponding time period, so that the power company can quickly and accurately verify users with serious suspicion of electricity theft, which is conducive to improving the efficiency and accuracy of anti-electricity theft monitoring of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] 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 some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 It is a schematic diagram of the principle of the multi-dimensional monitoring system against electricity theft of the present invention;

[0051] Figure 2 It is an overall flow chart of the multi-dimensional monitoring method against electricity theft of the present invention;

[0052] Figure 3 A flow chart for calculating the electricity theft suspicion coefficient in the present invention;

[0053] Figure 4 The present invention is a flow chart for determining the level of suspicion of electricity theft based on the predicted value of the suspicion coefficient of electricity theft. DETAILED DESCRIPTION

[0054] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than 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.

[0055] See also Figure 1-Figure 4 , a first aspect of the present invention provides a multi-dimensional monitoring system for anti-electricity theft, comprising: a data processing module, and a data acquisition module and an electricity theft warning module connected thereto;

[0056] Data collection module: used to obtain basic information of several users in the power supply area and the original power consumption data of each user in several consecutive cycles; pre-process the original power consumption data to obtain sample power consumption data; the basic information includes the user's address, name and payment record, and the original power consumption data and sample power consumption data both include the collection time, voltage data, current data and power consumption;

[0057] Data processing module: used to analyze the sample electricity consumption data to obtain the electricity theft suspicion coefficient; obtain the electricity theft suspicion coefficient prediction value of the user in the prediction period; compare the electricity theft suspicion coefficient prediction value with the preset suspicion range threshold to obtain the electricity theft suspicion level; wherein the electricity theft suspicion level includes no electricity theft suspicion, slight suspicion and severe suspicion;

[0058] Electricity theft warning module: generates electricity theft warning information corresponding to each user based on basic information and suspected electricity theft level; wherein the electricity theft warning information includes basic information and suspected electricity theft level.

[0059] In this embodiment, the original power consumption data is preprocessed, including:

[0060] A1: Extract raw electricity consumption data of several consecutive cycles;

[0061] A2: Eliminate abnormal data from the original electricity consumption data;

[0062] A3: Fill in the missing data in the original electricity consumption data;

[0063] A4: Integrate the pre-processed original electricity consumption data into sample electricity consumption data, and mark the serial number of the sample electricity consumption data of each user as i; wherein i=1, 2, ..., n, and n is the total number of user groups in the power supply area.

[0064] Exemplarily, the abnormal data in the original electricity consumption data include data on collection time, voltage data, current data, and electricity consumption that are obviously inconsistent with common sense, for example: the collection time is greater than the current time, the voltage data of the civil electricity voltage is greater than 250V, the current data, the electricity consumption is negative, etc.; the abnormal data in the original electricity consumption data is eliminated, the average value of each original electricity consumption data is obtained, and the average value of each original electricity consumption data is filled into the corresponding missing data.

[0065] In this embodiment, there are three users in the power supply area, which are marked as user 1, user 2 and user 3 respectively.

[0066] In this embodiment, the electricity theft suspicion coefficient is obtained based on the analysis of the sample electricity consumption data, including:

[0067] B1: Extract sample electricity consumption data of several consecutive cycles;

[0068] B2: Fitting the sample power consumption data of several consecutive cycles into a power consumption curve; wherein the power consumption curve includes a voltage curve, a current curve and a power consumption curve;

[0069] B3: Based on the power consumption curve, the power consumption change characteristics are obtained through data analysis; wherein the power consumption change characteristics include the maximum voltage fluctuation, the maximum current fluctuation and the maximum power consumption fluctuation;

[0070] B4: Calculate the electricity theft suspicion coefficient QDXi of user i by the formula QDXi=a×DYi+b×DLi+c×YDLi; where DYi is the voltage derivative of user i, DLi is the current derivative of user i, YDLi is the electricity consumption derivative of user i, and a, b, and c are proportional coefficients greater than 0.

[0071] Exemplarily, the proportional coefficients a=10, b=12, c=11 are set; the maximum voltage fluctuation of user 1 DY1=0.3, the maximum current fluctuation of user 1 DL1=0.2, the maximum power consumption fluctuation of user 1 YDL1=0.4; the electricity theft suspicion coefficient of user 1 QDX1=9.8 is calculated by the formula.

[0072] The present invention obtains electricity consumption variation characteristics by fitting sample electricity consumption data of several continuous cycles into electricity consumption curves and calculating the first-order derivative functions of each electricity consumption curve; the electricity theft suspicion coefficient of each user is obtained by formulating and calculating the electricity consumption variation characteristics, so that the electricity theft suspicion coefficient can quantify the correlation between the user's electricity consumption data and the probability of electricity theft, thereby facilitating improving the accuracy of anti-electricity theft monitoring of users.

[0073] In this embodiment, the sample power consumption data of several consecutive cycles are fitted into a power consumption curve, including:

[0074] Extract sample electricity consumption data of several consecutive periods; establish an electricity consumption curve with time as the independent variable and sample electricity consumption data as the dependent variable.

[0075] In this embodiment, the power consumption change characteristics are obtained through data analysis based on the power consumption curve, including

[0076] Extracting the power consumption curve; calculating the first-order derivative functions of the voltage curve, the current curve and the power consumption curve respectively, and obtaining the first derivative function of the voltage, the first derivative function of the current and the first derivative function of the power consumption;

[0077] Take the absolute values ​​of the first derivative of voltage, the first derivative of current and the first derivative of power consumption to obtain the second derivative of voltage, the second derivative of current and the second derivative of power consumption;

[0078] The maximum value of voltage derivative, current derivative and power consumption derivative are obtained from the second derivative of voltage, the second derivative of current and the second derivative of power consumption respectively, and the maximum value of voltage derivative is marked as the maximum value of voltage fluctuation, the maximum value of current derivative is marked as the maximum value of current fluctuation, and the maximum value of power consumption derivative is marked as the maximum value of power consumption fluctuation.

[0079] In this embodiment, obtaining the predicted value of the electricity theft suspicion coefficient of the user in the prediction period includes:

[0080] The electricity theft suspicion coefficient of the user in recent consecutive cycles is extracted and marked as a standard data set, and the standard data set is input into the electricity theft prediction model to obtain the predicted value of the electricity theft suspicion coefficient of the user in the prediction period; wherein, the electricity theft prediction model is obtained through artificial intelligence model training.

[0081] In this embodiment, the electricity theft prediction model is obtained by training an artificial intelligence model, including:

[0082] The electricity theft suspicion coefficients of several consecutive cycles are extracted and integrated into several groups of original data, 80% of the original data are used as training data and 20% as test data; the training data are used to train the artificial intelligence model; the test data are used to test the trained artificial intelligence model, and the artificial intelligence model is adjusted according to the test results; finally, an electricity theft prediction model is obtained, the input of which is the electricity theft suspicion coefficients of the most recent several consecutive cycles and the output is the electricity theft suspicion coefficient of the prediction cycle; wherein the artificial intelligence model includes a BP neural network model or an RBF neural network model.

[0083] The present invention extracts the predicted value of the electricity theft suspicion coefficient for training the artificial intelligence model. After the training is completed, the electricity theft prediction model is obtained. The electricity theft suspicion coefficient of a prediction period is obtained by inputting the electricity theft suspicion coefficient of several consecutive periods into the electricity theft prediction model. The electricity theft suspicion coefficient can be estimated in advance for the user in the corresponding time period, so that the power company can quickly and accurately verify the users with serious suspicion of electricity theft, which is conducive to improving the efficiency and accuracy of anti-electricity theft monitoring of users.

[0084] In this embodiment, the comparison between the predicted value of the electricity theft suspicion coefficient and the preset suspicion range threshold is performed, including:

[0085] C1: Extract the predicted value of the electricity theft suspicion coefficient;

[0086] C2: Set the suspect range threshold [XY1, XY2]; where XY1 <XY2;

[0087] C3: Determine whether the predicted value of the electricity theft suspicion coefficient is within the suspicion range threshold; if yes, mark the electricity theft suspicion level as mild suspicion; if no, jump to C4;

[0088] C4: Determine whether the predicted value of the electricity theft suspicion coefficient is greater than the maximum value XY2 of the suspicion range threshold; if yes, mark the electricity theft suspicion level as severe suspicion; if no, mark the electricity theft suspicion level as no suspicion.

[0089] For example, the predicted value QDY of the electricity theft suspicion coefficient is set 预测 =12.4, the threshold of the suspected range [XY1, XY2] = [10, 12]; due to the predicted value of the electricity theft suspicion coefficient QDY 预测 =12.4 is not within the suspected range threshold [XY1, XY2], so jump to C4; because the predicted value of the suspected coefficient of electricity theft QDY 预测 =12.4 is greater than the maximum value of the suspicion range threshold value XY2=12, so the corresponding user's electricity theft suspicion level is marked as serious suspicion.

[0090] In this embodiment, electricity theft warning information corresponding to each user is generated based on basic information and the suspected level of electricity theft, including:

[0091] Extract the basic information and the suspected level of electricity theft of each user; determine whether the suspected level of electricity theft is no suspicion of electricity theft; if yes, no electricity theft warning information will be generated; if no, the basic information and the suspected level of electricity theft will be integrated into the electricity theft warning information.

[0092] Exemplarily, the user's electricity theft suspicion level is set to severe suspicion; because the user's electricity theft suspicion level is not no electricity theft suspicion, the basic information and the electricity theft suspicion level are integrated into electricity theft warning information.

[0093] A second aspect of the present invention provides a multi-dimensional monitoring method for anti-electricity theft, including:

[0094] S1: Obtain basic information of several users in the power supply area and the original power consumption data of each user in several consecutive periods;

[0095] S2: preprocessing the original electricity consumption data to obtain sample electricity consumption data;

[0096] S3: Analyze the sample electricity consumption data to obtain the electricity theft suspicion coefficient;

[0097] S4: Obtain the predicted value of the user's electricity theft suspicion coefficient during the prediction period;

[0098] S5: Compare the electricity theft suspicion coefficient prediction value with the preset suspicion range threshold to obtain the electricity theft suspicion level;

[0099] S6: Generate electricity theft warning information corresponding to each user based on the basic information and the electricity theft suspicion level.

[0100] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0101] Working principle of the present invention:

[0102] The present invention obtains basic information of several users in a power supply area and original electricity consumption data of each user in several continuous periods; pre-processes the original electricity consumption data to obtain sample electricity consumption data; obtains a suspected electricity theft coefficient based on analysis of the sample electricity consumption data; obtains a predicted value of the suspected electricity theft coefficient of the user in a predicted period; obtains a suspected electricity theft level based on comparison of the predicted value of the suspected electricity theft coefficient and a preset suspected range threshold; and generates corresponding electricity theft warning information for each user based on the basic information and the suspected electricity theft level.

[0103] The above embodiments are only used to illustrate the technical method 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 method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A multi-dimensional monitoring system for anti-electricity theft, comprising: A data processing module, and a data acquisition module and a power theft warning module connected thereto; characterized in that the data acquisition module is used to obtain basic information of several users in the power supply area and the original power consumption data of each user in several consecutive cycles; pre-process the original power consumption data to obtain sample power consumption data; wherein the basic information includes the user's address, name and payment record, and the original power consumption data and the sample power consumption data both include the acquisition time, voltage data, current data and power consumption; The data processing module is used to analyze the sample electricity consumption data to obtain the electricity theft suspicion coefficient; obtain the electricity theft suspicion coefficient prediction value of the user in the prediction period; compare the electricity theft suspicion coefficient prediction value with the preset suspicion range threshold to obtain the electricity theft suspicion level; wherein the electricity theft suspicion level includes no electricity theft suspicion, slight suspicion and severe suspicion; The electricity theft warning module generates electricity theft warning information corresponding to each user based on basic information and the suspected level of electricity theft; wherein the electricity theft warning information includes basic information and the suspected level of electricity theft.

2. A multi-dimensional monitoring system for anti-electricity theft according to claim 1, characterized in that: The preprocessing of the original power consumption data includes: A1: Extract raw electricity consumption data of several consecutive cycles; A2: Eliminate abnormal data from the original electricity consumption data; A3: Fill in the missing data in the original electricity consumption data; A4: Integrate the pre-processed original electricity consumption data into sample electricity consumption data, and mark the serial number of the sample electricity consumption data of each user as i; wherein i=1, 2, ..., n, and n is the total number of user groups in the power supply area.

3. A multi-dimensional monitoring system for anti-electricity theft according to claim 2, characterized in that: The electricity theft suspicion coefficient obtained by analyzing the sample electricity consumption data includes: B1: Extract sample electricity consumption data of several consecutive cycles; B2: Fitting the sample power consumption data of several consecutive cycles into a power consumption curve; wherein the power consumption curve includes a voltage curve, a current curve and a power consumption curve; B3: Based on the power consumption curve, the power consumption change characteristics are obtained through data analysis; wherein the power consumption change characteristics include the maximum voltage fluctuation, the maximum current fluctuation and the maximum power consumption fluctuation; B4: Calculate the electricity theft suspicion coefficient QDXi of user i by the formula QDXi=a×DYi+b×DLi+c×YDLi; where DYi is the voltage derivative of user i, DLi is the current derivative of user i, YDLi is the electricity consumption derivative of user i, and a, b, and c are proportional coefficients greater than 0.

4. A multi-dimensional monitoring system for anti-electricity theft according to claim 3, characterized in that: The step of fitting the sample power consumption data of a plurality of consecutive cycles into a power consumption curve comprises: Extract sample electricity consumption data of several consecutive periods; establish an electricity consumption curve with time as the independent variable and sample electricity consumption data as the dependent variable.

5. The multi-dimensional monitoring system for anti-electricity theft according to claim 3, characterized in that: The power consumption change characteristics are obtained through data analysis based on the power consumption curve, including Extracting the power consumption curve; calculating the first-order derivative functions of the voltage curve, the current curve and the power consumption curve respectively, and obtaining the first derivative function of the voltage, the first derivative function of the current and the first derivative function of the power consumption; Take the absolute values ​​of the first derivative of voltage, the first derivative of current and the first derivative of power consumption to obtain the second derivative of voltage, the second derivative of current and the second derivative of power consumption; The maximum value of voltage derivative, current derivative and power consumption derivative are obtained from the second derivative of voltage, the second derivative of current and the second derivative of power consumption respectively, and the maximum value of voltage derivative is marked as the maximum value of voltage fluctuation, the maximum value of current derivative is marked as the maximum value of current fluctuation, and the maximum value of power consumption derivative is marked as the maximum value of power consumption fluctuation.

6. A multi-dimensional monitoring system for anti-electricity theft according to claim 1, characterized in that: The obtaining of the predicted value of the electricity theft suspicion coefficient of the user during the prediction period includes: The electricity theft suspicion coefficient of the user in recent consecutive cycles is extracted and marked as a standard data set, and the standard data set is input into the electricity theft prediction model to obtain the predicted value of the electricity theft suspicion coefficient of the user in the prediction period; wherein, the electricity theft prediction model is obtained through artificial intelligence model training.

7. A multi-dimensional monitoring system for anti-electricity theft according to claim 6, characterized in that: The electricity theft prediction model is obtained through artificial intelligence model training, including: Extract the electricity theft suspicion coefficients of several consecutive cycles and integrate them into several groups of training data and test data; use the training data to train the artificial intelligence model; use the test data to test the trained artificial intelligence model, and adjust the artificial intelligence model according to the test results; finally obtain an electricity theft prediction model whose input is the electricity theft suspicion coefficients of several recent consecutive cycles and whose output is the electricity theft suspicion coefficient of the prediction cycle; wherein the artificial intelligence model includes a BP neural network model or an RBF neural network model.

8. The multi-dimensional monitoring system for anti-electricity theft according to claim 1, characterized in that: The comparison between the predicted value of the electricity theft suspicion coefficient and a preset suspicion range threshold value includes: C1: Extract the predicted value of the electricity theft suspicion coefficient; C2: Set the suspect range threshold [XY1, XY2]; where XY1 <XY2; C3: Determine whether the predicted value of the electricity theft suspicion coefficient is within the suspicion range threshold; if yes, mark the electricity theft suspicion level as mild suspicion; if no, jump to C4; C4: Determine whether the predicted value of the electricity theft suspicion coefficient is greater than the maximum value XY2 of the suspicion range threshold; if yes, mark the electricity theft suspicion level as severe suspicion; if no, mark the electricity theft suspicion level as no suspicion.

9. A multi-dimensional monitoring system for anti-electricity theft according to claim 1, characterized in that: The generating of electricity theft warning information corresponding to each user based on basic information and electricity theft suspicion level includes: Extract the basic information and the suspected level of electricity theft of each user; determine whether the suspected level of electricity theft is no suspicion of electricity theft; if yes, no electricity theft warning information will be generated; if no, the basic information and the suspected level of electricity theft will be integrated into the electricity theft warning information.

10. A multi-dimensional monitoring method for anti-electricity theft, based on the multi-dimensional monitoring system for anti-electricity theft according to any one of claims 1 to 9, characterized in that: include: S1: Obtain basic information of several users in the power supply area and the original power consumption data of each user in several consecutive periods; S2: preprocessing the original electricity consumption data to obtain sample electricity consumption data; S3: Analyze the sample electricity consumption data to obtain the electricity theft suspicion coefficient; S4: Obtain the predicted value of the user's electricity theft suspicion coefficient during the prediction period; S5: Compare the electricity theft suspicion coefficient prediction value with the preset suspicion range threshold to obtain the electricity theft suspicion level; S6: Generate electricity theft warning information corresponding to each user based on the basic information and the electricity theft suspicion level.

Images