Analysis Method and System for Detecting Electricity Theft Based on the Algorithm of Wave Injection Without Power Outage

Abstract

The present invention discloses a technical analysis method and system for detecting electricity theft based on a non-power-off wave injection algorithm, which relates to the field of power grid technology. The analysis method includes: obtaining past power grid operation data, analyzing and processing the power grid operation data to obtain user electricity consumption data and climate power consumption characteristics, dividing the user electricity consumption data according to the climate power consumption characteristics to obtain user electricity consumption category groups, analyzing the periodic characteristics and trend characteristics of the user electricity consumption data under different user electricity consumption category groups, establishing a short-term energy consumption prediction model according to the user electricity consumption data and the periodic characteristics and trend characteristics, setting the type of wave injection modulation according to the output result of the short-term energy consumption prediction model, analyzing the fluctuation of the user energy consumption after wave injection, identifying abnormal fluctuations, determining electricity theft behavior, effectively improving the efficiency of electricity theft detection, accurately locating the electricity theft point, facilitating rapid processing, and maintaining the stability of the power grid.

Description

Technical Field

[0001] The invention relates to the field of power grid technology, and in particular to a method and system for analyzing power theft verification technology based on an uninterrupted power injection algorithm. Background Art

[0002] Electricity theft refers to the act of illegally occupying electricity and not or under-metering electricity consumption in order to avoid or under-pay electricity bills. It is a serious crime that not only causes huge economic losses to the country and enterprises, but also seriously disrupts the normal power supply order, causes damage to power facilities, creates major hidden dangers, and directly threatens the safety of the power grid.

[0003] Common acts of electricity theft include unauthorized wiring and bypassing metering devices. Electricity theft has caused huge economic losses to the country and enterprises. According to statistics, my country loses 20 billion yuan every year due to electricity theft. Electricity theft will also seriously disrupt the normal power supply order and affect the stability and reliability of power supply. It may also cause damage to power facilities, create major safety hazards, and directly threaten the safety of the power grid.

[0004] The non-stop wave injection algorithm can identify non-standard current fluctuation patterns by analyzing the current waveform in the power grid in real time. These patterns may be indicators of power theft. It can continuously monitor the power grid status and dynamically adjust monitoring parameters to adapt to different power grid conditions and user behaviors, thereby achieving instant detection and early warning of power theft. The application of the non-stop wave injection algorithm in the field of power theft verification shows great potential and value.

[0005] Therefore, there is an urgent need for a technical analysis method and system for electricity theft verification based on an uninterrupted power injection algorithm to solve the above problems. Summary of the invention

[0006] To solve the above problems, the present application provides a technical analysis method and system for electricity theft verification based on a non-stop power injection algorithm, with the goal of preventing users from stealing electricity. It aims to improve the efficiency of electricity theft verification, manage and control users, and thus achieve the purpose of maintaining the stability and security of the power grid.

[0007] In the present application, a technical analysis method for verifying electricity theft based on a non-stop wave injection algorithm is provided, including: S1, obtaining past power grid operation data, analyzing and processing the power grid operation data to obtain user power consumption data and climate power consumption characteristics;

[0008] S2. Divide the user power consumption data into user power consumption category groups according to the climate power consumption characteristics, and analyze the periodic characteristics and trend characteristics of the user power consumption data under different user power consumption category groups;

[0009] S3. Establish a short-term energy consumption prediction model based on user power consumption data, cycle characteristics, and trend characteristics;

[0010] S4. Setting the type of injection wave modulation according to the output result of the short-term energy consumption prediction model;

[0011] S5. Analyze the fluctuation of user energy consumption after wave injection, identify abnormal fluctuations, and determine electricity theft behavior.

[0012] Preferably, in step S2, the specific contents of the user power consumption category groups obtained by dividing the user power consumption data according to the climate power consumption characteristics are as follows:

[0013] The climate power consumption characteristics include load rate characteristics, peak-valley difference characteristics and flexible power consumption characteristics;

[0014] Dividing the user power consumption data into a range according to the load rate characteristics to obtain a number of first category groups;

[0015] Divide the user power consumption data into several second category groups according to the peak-to-valley difference characteristics;

[0016] Merge and update the data of the first category group and the second category group according to the flexible power consumption characteristics to obtain a first updated category group and a second updated category group respectively;

[0017] The data of the first update category group and the second update category group are checked for duplication, and the duplicate data are sorted out to obtain user power consumption category groups.

[0018] Preferably, in step 2, the specific method of establishing a short-term energy consumption prediction model based on user power consumption data and cycle characteristics and trend characteristics includes:

[0019] Standardize the data of user electricity consumption category groups to eliminate abnormal data;

[0020] Create a sequential model as the model basis for the short-term energy consumption prediction model;

[0021] Add a long short-term memory network layer to the sequential model, and determine the time step and number of features of the long short-term memory network layer according to real-time requirements;

[0022] Add a long fully connected layer as the output layer in the sequential model;

[0023] The data, periodic characteristics and trend characteristics of the user electricity consumption category group are respectively put into the sequential model for model training to obtain the initial sequential model;

[0024] Use the initial sequential model to estimate the electricity consumption in the future to obtain the estimated user electricity consumption;

[0025] A time reference line is established for user electricity consumption, and a user electricity consumption prediction curve is generated with the time reference line as the horizontal axis.

[0026] Preferably, the content of establishing the short-term energy consumption prediction model also includes:

[0027] The short-term energy consumption prediction model is compiled, wherein the model compilation process includes defining an optimizer and a loss function.

[0028] Preferably, the loss function is expressed as:

[0029] ;

[0030] in, is the actual distribution of data of user electricity category groups, is the true distribution of periodic characteristics, is the true distribution of trend characteristics, is the periodic feature prediction distribution, is the trend feature prediction distribution, Predict the distribution of data for user electricity consumption category groups.

[0031] Preferably, in step S4, the specific content of setting the type of injection modulation according to the output result of the short-term energy consumption prediction model is:

[0032] Analyze the user's power consumption forecast curve to obtain the predicted power fluctuation, predicted power consumption structure, predicted power output, and predicted grid frequency in the predicted time period;

[0033] The injection wave modulation standard criteria are preset;

[0034] Select the type of injection wave modulation in the injection wave modulation specification criteria according to the predicted power fluctuation and predicted power consumption structure;

[0035] Determine the primary modulation depth in the injection wave modulation process according to the predicted power output;

[0036] Determine the primary modulation frequency in the injection wave modulation process according to the predicted grid frequency condition;

[0037] The influence coefficients of different climate change conditions on different injection wave modulation types are preset;

[0038] Acquire the climate change situation at the prediction time, and generate the climate impact factor according to the influence degree coefficient of the climate change situation on different injection wave modulation types, wherein the climate impact factor includes a climate-depth impact factor and a climate-frequency impact factor;

[0039] The injection wave modulation depth and injection wave modulation frequency are obtained by adjusting the primary modulation depth and the primary modulation frequency according to the climate influencing factors.

[0040] Preferably, the expression of the beam modulation depth is:

[0041] ;

[0042] The expression of the beam wave modulation frequency is:

[0043] ;

[0044] in, is the injection wave modulation depth, is the climate-depth impact factor, is the amplitude of the carrier, For different load power output values, is the injection wave modulation frequency, is the climate-frequency impact factor, is the proportionality coefficient, is the integration coefficient, is the differential coefficient, is the actual frequency of the power grid at time t, is the nominal frequency.

[0045] Preferably, in step S5, the fluctuation of the user's energy consumption after wave injection is analyzed, abnormal fluctuations are identified, and the specific content of the electricity theft behavior is determined as follows:

[0046] Identify the fluctuation of user energy consumption after wave injection, and determine safe fluctuations and abnormal fluctuations. If a one-time fluctuation occurs, it is defined as a safe fluctuation, and if a periodic fluctuation or continuous fluctuation occurs, it is defined as an abnormal fluctuation.

[0047] Explore the causes of fluctuations, segment electricity consumption periods according to user attributes, and find the specific sources of abnormalities;

[0048] Type analysis of abnormal sources is performed to confirm standard fluctuations of that type, thereby identifying potential electricity theft.

[0049] In the present application, a technical analysis system for verifying electricity theft based on a non-stop wave injection algorithm is also provided, including: an information acquisition module: acquiring past power grid operation data, analyzing and processing the power grid operation data to obtain user power consumption data and climate power consumption characteristics;

[0050] Initial analysis module: divide the user power consumption data into user power consumption category groups according to the climate power consumption characteristics, and analyze the periodic characteristics and trend characteristics of the user power consumption data under different user power consumption category groups;

[0051] Model prediction module: establish a short-term energy consumption prediction model based on user power consumption data, cycle characteristics, and trend characteristics, and set the type of injection wave modulation based on the output results of the short-term energy consumption prediction model;

[0052] Electricity theft analysis module: Analyzes the fluctuation of user energy consumption after wave injection, identifies abnormal fluctuations, and determines electricity theft behavior.

[0053] To summarize, compared with the traditional power grid power theft verification technology, the present invention, a technical analysis method and system for power theft verification based on the non-stop wave injection algorithm, constructs a short-term energy consumption prediction model, studies the power grid power theft evidence solidification method and technology based on the non-stop wave injection algorithm, sets the wave injection modulation type and main parameters based on the analysis of the time characteristics of the user's power time series data and the short-term energy consumption prediction curve, analyzes the fluctuation of the user's energy consumption after the wave injection, identifies abnormal fluctuations, determines the power theft behavior, effectively improves the power theft verification efficiency, accurately locates the power theft point, facilitates rapid processing, and maintains the stability of the power grid.

[0054] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A method step diagram of a method for analyzing a power theft verification technology based on a non-stop power injection algorithm according to the present invention;

[0056] Figure 2 The present invention is a module schematic diagram of a power theft verification technology analysis system based on a non-stop power injection algorithm. DETAILED DESCRIPTION

[0057] The technical solution of the present invention is further described below by means of the accompanying drawings and embodiments. It should be noted that unless otherwise specifically stated, the relative arrangement of components and steps, numerical expressions and numerical values ​​described in these embodiments do not limit the scope of the present application.

[0058] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present application, its application, or uses.

[0059] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0060] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0061] Unless otherwise defined, technical or scientific terms used in the present invention shall have the common meanings understood by one having ordinary skills in the field to which the present invention belongs.

[0062] like Figure 1As shown, the present invention provides a technical analysis method for verifying electricity theft based on a non-stop wave injection algorithm, which is specifically as follows: Step S1, obtaining past power grid operation data, analyzing and processing the power grid operation data to obtain user power consumption data and climate power consumption characteristics.

[0063] Step S2, dividing the user power consumption data into user power consumption category groups according to the climate power consumption characteristics, and analyzing the periodic characteristics and trend characteristics of the user power consumption data under different user power consumption category groups.

[0064] It is understandable that the cyclical characteristics of user electricity consumption may be affected by climatic conditions and produce seasonal changes, such as high load caused by the use of air conditioners in summer and increased heating demand in winter. There may also be peaks in the morning and evening, resulting in intraday changes, which are closely related to people's daily activity patterns. In addition, the difference in electricity consumption between weekdays and weekends, usually lower electricity consumption on weekends, causes intra-week changes. In addition, during specific holidays or special events, electricity consumption may fluctuate abnormally, resulting in holiday effects.

[0065] Trend characteristics of user electricity consumption: With economic development and population growth, the total electricity consumption shows an increasing trend year by year, that is, a long-term growth trend. It will also be affected by economic activities, market changes or policy adjustments. The electricity consumption may rise or fall rapidly in the short term, that is, a short-term fluctuation trend. In some mature markets or when technological innovations reach a balance, the electricity consumption may show a relatively stable trend.

[0066] Among them, in step 2, in step S2, the specific content of the user electricity consumption category group obtained by dividing the user electricity consumption data according to the climate power consumption characteristics is as follows: the climate power consumption characteristics include load rate characteristics, peak-valley difference characteristics and flexible power consumption characteristics.

[0067] The user power consumption data is divided into ranges according to the load rate characteristics to obtain a plurality of first category groups.

[0068] The user power consumption data is divided into ranges according to the peak-to-valley difference characteristics to obtain a number of second category groups.

[0069] The data of the first category group and the second category group are merged and updated according to the flexible power consumption characteristics to obtain the first updated category group and the second updated category group respectively.

[0070] The data of the first update category group and the second update category group are checked for duplication, and the duplicate data are sorted out to obtain user power consumption category groups.

[0071] Step S3, establishing a short-term energy consumption prediction model based on the user's electricity consumption data and cycle characteristics and trend characteristics.

[0072] Among them, in step 2, the specific method of establishing the short-term energy consumption prediction model based on the user's electricity consumption data and periodic characteristics and trend characteristics includes: standardizing the data of the user's electricity consumption category group to eliminate abnormal data.

[0073] It is understandable that in the process of constructing the short-term energy consumption prediction curve, historical power time series data can be used to construct the short-term energy consumption prediction curve and time series prediction methods can be used, such as autoregressive integrated moving average model, long short-term memory network, etc., to predict energy consumption in the future.

[0074] In this application, a sequential model is created as the model basis of the short-term energy consumption prediction model. The sequential model is a common model construction method in the deep learning framework.

[0075] Add a long short-term memory network layer to the sequential model, and determine the time step and number of features of the long short-term memory network layer based on real-time requirements.

[0076] Add a long fully connected layer as the output layer in the Sequential model.

[0077] The data, periodic characteristics and trend characteristics of the user electricity consumption category group are respectively put into the sequential model for model training to obtain the initial sequential model.

[0078] The initial sequential model is used to estimate the electricity consumption in the future to obtain the estimated user electricity consumption.

[0079] A time reference line is established for user electricity consumption, and a user electricity consumption prediction curve is generated with the time reference line as the horizontal axis.

[0080] Furthermore, the content of establishing the short-term energy consumption prediction model also includes: compiling the short-term energy consumption prediction model, wherein the model compilation process includes defining an optimizer and a loss function.

[0081] Furthermore, the loss function is expressed as:

[0082] .

[0083] in, is the actual distribution of data of user electricity category groups, is the true distribution of periodic characteristics, is the true distribution of trend characteristics, is the periodic feature prediction distribution, is the trend feature prediction distribution, Predict the distribution of data for user electricity consumption category groups.

[0084] Step S4, setting the type of injection wave modulation according to the output result of the short-term energy consumption prediction model.

[0085] Further, in step S4, the specific content of setting the type of injection modulation according to the output result of the short-term energy consumption prediction model is: analyzing the user's electricity consumption prediction curve to obtain the predicted electricity fluctuation, predicted electricity consumption structure, predicted power output, and predicted grid frequency in the predicted time period.

[0086] The injection wave modulation standard criteria are preset.

[0087] The type of injection wave modulation is selected in the injection wave modulation specification criteria according to the predicted power fluctuation and predicted power consumption structure.

[0088] The primary modulation depth in the injection wave modulation process is determined according to the predicted power output.

[0089] The primary modulation frequency in the injection wave modulation process is determined according to the predicted grid frequency conditions.

[0090] The influence coefficients of different climate change conditions on different injection wave modulation types are preset.

[0091] The climate change situation at the prediction time is obtained, and the climate impact factor is generated according to the influence degree coefficient of the climate change situation on different injection wave modulation types. The climate impact factor includes a climate-depth impact factor and a climate-frequency impact factor.

[0092] The injection wave modulation depth and injection wave modulation frequency are obtained by adjusting the primary modulation depth and the primary modulation frequency according to the climate influencing factors.

[0093] It can be understood that the modulation process includes amplitude modulation and frequency modulation. Amplitude modulation transmits information by changing the amplitude of the carrier and is suitable for application scenarios where bandwidth requirements are not high and the signal-to-noise ratio can be high. Frequency modulation transmits information by changing the frequency of the carrier and is suitable for environments that require high bandwidth and better interference suppression.

[0094] Furthermore, the expression of the beam modulation depth is:

[0095] .

[0096] It is understandable that the modulation depth determines the degree of change in the intensity of the modulated waveform. For amplitude modulation, the modulation depth directly affects the signal transmission quality and the stability of the system. Too high a modulation depth may cause signal distortion, while too low a modulation depth may cause low signal transmission efficiency. Therefore, it is necessary to select an appropriate modulation depth based on specific application requirements and environmental conditions.

[0097] The expression of the beam wave modulation frequency is:

[0098] .

[0099] in, is the injection wave modulation depth, is the climate-depth impact factor, is the amplitude of the carrier, For different load power output values, is the injection wave modulation frequency, is the climate-frequency impact factor, is the proportionality coefficient, is the integration coefficient, is the differential coefficient, is the actual frequency of the power grid at time t, is the nominal frequency.

[0100] It is understandable that the modulation frequency determines the speed of change of the signal per unit time. Selecting an appropriate modulation frequency can optimize the energy capture efficiency and dynamic response characteristics of the system. Too high a modulation frequency may cause the system to overload, while too low a modulation frequency may not be able to effectively track changes in the input signal.

[0101] S5. Analyze the fluctuation of user energy consumption after wave injection, identify abnormal fluctuations, and determine electricity theft behavior.

[0102] Among them, the specific content of analyzing the fluctuation of user energy consumption after wave injection, identifying abnormal fluctuations, and determining the electricity theft behavior in step S5 is: identifying the fluctuation of user energy consumption after wave injection, determining safe fluctuations and abnormal fluctuations, if a one-time fluctuation occurs, it is defined as a safe fluctuation, if periodic fluctuations and continuous fluctuations occur, they are defined as abnormal fluctuations.

[0103] Explore the causes of fluctuations, segment electricity consumption periods according to user attributes, and find the specific sources of anomalies.

[0104] Type analysis of abnormal sources is performed to confirm standard fluctuations of that type, thereby identifying potential electricity theft.

[0105] like Figure 2 As shown, the present invention provides a technical analysis system for verifying electricity theft based on a non-stop wave injection algorithm, including an information acquisition module: acquiring past power grid operation data, analyzing and processing the power grid operation data to obtain user power consumption data and climate power consumption characteristics.

[0106] Initial analysis module: divide the user electricity consumption data into user electricity consumption category groups according to the climate power consumption characteristics, and analyze the periodic characteristics and trend characteristics of the user electricity consumption data under different user electricity consumption category groups.

[0107] Model prediction module: establish a short-term energy consumption prediction model based on user power consumption data as well as cycle characteristics and trend characteristics, and set the type of injection wave modulation based on the output results of the short-term energy consumption prediction model.

[0108] Electricity theft analysis module: Analyzes the fluctuation of user energy consumption after wave injection, identifies abnormal fluctuations, and determines electricity theft behavior.

[0109] To summarize, compared with the traditional power grid power theft verification technology, the present invention, a technical analysis method and system for power theft verification based on the non-stop wave injection algorithm, constructs a short-term energy consumption prediction model, studies the power grid power theft evidence solidification method and technology based on the non-stop wave injection algorithm, sets the wave injection modulation type and main parameters based on the analysis of the time characteristics of the user's power time series data and the short-term energy consumption prediction curve, analyzes the fluctuation of the user's energy consumption after the wave injection, identifies abnormal fluctuations, determines the power theft behavior, effectively improves the power theft verification efficiency, accurately locates the power theft point, facilitates rapid processing, and maintains the stability of the power grid.

[0110] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution 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 they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A technical analysis method for verifying electricity theft based on non-stop power injection algorithm, characterized in that: The following steps are involved: S1. Obtain past grid operation data, analyze and process the grid operation data to obtain user power consumption data and climate power consumption characteristics; S2. Divide the user power consumption data into user power consumption category groups according to the climate power consumption characteristics, and analyze the periodic characteristics and trend characteristics of the user power consumption data under different user power consumption category groups; S3. Establish a short-term energy consumption prediction model based on user power consumption data, cycle characteristics, and trend characteristics; S4. Setting the type of injection wave modulation according to the output result of the short-term energy consumption prediction model; S5. Analyze the fluctuation of user energy consumption after wave injection, identify abnormal fluctuations, and determine electricity theft behavior; In step S2, the user power consumption data is divided according to the climate power consumption characteristics to obtain the user power consumption category groups, and the specific contents are as follows: The climate power consumption characteristics include load rate characteristics, peak-valley difference characteristics and flexible power consumption characteristics; Dividing the user power consumption data into a range according to the load rate characteristics to obtain a number of first category groups; Divide the user power consumption data into several second category groups according to the peak-to-valley difference characteristics; Merge and update the data of the first category group and the second category group according to the flexible power consumption characteristics to obtain a first updated category group and a second updated category group respectively; The data of the first update category group and the second update category group are checked for duplication, and the duplicate data are sorted out to obtain user power consumption category groups.

2. The technical analysis method for verifying electricity theft based on non-stop power injection algorithm according to claim 1 is characterized in that: In step 3, the specific method of establishing a short-term energy consumption prediction model based on user power consumption data and cycle characteristics and trend characteristics includes: Standardize the data of user electricity consumption category groups to eliminate abnormal data; Create a sequential model as the model basis for the short-term energy consumption prediction model; Add a long short-term memory network layer to the sequential model, and determine the time step and number of features of the long short-term memory network layer according to real-time requirements; Add a long fully connected layer as the output layer in the sequential model; The data, periodic characteristics and trend characteristics of the user electricity consumption category group are respectively put into the sequential model for model training to obtain the initial sequential model; Use the initial sequential model to estimate the electricity consumption in the future to obtain the estimated user electricity consumption; A time reference line is established for user electricity consumption, and a user electricity consumption prediction curve is generated with the time reference line as the horizontal axis.

3. The technical analysis method for verifying electricity theft based on non-stop power injection algorithm according to claim 2 is characterized in that: The contents of establishing the short-term energy consumption prediction model also include: The short-term energy consumption prediction model is compiled, wherein the model compilation process includes defining an optimizer and a loss function.

4. The technical analysis method for verifying electricity theft based on non-stop power injection algorithm according to claim 3 is characterized in that: The expression of the loss function is: ； in, is the actual distribution of data of user electricity category groups, is the true distribution of periodic characteristics, is the true distribution of trend characteristics, is the periodic feature prediction distribution, is the trend feature prediction distribution, Predict the distribution of data for user electricity consumption category groups.

5. The technical analysis method for verifying electricity theft based on non-stop power injection algorithm according to claim 1 is characterized in that: In step S4, the specific content of setting the type of injection wave modulation according to the output result of the short-term energy consumption prediction model is: Analyze the user's power consumption forecast curve to obtain the predicted power fluctuation, predicted power consumption structure, predicted power output, and predicted grid frequency in the predicted time period; The injection wave modulation standard criteria are preset; Select the type of injection wave modulation in the injection wave modulation specification criteria according to the predicted power fluctuation and predicted power consumption structure; Determine the primary modulation depth in the injection wave modulation process according to the predicted power output; Determine the primary modulation frequency in the injection wave modulation process according to the predicted grid frequency condition; The influence coefficients of different climate change conditions on different injection wave modulation types are preset; Acquire the climate change situation at the prediction time, and generate the climate impact factor according to the influence degree coefficient of the climate change situation on different injection wave modulation types, wherein the climate impact factor includes a climate-depth impact factor and a climate-frequency impact factor; The injection wave modulation depth and injection wave modulation frequency are obtained by adjusting the primary modulation depth and the primary modulation frequency according to the climate influencing factors.

6. The technical analysis method for verifying electricity theft based on non-stop power injection algorithm according to claim 5 is characterized in that: The expression of the beam modulation depth is: ； The expression of the beam wave modulation frequency is: ； in, is the injection wave modulation depth, is the climate-depth impact factor, is the amplitude of the carrier, For different load power output values, is the injection wave modulation frequency, is the climate-frequency impact factor, is the proportionality coefficient, is the integration coefficient, is the differential coefficient, is the actual frequency of the power grid at time t, is the nominal frequency.

7. The technical analysis method for verifying electricity theft based on non-stop power injection algorithm according to claim 1 is characterized in that: In step S5, the fluctuation of the user's energy consumption after wave injection is analyzed, abnormal fluctuations are identified, and the specific content of the electricity theft behavior is determined as follows: Identify the fluctuation of user energy consumption after wave injection, and determine safe fluctuations and abnormal fluctuations. If a one-time fluctuation occurs, it is defined as a safe fluctuation, and if a periodic fluctuation or continuous fluctuation occurs, it is defined as an abnormal fluctuation. Explore the causes of fluctuations, segment electricity consumption periods according to user attributes, and find the specific sources of abnormalities; Type analysis of abnormal sources confirms standard fluctuations of that type, thereby identifying potential electricity theft.

8. A technical analysis system for power theft verification based on non-stop power injection algorithm, characterized in that: include: Information acquisition module: obtains past grid operation data, analyzes and processes the grid operation data to obtain user power consumption data and climate power consumption characteristics; Initial analysis module: divide the user power consumption data into user power consumption category groups according to the climate power consumption characteristics, and analyze the periodic characteristics and trend characteristics of the user power consumption data under different user power consumption category groups; The specific contents of the user electricity consumption category groups obtained by dividing the user electricity consumption data according to the climate power consumption characteristics are as follows: The climate power consumption characteristics include load rate characteristics, peak-valley difference characteristics and flexible power consumption characteristics; Dividing the user power consumption data into a range according to the load rate characteristics to obtain a number of first category groups; Divide the user power consumption data into several second category groups according to the peak-to-valley difference characteristics; Merge and update the data of the first category group and the second category group according to the flexible power consumption characteristics to obtain a first updated category group and a second updated category group respectively; Check the data of the first update category group and the second update category group for duplication, and sort the duplicate data to obtain user power consumption category groups; Model prediction module: establish a short-term energy consumption prediction model based on user power consumption data, cycle characteristics, and trend characteristics, and set the type of injection wave modulation based on the output results of the short-term energy consumption prediction model; Electricity theft analysis module: Analyzes the fluctuation of user energy consumption after wave injection, identifies abnormal fluctuations, and determines electricity theft behavior.

Images