Method and device for identifying electricity stealing behavior of user

Abstract

The invention discloses a method and a device for identifying an electricity stealing behavior of a user, and belongs to the technical field of intelligent power grids. In the method, a user is equipped with a user electric energy meter for recording an electric energy meter abnormal event and user power consumption data; the electric energy meter abnormal event is used for representing potential electricity stealing operation on the user electric energy meter; and in response to an electric energy meter abnormal event recorded by the user electric energy meter, identifying the electricity stealing behavior of the user according to the electric energy meter abnormal event and the user electricity consumption data. According to the method disclosed by the invention, firstly, the potential electricity stealing operation executed on the electric energy meter of the user is represented through the electric energy meter abnormal event recorded by the electric energy meter, and then whether the electric energy meter abnormal event is caused by the electricity stealing behavior or not is judged in combination with the analysis of the electricity utilization data of the user, so that the electricity stealing behavior of the user is accurately identified.

Description

Technical Field

[0001] The present invention generally relates to the field of smart grid technology. More specifically, the present invention relates to a method and device for identifying a user's electricity theft behavior. Background Art

[0002] Electricity theft will cause economic losses to power supply companies. If the monitoring and crackdown on electricity theft is not strengthened, the interests of power supply companies will not be guaranteed.

[0003] In the existing technology, the method of manually checking all users in high-loss areas and high-loss power supply lines, or using machine learning algorithms to identify electricity theft, is usually adopted. However, with the upgrading of electricity theft methods, the above two methods cannot adapt to new electricity theft scenarios due to their low recognition efficiency and lack of targeting of new types of electricity theft.

[0004] In view of this, there is an urgent need to provide a method for identifying a user's electricity theft behavior, so as to effectively and accurately identify the user's electricity theft behavior according to the characteristics of the new type of electricity theft behavior. Summary of the invention

[0005] In order to at least solve one or more technical problems described in the above background technology section, the present invention proposes the following technical solutions and multiple embodiments thereof.

[0006] In a first aspect, the present invention discloses a method for identifying user electricity theft behavior, wherein the user is equipped with a user electric energy meter for recording abnormal events of the electric energy meter and the user's electricity usage data; the abnormal events of the electric energy meter are used to characterize potential electricity theft operations on the user's electric energy meter; the method comprises: in response to the user's electric energy meter recording the abnormal events of the electric energy meter, identifying the user's electricity theft behavior according to the abnormal events of the electric energy meter and the user's electricity usage data.

[0007] In a second aspect, the present invention discloses a device for identifying a user's electricity theft behavior, comprising a processor configured to execute program instructions; and a memory configured to store program instructions, so that when the program instructions are loaded and executed by the processor, the device executes the method described in the first aspect.

[0008] The method and device for identifying the user's electricity theft behavior proposed in the present invention first characterizes the potential electricity theft operation performed on the user's electricity meter through the abnormal events of the electricity meter recorded by the electricity meter, and then judges whether the abnormal events of the electricity meter are caused by the electricity theft behavior in combination with the analysis of the user's electricity consumption data, thereby accurately identifying the user's electricity theft behavior. Furthermore, by controlling the electricity meter to monitor and record different abnormal events of the electricity meter, multiple user electricity theft behaviors can be identified. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein: Figure 1 A block diagram showing a hardware configuration that can implement the apparatus in the embodiment of the present invention; Figure 2 A logic flow chart showing a method for identifying electricity theft behavior of a user according to some embodiments of the present invention is shown; Figure 3 A logic flow chart of a method for identifying electricity theft behavior of a user according to some other embodiments of the present invention is shown. DETAILED DESCRIPTION

[0010] 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 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0011] It should be understood that the terms "include" and "comprising" used in the description and claims of the present invention indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0012] It should also be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the claims, the singular forms of "a", "an" and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should also be further understood that the term "and / or" used in the specification of the present invention and the claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0013] As used in this specification and claims, the term "if" can be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrases "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.

[0014] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0015] The power supply system is a complex and huge system that is responsible for transmitting electricity from power plants to end users to ensure the normal operation of social production and life. Generally speaking, the electricity produced by power plants is first boosted to a higher voltage level in a step-up substation, and then transmitted to a step-down substation over a long distance through a high-voltage transmission line. The step-down substation steps down the high-voltage electricity to a medium or low voltage level for transmission to different users. In the power distribution link, the step-down substation outputs electricity through the main distribution line. The power supply line can be mounted on the distribution bus line. The power supply line supplies power to dedicated transformers and substation transformers. Among them, dedicated transformers are used to supply power to dedicated transformer users such as large factories and commercial complexes, and substation transformers are used to supply power to low-voltage users such as residential areas and small commercial users.

[0016] Driven by economic interests, some users steal electricity to reduce electricity costs. Theft will cause economic losses to power supply companies. If the monitoring and crackdown on theft is not strengthened, the interests of power supply companies will not be guaranteed. In the prior art, the following two methods are mainly used to identify theft of electricity by users: (1) Conduct manual inspections on a household basis for all users in high-loss substations and high-loss power supply lines.

[0017] The line loss rate is an indicator used to evaluate the difference between the power supply and sales volume of the substation / power supply line. Its calculation method is shown in formula (1):

[0018] When electricity is transmitted in transmission lines and transformers and other equipment, there will be normal power loss due to resistance loss, magnetic field loss and other reasons, resulting in the amount of electricity sold being lower than the amount of power supplied. However, if the user steals electricity, the magnitude of the amount of electricity sold will be lower than the amount of power supplied. For example, for a power supply area, the normal area line loss rate caused by resistance loss, magnetic field loss and other reasons is 5%. When the measured area line loss rate is greater than the high line loss rate of 5%, the possibility of electricity theft in the area is relatively high. Once a high-loss area or power supply line is detected, on-site inspections can be carried out house by house, including checking whether the appearance of the user's electricity meter is damaged, whether the seal is intact, whether the user's household line and the wiring of the electricity meter are normal, and whether there is private wiring, bypassing the electricity meter, etc.

[0019] (2) Identify electricity theft based on machine learning algorithms.

[0020] This type of method first collects a large amount of data related to the user's electricity usage behavior and trains a model based on the collected data. The trained model is then used to analyze whether there is electricity theft in the user's electricity usage behavior.

[0021] For the first method mentioned above, when a large number of users need to be checked, a lot of manpower and time are required, resulting in low efficiency in identifying electricity theft and failure to detect electricity theft in a timely manner. Although the second method uses data-driven machine learning algorithms to solve the time-consuming and labor-intensive problem of manual inspection, the accuracy of the analysis results of the machine learning algorithm depends on the quality and quantity of the collected data. How to collect a large amount of high-quality data on complex user electricity behaviors (especially electricity theft) has become an important issue that limits the widespread implementation of this type of method.

[0022] The inventor of the present invention found that: with the rapid development of electric power technology, the means of stealing electricity have been continuously upgraded and become more hidden and complex; traditional ways of stealing electricity, such as privately pulling wires and destroying electricity meter measurement, have gradually been eliminated; they have been replaced by high-tech means of stealing electricity, such as privately modifying metering devices, installing remote control stealing appliances, current-boosting stealing appliances and other equipment, and even stealing electricity by illegal card writing and other methods; and these new means of stealing electricity often require the electric energy meter to be powered off first, and then the electric energy meter cover or terminal cover is opened to change the metering method of the electric energy meter or install electricity stealing equipment. However, the existing machine learning algorithms for identifying electricity theft behaviors are intended to detect electricity theft behaviors that conform to general rules, such as: identifying meters with negative errors based on the law of conservation of electric energy, and using clustering algorithms to identify meters that conform to the law of electricity theft. Therefore, the existing machine learning algorithms for identifying electricity theft behaviors lack the pertinence to the cover-opening electricity theft behaviors and cannot adapt to new electricity theft scenarios. In view of this, in order to solve the problem that the prior art cannot effectively identify new types of electricity theft, the present invention provides a method for identifying user electricity theft, so as to effectively and accurately identify user electricity theft based on the characteristics of new types of electricity theft.

[0023] In some embodiments, the user is equipped with a user electricity meter for recording abnormal events of the electricity meter and the user's electricity usage data; the abnormal events of the electricity meter are used to characterize potential electricity theft operations on the user's electricity meter; in response to the user's electricity meter recording abnormal events of the electricity meter, the user's electricity theft behavior is identified based on the abnormal events of the electricity meter and the user's electricity usage data.

[0024] In the embodiment of the present invention, the targeted user can be any electricity user, including but not limited to the special transformer user and low-voltage user mentioned above. The user's electric energy meter not only has the general electricity data metering function, but also has the function of monitoring abnormal events of the electric energy meter. The abnormal event of the electric energy meter refers to a type of operation performed on the electric energy meter, such as opening the electric energy meter cover, disconnecting the power supply of the electric energy meter, etc. These operations belong to the characteristic steps of the electricity theft process, and therefore can represent the potential electricity theft operation performed on the user's electric energy meter. The reason why it is a "potential" electricity theft operation is that the execution of this type of operation does not necessarily mean that the electricity theft behavior has occurred. In the normal maintenance and replacement of the electric energy meter, this type of operation will also be performed. In this regard, in order to determine whether the abnormal event of the electric energy meter is caused by the electricity theft behavior, the embodiment of the present invention further combines the user's electricity data recorded by the electric energy meter to identify the user's electricity theft behavior. It can be understood that the electricity theft behavior will change the characteristics of the user's electricity data, so that the electricity data no longer conforms to normal characteristics such as energy conservation, but conforms to the electricity theft characteristics, so the user's electricity theft behavior can be identified through the user's electricity data.

[0025] The method for identifying the user's electricity theft behavior disclosed in the present invention first characterizes the potential electricity theft operation performed on the user's electricity meter through the abnormal events of the electricity meter recorded by the electricity meter, and then judges whether the abnormal events of the electricity meter are caused by the electricity theft behavior in combination with the analysis of the user's electricity consumption data, so as to accurately identify the user's electricity theft behavior. Furthermore, by controlling the electricity meter to monitor and record different abnormal events of the electricity meter, multiple user electricity theft behaviors can be identified. Therefore, the method disclosed in the present invention integrates the data characteristics and behavioral characteristics of the electricity theft behavior, so as to identify new types of electricity theft behaviors in a targeted manner.

[0026] In some embodiments, abnormal events of the electric energy meter include at least one of the following: an electric energy meter power failure event and an electric energy meter cover opening event. The electric energy meter power failure refers to an interruption in the power supply line of the electric energy meter, which causes the internal circuit of the electric energy meter to fail to work normally. The electric energy meter cover opening event includes the electric energy meter cover opening event (opening the electric energy meter cover) and the electric energy meter terminal button cover opening event (opening the electric energy meter terminal button cover). The electric energy meter cover is usually located on the front of the electric energy meter. Its main function is to protect the internal integrated circuit board, display screen, metering chip and other key components of the electric energy meter, and to facilitate users and meter readers to observe the display screen and indicator light and other information of the electric energy meter. The terminal button cover of the electric energy meter is located at the wiring terminal of the electric energy meter. Its main function is to protect the wiring terminal of the electric energy meter from external influences. As mentioned above, new methods of electricity theft often require the electricity meter to be powered off and then the meter cover or terminal cover to be opened to change the metering method or install electricity theft equipment. Therefore, the meter power-off event and the meter cover opening event can represent potential electricity theft operations performed on the user's electricity meter.

[0027] In some embodiments, the user electricity consumption data includes at least one of the following: daily frozen indication data, voltage data, current data, electricity consumption data, area line loss rate data, and power supply line line loss rate data.

[0028] It can be understood that the user's electricity consumption data is the time series data collected by the electric energy meter during the actual operation process, including daily frozen indication data, voltage data, current data, electricity consumption data, substation line loss rate data and power line line loss rate data. Among them, the daily frozen indication refers to the electric energy metering indication obtained and recorded by the electric energy meter at a specific time every day, which indicates the user's cumulative electricity consumption. Voltage data and current data refer to the phase voltage and phase current data collected by the electric energy meter. Corresponding to the three-phase four-wire power supply system, the voltage data and current data include three-phase voltage data and three-phase current data; corresponding to the single-phase power supply system, the voltage data and current data include single-phase voltage data and single-phase current data. The electricity consumption data can include daily electricity consumption data and monthly electricity consumption data; among them, the daily frozen indications of two adjacent days are subtracted to obtain the daily electricity consumption; the daily electricity consumption of each day in a month is accumulated to obtain the monthly electricity consumption. The concept of line loss rate was introduced in the previous article. Similarly, the line loss rate of the substation is an indicator for evaluating the gap between the power supply and the power sales of the substation, and the line loss rate of the power supply line is an indicator for evaluating the gap between the power output of the step-down substation and the power sales.

[0029] In some embodiments, the user power meter includes a sensor for detecting the opening and closing actions of the user power meter; in response to the sensor detecting the cover opening action, the user power meter records the power meter cover opening event and the start time corresponding to the power meter cover opening event; in response to the sensor detecting the cover closing action, the user power meter records the end time corresponding to the power meter cover opening event.

[0030] As mentioned above, the energy meter cover opening event includes the energy meter cover opening event and the energy meter end button cover opening event. For the energy meter cover opening event, a cover sensor can be set in the energy meter for detection. The main function of the cover sensor is to detect the opening and closing actions of the user's energy meter. In some embodiments, the cover sensor can be a mechanical micro switch (for the convenience of description, it is recorded as the first micro switch). When the energy meter cover is closed, the cover will apply a certain pressure to the first micro switch, so that the first micro switch is in a closed state; once the cover is opened, the pressure disappears, and the first micro switch will return to the disconnected state due to its own elasticity; when the first micro switch enters the disconnected state from the closed state, it can be considered that the cover opening action is detected; when the first micro switch enters the closed state from the disconnected state, it can be considered that the cover closing action is detected. In these embodiments, the cover sensor can be set on the integrated circuit board inside the energy meter, and can contact the cover when the cover of the energy meter is in the closed state.

[0031] As for the event of opening the terminal knob cover of the electric energy meter, a terminal knob cover sensor can be set in the electric energy meter for detection. The main function of the terminal knob cover sensor is to detect the opening and closing actions of the terminal knob cover of the user's electric energy meter. Similar to the setting of the meter cover sensor, the terminal knob cover sensor can also be a mechanical micro switch (for the convenience of description, it is recorded as the second micro switch). When the terminal knob cover of the electric energy meter is closed, the terminal knob cover will exert a certain pressure on the second micro switch, so that the second micro switch is in a closed state; once the terminal knob cover of the electric energy meter is opened, the pressure disappears, and the second micro switch will return to the open state due to its own elasticity; when the second micro switch enters the open state from the closed state, it can be considered that the opening action of the terminal knob cover is detected; when the second micro switch enters the closed state from the open state, it can be considered that the closing action of the terminal knob cover is detected.

[0032] In some embodiments, identifying a user's electricity stealing behavior based on an abnormal event of an electric energy meter and the user's electricity usage data includes: in response to identifying a mutation point from the user's electricity usage data, determining that the corresponding user has committed electricity stealing behavior.

[0033] As mentioned above, user electricity consumption data is time series data collected by the electric energy meter during actual operation. Then, by inputting the user electricity consumption data into the time series mutation point detection algorithm, the mutation points that may be contained in the data can be identified. It can be understood that in time series data, the mutation point refers to the point where the statistical characteristics of the data (such as mean, variance, etc.) mutate, which marks the transition of the time series from one relatively stable state to another. Then for the embodiment of the present invention, when the user's electric energy meter records an abnormal event of the electric energy meter, if the mutation point is detected in the user's electricity consumption data, it can be considered that the abnormal time of the electric energy meter recorded by the user's electric energy meter is caused by electricity theft.

[0034] It is understandable that the user power consumption data may include one or more variables, and the corresponding time series data may be univariate time series data or multivariate time series data. As for the time series mutation point detection algorithm that can be used, those skilled in the art can select adaptively as needed, and the present invention does not limit this. Exemplarily, a long short-term memory (LSTM) model or a gated recurrent unit (GRU) model can be used as a detection algorithm. LSTM / GRU are two variants of recurrent neural networks (RNN), which can learn long-term dependencies in time series. For the mutation point detection task of user power consumption data, the model parameters can be adjusted using the training data set first, and then the user power consumption data to be processed is used as input, and the next value of the time series is predicted by the trained model, and the actual observed value is compared with the predicted value. If the difference is large, it is determined that a mutation point is detected.

[0035] In some embodiments, in response to identifying a mutation point from the user's electricity usage data, determining that the corresponding user has engaged in electricity theft includes: calculating the time correlation between the mutation point and the abnormal event of the electricity meter; and in response to the time correlation being not less than a set time correlation threshold, determining that the corresponding user has engaged in electricity theft.

[0036] It can be understood that after using the time series mutation point detection algorithm to identify the mutation point in the user's electricity consumption data, the mutation time corresponding to the mutation point can also be determined. The temporal correlation between the mutation point and the abnormal event of the electric energy meter refers to the degree of proximity between the process time corresponding to the abnormal event of the electric energy meter and the mutation time. If the temporal correlation is relatively high, it can be considered that the abnormal event of the electric energy meter caused the user's electricity consumption data to mutate, and the abnormal event of the electric energy meter was caused by electricity theft. In some embodiments, the temporal correlation between the mutation point and the abnormal event of the electric energy meter can be calculated based on the cosine similarity between the start time and the mutation time of the abnormal event of the electric energy meter. In these embodiments, a vector Time with a length of 6 is selected to represent the start time and the mutation time, that is, Time = (Year, Month, Day, Hour, Minute, Second). Let the start time of the abnormal event of the electric energy meter be Time1 and the mutation time be Time2, then the temporal correlation (Temporal Correlation, TC) between the mutation point and the abnormal event of the electric energy meter can be calculated by the following formula (2):

[0037] Similarly, the time correlation can also be calculated for the end time Time3 and the mutation time Time2 of the abnormal event of the electric energy meter, or the intermediate time ( ) and the mutation time Time2 to calculate the temporal correlation.

[0038] It can be understood that the time correlation calculated according to the above formula (2) is essentially the cosine similarity of two times, and the value range is [0,1]. The larger the value of the calculated time correlation is, the higher the time correlation is. In some embodiments, the time correlation threshold can be set to 0.6 or 0.8.

[0039] In some embodiments, identifying a user's electricity theft behavior based on abnormal events of an electric energy meter and the user's electricity usage data includes: identifying the user's electricity theft behavior based on at least one of a first indicator, a second indicator, and a third indicator; wherein the first indicator is used to characterize: the degree of change in the user's electricity usage data before and after the abnormal event of the electric energy meter occurs; the second indicator is used to characterize: the time correlation between the abnormal events of the electric energy meter; the third indicator is used to characterize: the time correlation between the abnormal events of the electric energy meter and the missing points of the user's electricity usage data.

[0040] As for the first indicator, it is understandable that if the abnormal event of the electric energy meter is caused by the user's electricity theft, then before and after the abnormal event of the electric energy meter occurs, the user's electricity consumption data will change significantly in the direction of reducing electricity consumption. Therefore, the user's electricity theft behavior can be identified based on the degree of change in the user's electricity consumption data.

[0041] As for the second indicator, it can be understood that the abnormal events of the electric energy meter include the power-off event of the electric energy meter and the cover-opening event of the electric energy meter. More specifically, the electric energy meter can record the power-off time and power-on time of the electric energy meter for the power-off event of the electric energy meter, and record the time of opening the cover / terminal button cover and closing the cover / terminal button cover of the electric energy meter for the cover-opening event of the electric energy meter. If the time correlation between the power-off event of the electric energy meter and the cover-opening event of the electric energy meter is relatively high, it can be considered that: after the power-off operation of the electric energy meter, the cover-opening operation is immediately performed, and the abnormal events of the electric energy meter occur so closely, which means that there is electricity theft. In some embodiments, the intermediate time between the power-off time and the power-on time is calculated for the power-off event of the electric energy meter to obtain the first intermediate time; the intermediate time between the cover-opening time and the cover-closing time is calculated for the cover-opening time (including the cover-opening and the terminal button cover-opening) time of the electric energy meter to obtain the second intermediate time; the similarity of the first intermediate time and the second intermediate time is calculated using the cosine similarity calculation formula, and it is used as the time correlation between the power-off event of the electric energy meter and the cover-opening time of the electric energy meter.

[0042] As for the third indicator, it can be understood that the missing point in the user's electricity consumption data means that at the corresponding time point, the electric energy meter fails to obtain valid user electricity consumption data. If the time correlation between the abnormal event of the electric energy meter and the missing point in the user's electricity consumption data is high (that is, the time of the abnormal event of the electric energy meter is close to the time of the missing point), it can be considered that the abnormal event of the electric energy meter is caused by electricity theft. In some embodiments, the cosine similarity calculation formula is used to calculate the similarity of the start time and the missing point time of the abnormal event of the electric energy meter, and it is used as the time correlation between the abnormal event of the electric energy meter and the missing point in the user's electricity consumption data.

[0043] In some embodiments, corresponding indicator thresholds can be set for the first indicator, the second indicator, and the third indicator respectively, and then the values ​​of the first indicator, the second indicator, and the third indicator are calculated based on abnormal events of the electric energy meter and the user's electricity consumption data. When the values ​​of the first indicator, the second indicator, and the third indicator all exceed the set indicator thresholds, it is considered that electricity theft has occurred, or when the value of one of the first indicator, the second indicator, and the third indicator exceeds the set threshold, it is considered that electricity theft has occurred.

[0044] In some embodiments, the degree of change in the user's electricity consumption data includes at least one of the following: the degree of change in electricity consumption, the degree of change in the line loss rate of the station area, and the degree of change in the line loss rate of the power supply line.

[0045] It is understandable that the degree of change in electricity consumption includes the degree of change in daily electricity consumption and the degree of change in monthly electricity consumption. Taking daily electricity consumption as an example, first select the day before the abnormal event of the electric energy meter occurs and the day after the abnormal event of the electric energy meter occurs, and calculate the absolute value of the daily electricity consumption difference for these two days to evaluate the degree of change in daily electricity consumption. When the absolute value of the daily electricity consumption difference is greater than the set daily electricity consumption change threshold, it is considered that electricity theft has occurred. Exemplarily, the first day before the abnormal event of the electric energy meter occurs is selected as Day1, and the first day after the abnormal event of the electric energy meter occurs is selected as Day2, and the calculation is performed. ,when When the daily power consumption change threshold is greater than the set value, it is considered that power theft has occurred. Regarding the setting of the daily power consumption change threshold, those skilled in the art can set it adaptively as needed, and the present invention does not limit this.

[0046] Similar to the degree of change in daily electricity consumption: for the substation, one day before the abnormal event of the electric energy meter occurs and one day after the abnormal event of the electric energy meter occurs are selected, and the absolute value of the difference in the line loss rate of the substation is calculated for these two days to evaluate the degree of change in the line loss rate of the substation. When the absolute value of the difference in the line loss rate of the substation is greater than the set threshold value of the change in the line loss rate of the substation, it is considered that electricity theft has occurred; for the power supply line, one day before the abnormal event of the electric energy meter occurs and one day after the abnormal event of the electric energy meter occurs are selected, and the absolute value of the difference in the line loss rate of the power supply line is calculated for these two days to evaluate the degree of change in the line loss rate of the power supply line. When the absolute value of the difference in the line loss rate of the power supply line is greater than the set threshold value of the change in the line loss rate of the power supply line, it is considered that electricity theft has occurred.

[0047] In some embodiments, based on at least one of the first indicator, the second indicator, and the third indicator, identifying the user's electricity stealing behavior includes: fusing at least two of the first indicator, the second indicator, and the third indicator to obtain a fused indicator; in response to the value of the fused indicator being greater than a set fused indicator threshold, determining that the corresponding user has engaged in electricity stealing behavior.

[0048] It is understandable that the first indicator ( )、Second indicator( )、The third indicator( ) are fused to obtain a fusion indicator. For example, the three indicators can be fused by the following formula (3) to obtain an integrated indicator (Integrated-Indicator, II):

[0049] in, , , Represent the weights of the three indicators respectively, and In some embodiments, after respectively calculating the first indicator, the second indicator and the third indicator, the indicator values ​​are substituted into formula (3) to calculate the value of the fusion indicator, and when the value of the fusion indicator is greater than the set fusion indicator threshold, it is determined that the corresponding user has committed electricity theft.

[0050] Figure 1 1 is a block diagram showing a hardware configuration of the apparatus 100 that can implement an embodiment of the present invention. Figure 1 As shown, the device 100 may include a processor 101 and a memory 102. The processor is configured to execute program instructions, and the memory is configured to store program instructions. When the program instructions are loaded and executed by the processor, the device executes the method for identifying the user's electricity theft behavior according to any of the above embodiments. Figure 1In the device 100, only the components related to this embodiment are shown. Therefore, it is obvious to those skilled in the art that the device 100 may also include Figure 1 The specific functions implemented by the memory 102 and the processor 101 of the device 100 provided in the embodiments of this specification can be explained in comparison with the aforementioned embodiments in this specification, and can achieve the technical effects of the aforementioned embodiments, so they will not be repeated here.

[0051] The apparatus 100 may correspond to a computing device having various processing functions. For example, the apparatus 100 may be implemented as various types of devices, such as a personal computer (PC), a server device, a mobile device, etc.

[0052] The processor 101 can control the operation of the device 100. For example, the processor 101 controls the operation of the device 100 by executing a program stored in the memory 102 on the device 100. The processor 101 can be implemented by a central processing unit (CPU), a graphics processing unit (GPU), an application processor (AP), an artificial intelligence processor chip (IPU), etc. provided in the device 100. However, the present invention is not limited to this. In this embodiment, the processor 101 can be implemented in any appropriate manner. For example, the processor 101 can take the form of a computer-readable medium, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller, etc., such as a microprocessor or a processor and a computer-readable program code (such as software or firmware) that can be executed by the (micro) processor.

[0053] The memory 102 may be used to store hardware of various data and instructions processed in the device 100. For example, the memory 102 may store processed data and data to be processed in the device 100. The memory 102 may store data that has been processed or to be processed by the processor 101, for example, user electricity consumption data collected by the user's electric energy meter, etc. In addition, the memory 102 may store applications, drivers, etc. to be driven by the device 100. For example: the memory 102 may store various programs related to the method for identifying the user's electricity theft behavior to be executed by the processor 101, etc. The memory 102 may be a DRAM, but the present invention is not limited thereto. The memory 102 may include at least one of a volatile memory or a non-volatile memory. The non-volatile memory may include a read-only memory (ROM), a programmable ROM (PROM), an electrically programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a flash memory, a phase change RAM (PRAM), a magnetic RAM (MRAM), a resistive RAM (RRAM), a ferroelectric RAM (FRAM), etc. The volatile memory may include dynamic RAM (DRAM), static RAM (SRAM), synchronous DRAM (SDRAM), PRAM, MRAM, RRAM, ferroelectric RAM (FeRAM), etc. In an embodiment, the memory 102 may include at least one of a hard disk drive (HDD), a solid state drive (SSD), a high-density flash memory (CF), a secure digital (SD) card, a micro secure digital (Micro-SD) card, a mini secure digital (Mini-SD) card, an extreme digital (xD) card, caches, or a memory stick.

[0054] The specific functions implemented by the memory 102 and the processor 101 of the device 100 provided in the implementation mode of this specification can be explained in comparison with the aforementioned implementation modes in this specification, and can achieve the technical effects of the aforementioned implementation modes, and will not be repeated here.

[0055] Example Embodiment-1 Figure 2 The following is a logic flow chart of a method for identifying a user's electricity theft behavior according to some embodiments of the present invention. The figure fully illustrates the process of identifying a user's electricity theft behavior based on abnormal events recorded by the user's electricity meter and the user's electricity consumption data, so as to better understand the spirit of the embodiments of the present invention.

[0056] like Figure 2As shown, in this exemplary embodiment, the method for identifying the user's electricity theft behavior includes: step 201, obtaining the user's electricity meter data through the electricity consumption information collection 2.0 system; step 202, performing data preprocessing on the user's electricity meter data; step 203, calculating the fusion index based on the preprocessed data; step 204, outputting whether the user has committed electricity theft.

[0057] As for step 201, it can be understood that the user information collection 2.0 system is a system specially designed for collecting information related to power users. The system can communicate with various types of electric energy meters to collect data generated by the electric energy meters, including record data of abnormal events of the electric energy meters and operation load curve data. Among them, the abnormal events of the electric energy meters include power-off events of the electric energy meters, opening button cover events of the electric energy meters, and opening cover events of the electric energy meters; the operation load curve data includes daily frozen indication data, voltage data, current data, power consumption data, line loss rate data of the substation area, and line loss rate data of the power supply line. There are three ways for the Electricity Information 2.0 Collection System to obtain the data from the electricity meters: (1) The electricity meters actively report data to the Electricity Information 2.0 Collection System periodically; (2) The Electricity Information 2.0 Collection System is configured with batch patrol and reading tasks to periodically read the data from the electricity meters; (3) A call interface program is developed in the Electricity Information 2.0 Collection System. When the collection system operator finds that the data from some users' electricity meters have not been fully obtained, or believes that there are problems with the data from some users' electricity meters, the collection system can use the call interface program to actively read the data from these users' electricity meters.

[0058] For step 202, preprocessing the user's electric energy meter data includes: (1) sorting the record data of abnormal events of the electric energy meter and the operating load curve data on a user-by-user basis, that is, classifying the abnormal events of the electric energy meter and the operating load curve data belonging to the same user as the electric energy meter data of the same user, thereby logically and physically concentrating the electric energy meter data, and then analyzing whether there is electricity theft for each user. (2) Data cleaning. During the collection, transmission and storage process, the electric energy meter data may be interfered by various factors, resulting in erroneous or non-compliant data. Data cleaning of user electric energy meter data mainly involves detecting abnormal and missing data in the original data through statistical methods and constructed business rule logic, and then processing the collected data through outlier processing and missing value filling methods to improve the accuracy, completeness and consistency of the data. In particular, the data cleaning work for abnormal events of the electricity meter includes determining whether the number of event records, the date and time of event records, and the duration of events are reasonable, and making corrections if the data is unreasonable; as well as analyzing the validity of the meter cover / terminal cover opening events and power-off events recorded by the electricity meter, so as to determine whether the events recorded by the electricity meter are accurate enough.

[0059] For step 203, it can be understood that for the pre-processed data, the index value can be calculated for the established evaluation index, and then multiple evaluation indexes can be fused to obtain a fused index. The evaluation indexes include: correlation analysis index between the meter cover opening event and the power-off event of the electric energy meter; daily average power consumption change index before and after the meter cover opening date; monthly average power consumption change index before and after the meter cover opening date; power-off event of the electric energy meter and the lack of consistency index of the load curve; line loss rate change index of the back-end area before and after the record date of the abnormal event of the electric energy meter, and line loss rate change index of the power supply line before and after the record date of the abnormal event of the electric energy meter. By assigning weights to each evaluation index, the evaluation index is fused to obtain a fused index.

[0060] As for step 204, it can be understood that, according to the value of the integrated index, the suspicion level of the user's electricity theft behavior can be determined, and the user with high suspicion can be determined as the electricity theft user.

[0061] Example Embodiment-2 Figure 3 The following is a logic flow chart of a method for identifying a user's electricity theft behavior according to some other embodiments of the present invention. The figure fully illustrates the process of identifying a user's electricity theft behavior based on abnormal events recorded by the user's electricity meter and the user's electricity consumption data, so as to better understand the spirit of the embodiments of the present invention.

[0062] like Figure 3 As shown, in this exemplary embodiment, the method for identifying the user's electricity theft behavior includes: step 301, obtaining the user's electricity meter data through the electricity consumption information collection 2.0 system; step 302, performing data preprocessing on the user's electricity meter data; step 303, identifying the mutation interval in the user's electricity consumption data, and determining whether there is an abnormal event of the electricity meter recorded in the mutation interval; step 304, outputting whether the user has committed electricity theft.

[0063] Step 301 and step 302 are respectively consistent with step 201 and step 202 in the above-mentioned exemplary embodiment-1, and will not be repeated here. In this exemplary embodiment, after preprocessing the data, the mutation interval in the user's electricity consumption data is first identified. The user's electricity consumption data is time series data, which can be input into the time series anomaly detection algorithm to identify mutation points. The mutation points constitute the mutation interval. It can be understood that the number of mutation intervals obtained can be multiple. Then, it is determined whether there is an abnormal event of the electric energy meter recorded in the mutation interval. If there is an abnormal event of the electric energy meter in the mutation interval, it can be considered that there is electricity theft.

[0064] Therefore, the present invention discloses a method for diagnosing and analyzing electricity theft by high and low voltage users based on electricity meter events. By conducting a fusion analysis of the abnormal event records of the electricity meter and the operating load curve data, the method can specifically identify electricity theft methods such as malicious opening of the electricity meter, installing a remote control inside the electricity meter, changing the resistance of the voltage and current sampling loop, corroding components with chemicals, and adding magnetic field interference, so as to accurately locate the electricity theft users and improve the efficiency of investigating and punishing electricity theft.

[0065] It should be noted that, for the purpose of simplicity, the present invention describes some methods and embodiments thereof as a series of actions and combinations thereof, but those skilled in the art will appreciate that the scheme of the present invention is not limited by the order of the described actions. Therefore, based on the disclosure or teaching of the present invention, those skilled in the art will appreciate that some of the steps therein may be performed in other orders or simultaneously. Further, those skilled in the art will appreciate that the embodiments described in the present invention may be regarded as optional embodiments, i.e., the actions or modules involved therein are not necessarily necessary for the implementation of one or some schemes of the present invention. In addition, depending on the different schemes, the present invention also has different emphases on the description of some embodiments. In view of this, those skilled in the art will appreciate that the parts not described in detail in a certain embodiment of the present invention may also refer to the relevant descriptions of other embodiments.

Claims

1. A method for identifying a user's electricity theft behavior, characterized in that: The user is equipped with a user electric energy meter to record abnormal events of the electric energy meter and the user's electricity consumption data; The abnormal event of the electric energy meter is used to characterize the potential electricity theft operation on the user's electric energy meter; the method comprises: In response to the user's electric energy meter recording an abnormal electric energy meter event, The user's electricity theft behavior is identified based on the abnormal event of the electric energy meter and the user's electricity usage data.

2. The method according to claim 1, characterized in that The abnormal event of the electric energy meter includes at least one of the following: a power failure event of the electric energy meter and a cover opening event of the electric energy meter.

3. The method according to claim 1, characterized in that The user electricity consumption data includes at least one of the following: daily frozen indication data, voltage data, current data, electricity consumption data, substation line loss rate data, and power supply line line loss rate data.

4. The method according to claim 2, characterized in that: The user electric energy meter includes a sensor for detecting the opening and closing actions of the cover of the user electric energy meter; the method includes: In response to the sensor detecting the cover opening action, the user electric energy meter records the electric energy meter cover opening event and the start time corresponding to the electric energy meter cover opening event; In response to the sensor detecting the cover closing action, the user electric energy meter records the end time corresponding to the electric energy meter cover opening event.

5. The method according to claim 1, characterized in that Identifying the user's electricity theft behavior according to the abnormal event of the electric energy meter and the user's electricity consumption data, including: In response to identifying a mutation point from the user's electricity usage data, it is determined that the corresponding user has engaged in electricity theft behavior.

6. The method according to claim 5, characterized in that In response to identifying a mutation point from the user's electricity usage data, determining that the corresponding user has electricity theft behavior includes: Calculating the time correlation between the mutation point and the abnormal event of the electric energy meter; In response to the time correlation being not less than a set time correlation threshold, it is determined that the corresponding user has committed electricity theft.

7. The method according to claim 1, characterized in that Identifying the user's electricity theft behavior according to the abnormal event of the electric energy meter and the user's electricity consumption data, including: Identifying the user's electricity theft behavior according to at least one of the first indicator, the second indicator, and the third indicator; The first indicator is used to characterize: the degree of change of the user's electricity consumption data before and after the abnormal event of the electric energy meter occurs; The second indicator is used to characterize: the time correlation between abnormal events of the electric energy meter; The third indicator is used to characterize: the time correlation between the abnormal event of the electric energy meter and the missing point of the user's electricity consumption data.

8. The method according to claim 7, characterized in that The degree of change of the user's electricity consumption data includes at least one of the following: the degree of change of electricity consumption, the degree of change of the line loss rate of the substation area, and the degree of change of the line loss rate of the power supply line.

9. The method according to claim 7, characterized in that: Identifying the user's electricity theft behavior according to at least one of the first indicator, the second indicator, and the third indicator includes: Fusing at least two of the first indicator, the second indicator, and the third indicator to obtain a fused indicator; In response to the value of the fusion indicator being greater than a set fusion indicator threshold, it is determined that the corresponding user has committed electricity theft.

10. A device for identifying a user's electricity theft behavior, characterized in that: include: a processor configured to execute program instructions; as well as A memory configured to store the program instructions, which, when loaded and executed by the processor, causes the apparatus to perform the method according to any one of claims 1 to 9.

Images