Electricity stealing detection method, device, equipment, medium and program product

Abstract

The embodiment of the invention provides an electricity larceny detection method, device and equipment, a medium and a program product. In the method, after detection data of a first preset number of continuous detection days are obtained, a detection index is generated according to the detection data. And generating a reference index according to the target generative adversarial network model. And finally, according to the detection index, the reference index and a preset index threshold, determining whether an electricity stealing behavior occurs. According to the scheme, electricity stealing detection is carried out through the detection index, the reference index and the preset index threshold, and the detection accuracy is effectively improved.

Description

Technical Field

[0001] The present application relates to the technical field of electricity theft detection, and in particular to an electricity theft detection method, device, equipment, medium and program product. Background Art

[0002] With the development of technology, electric meters have more and more functions. They can not only send electricity consumption remotely, but also send voltage, current, power factor and other data remotely. It is no longer necessary for staff to manually obtain these data, which greatly reduces the workload of staff. In order to determine whether there is electricity theft, it can also be achieved through the data sent by the electric meter.

[0003] In the prior art, workers usually manually determine whether there is abnormal data in the data transmitted by the meter based on their own experience. If there is abnormal data, it is determined that electricity theft has occurred.

[0004] In summary, the existing electricity theft detection method uses manual detection, resulting in low detection accuracy. Summary of the invention

[0005] The electricity theft detection method, device, equipment, medium and program product provided in the embodiments of the present application are used to solve the problem of low detection accuracy caused by manual detection in existing electricity theft detection methods.

[0006] In a first aspect, an embodiment of the present application provides a method for detecting electricity theft, which includes:

[0007] Acquire detection data of a first preset number of consecutive detection days, the detection data of each detection day including power consumption and line loss rate of a second preset number of days before the detection day, power consumption and line loss rate of the second preset number of days after the detection day, and power consumption, line loss rate and number of alarms on the detection day;

[0008] Generate detection indicators based on the detection data of each detection day, wherein the detection indicators include a power consumption indicator to be detected, a line loss indicator to be detected, and an alarm indicator to be detected;

[0009] Generate benchmark indicators according to the target generation adversarial network model, wherein the benchmark indicators include benchmark power consumption indicators, benchmark line loss indicators and benchmark alarm indicators; the target generation adversarial network model is a model trained based on power data when power theft occurs;

[0010] Whether electricity theft occurs is determined based on the detection index, the benchmark index and a preset index threshold.

[0011] In a possible implementation manner, generating a detection index according to the detection data of each detection day includes:

[0012] Determine the power consumption index to be detected according to the power consumption of the second preset number of days before each of the detection days, the power consumption of the second preset number of days after each of the detection days, and the power consumption of each of the detection days;

[0013] Determining the line loss index to be detected according to the line loss rate of the second preset number of days before each of the detection days, the line loss rate of the second preset number of days after each of the detection days, and the line loss rate of each of the detection days;

[0014] The number of alarms on each of the detection days is added together to obtain the alarm index to be detected.

[0015] In a possible implementation manner, determining the power consumption index to be detected according to the power consumption of the second preset number of days before each of the detection days, the power consumption of the second preset number of days after each of the detection days, and the power consumption of each of the detection days includes:

[0016] For each of the detection days, the power consumption slope of the detection day is calculated according to the power consumption of the second preset number of days before the detection day, the power consumption of the second preset number of days after the detection day, and the power consumption of the detection day;

[0017] For each of the detection days, if the power consumption slope of the day before the detection day exists, determine the power consumption decreasing mark of the detection day according to the power consumption slope of the detection day and the power consumption slope of the day before the detection day, the power consumption decreasing mark is 1 or 0, when the power consumption decreasing mark is 1, it indicates that the power consumption is decreasing, and when the power consumption decreasing mark is 0, it indicates that the power consumption is not decreasing;

[0018] For each of the detection days, if the power consumption slope of the day before the detection day does not exist, the power consumption decreasing mark of the detection day is determined to be 0;

[0019] The power consumption decreasing mark of each detection day is added together to obtain the power consumption index to be detected.

[0020] In a possible implementation manner, determining the line loss index to be detected according to the line loss rate of the second preset number of days before each of the detection days, the line loss rate of the second preset number of days after each of the detection days, and the line loss rate of each of the detection days includes:

[0021] For each of the detection days, calculating an average value of the first line loss rate of the detection day according to the line loss rates of the second preset number of days before the detection day and the line loss rate of the detection day;

[0022] For each of the detection days, calculating a second average value of the line loss rate for the detection day according to the line loss rates of a second preset number of days after the detection day and the line loss rate of the detection day;

[0023] For each of the detection days, determining a sub-line loss index for the detection day according to a first line loss rate average value and a second line loss rate average value of the detection day;

[0024] The line loss index to be detected is generated according to the sub-line loss index of each detection day.

[0025] In a possible implementation manner, the preset indicator threshold includes a power consumption day threshold, a line loss threshold, and an alarm threshold, and determining whether power theft occurs according to the detection indicator, the benchmark indicator, and the preset indicator threshold includes:

[0026] Generate a first detection result according to the detection index, the benchmark index, the preset power consumption weight, the preset line loss weight, and the preset alarm weight, wherein the first detection result is used to indicate whether electricity theft occurs;

[0027] Generate a second detection result according to the detection index, the power consumption day threshold, the line loss threshold and the alarm threshold, wherein the second detection result is used to indicate whether power theft occurs;

[0028] Whether electricity theft occurs is determined according to the first detection result and the second detection result.

[0029] In a possible implementation, the method further includes:

[0030] If the number of electricity thefts is 0 and it is determined that electricity theft has occurred, an alarm is issued, the number of electricity thefts is increased by one, and the current first electricity theft time is recorded.

[0031] In a possible implementation, the method further includes:

[0032] If the number of electricity thefts is 1, and it is determined that the electricity theft has not occurred, and the time between the current moment and the first electricity theft moment is less than or equal to a first preset number of days, the number of electricity thefts is updated to 0;

[0033] If the number of electricity thefts is 1, and it is determined that electricity theft has occurred, and the time between the current time and the first electricity theft time is equal to the first preset number of days, power restriction is performed, the number of electricity thefts is increased by one, and the current second electricity theft time is recorded.

[0034] In a possible implementation, the method further includes:

[0035] If the number of electricity thefts is 2, and it is determined that no electricity theft has occurred, and the time between the current moment and the second electricity theft moment is less than or equal to the second preset number of days, the number of electricity thefts is updated to 0;

[0036] If the number of electricity thefts is 2, and it is determined that electricity theft has occurred, and the time between the current moment and the second electricity theft moment is equal to the second preset number of days, a reporting process is performed and the number of electricity thefts is updated to 0.

[0037] In a second aspect, an embodiment of the present application provides an electricity theft detection device, comprising:

[0038] An acquisition module, used to acquire detection data of a first preset number of consecutive detection days, the detection data of each detection day including power consumption and line loss rate of a second preset number of days before the detection day, power consumption and line loss rate of the second preset number of days after the detection day, and power consumption, line loss rate and number of alarms on the detection day;

[0039] Processing modules for:

[0040] Generate detection indicators based on the detection data of each detection day, wherein the detection indicators include a power consumption indicator to be detected, a line loss indicator to be detected, and an alarm indicator to be detected;

[0041] Generate benchmark indicators according to the target generation adversarial network model, wherein the benchmark indicators include benchmark power consumption indicators, benchmark line loss indicators and benchmark alarm indicators; the target generation adversarial network model is a model trained based on power data when power theft occurs;

[0042] The detection module is used to determine whether electricity theft occurs based on the detection index, the benchmark index and a preset index threshold.

[0043] In a third aspect, an embodiment of the present application provides an electronic device, including:

[0044] Processor, memory, communication interface;

[0045] The memory is used to store executable instructions of the processor;

[0046] The processor is configured to execute the electricity theft detection method according to any one of the first aspects by executing the executable instructions.

[0047] In a fourth aspect, an embodiment of the present application provides a readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the electricity theft detection method described in any one of the first aspects.

[0048] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, is used to implement the electricity theft detection method described in any one of the second aspects.

[0049] The electricity theft detection method, device, equipment, medium and program product provided in the embodiment of the present application, after obtaining the detection data of the first preset number of consecutive detection days, generates a detection index according to the detection data. Then, an adversarial network model is generated according to the target to generate a benchmark index. Finally, it is determined whether the electricity theft behavior has occurred based on the detection index, the benchmark index and the preset index threshold. This solution performs electricity theft detection through detection indicators, benchmark indicators and preset index thresholds, which effectively improves the detection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0051] Figure 1 A schematic diagram of a flow chart of a first embodiment of a method for detecting electricity theft provided in the present application;

[0052] Figure 2 A schematic diagram of the flow chart of Embodiment 2 of the electricity theft detection method provided in this application;

[0053] Figure 3 A flow chart of the server processing electricity theft provided for this application;

[0054] Figure 4 A schematic diagram of the structure of an embodiment of an electricity theft detection device provided in this application;

[0055] Figure 5 A schematic diagram of the structure of an electronic device provided in this application.

[0056] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0057] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0058] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0059] With the development of technology, electric meters have the function of sending data remotely, and can send power consumption, voltage, current, line loss rate, etc., reducing the workload of staff.

[0060] In the prior art, the staff can use the data sent by the meter to determine whether there is abnormal data based on their own experience. If there is abnormal data, it is determined that electricity theft has occurred. Since the detection is done manually, the detection accuracy is low.

[0061] In view of the problems existing in the prior art, the inventors found in the process of studying the electricity theft detection method that when electricity theft occurs, some data will be abnormal. These data are used as detection data, and then the power consumption index to be detected, the line loss index to be detected and the alarm index to be detected are generated according to the detection data. The adversarial network model can be generated according to the target to generate the benchmark power consumption index, the benchmark line loss index and the benchmark alarm index. These three indicators are the indicators when electricity theft occurs. Then, according to the power consumption index to be detected, the line loss index to be detected, the alarm index to be detected, the benchmark power consumption index, the benchmark line loss index and the benchmark alarm index, a first detection result is generated; according to the power consumption index to be detected, the line loss index to be detected, the alarm index to be detected and the alarm threshold, a second detection result is generated; and then according to the first detection result and the second detection result, it is determined whether the electricity theft occurs, which effectively improves the detection accuracy. Based on the above-mentioned inventive concept, the electricity theft detection scheme in this application is designed.

[0062] The execution subject of the electricity theft detection method in the present application can be a server, or a computer, a terminal device, a smart meter, etc. This application does not limit it. The following description takes the server as an example.

[0063] The following is an example of an application scenario of the electricity theft detection method provided in this application.

[0064] For example, in this application scenario, the electric meter will send power data to the server, and the power data packet includes daily power consumption, line loss rate and number of alarms.

[0065] The server will perform a test once a day. During each test, the test data from m+n days before the day to m+1 days will be obtained, where n is the first preset number and m is the second preset number. Each of these days is called a test day. The test data for each test day includes the power consumption and line loss rate m days before the test day, the power consumption and line loss rate m days after the test day, and the power consumption, line loss rate and number of alarms on the test day.

[0066] Then, the server generates detection indicators based on the detection data of each detection day, and the detection indicators include power consumption indicators to be detected, line loss indicators to be detected, and alarm indicators to be detected.

[0067] The server generates benchmark indicators based on the target generative adversarial network model, and the benchmark indicators include benchmark electricity consumption indicators, baseline line loss indicators and benchmark alarm indicators. Since the target generative adversarial network model is a model trained based on the power data when electricity theft occurs, the benchmark indicators can be generated based on the generator in the target generative adversarial network model, and the benchmark indicators are highly similar to the indicators when electricity theft occurs.

[0068] Finally, the server generates a first detection result based on the power consumption index to be detected, the line loss index to be detected, the alarm index to be detected, the benchmark power consumption index, the benchmark line loss index and the benchmark alarm index; generates a second detection result based on the power consumption index to be detected, the line loss index to be detected, the alarm index to be detected and the alarm threshold; and determines whether electricity theft occurs based on the first detection result and the second detection result.

[0069] It should be noted that the above scenario is only an example of an application scenario provided by an embodiment of the present application. The embodiment of the present application does not limit the actual form of the various devices included in the scenario, nor does it limit the interaction method between the devices. In the specific application of the solution, it can be set according to actual needs.

[0070] The technical solution of the present application is described in detail below through specific embodiments. It should be noted that the following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0071] Figure 1 This is a flow chart of the first embodiment of the electricity theft detection method provided by the present application. The present embodiment of the present application describes the situation in which the server performs electricity theft detection based on the detection data, the target generation adversarial network model and the preset indicator threshold. The method in this embodiment can be implemented by software, hardware or a combination of software and hardware. Figure 1As shown, the electricity theft detection method specifically includes the following steps:

[0072] S101: Acquire detection data of a first preset number of consecutive detection days.

[0073] In this step, the server can perform a power theft detection once a day. During each detection, the detection data of the first preset number of detection days will be obtained, that is, the detection data from m+n days before the current day to m+1 days will be obtained, where n is the first preset number and m is the second preset number. Each of these days is called a detection day.

[0074] The detection data of each detection day includes the power consumption and line loss rate m days before the detection day, the power consumption and line loss rate m days after the detection day, and the power consumption, line loss rate and number of alarms on the detection day.

[0075] It should be noted that the first preset number can be 2, 3, 5, 10, etc., and the second preset number can be 3, 5, 7, 10, etc. The embodiment of the present application does not limit the first preset number and the second preset number, and can be determined according to actual conditions.

[0076] It should be noted that the meter will generate alarms during operation, such as alarms caused by voltage missing phase, voltage interruption phase, and current polarity reversal. The meter will count the number of alarms per day and send it to the server, which can obtain the number of alarms per day.

[0077] It should be noted that if some data are missing from the test data of a test day, they can be supplemented by interpolation.

[0078] S102: Generate detection indicators based on the detection data of each detection day.

[0079] In this step, after the server obtains the detection data of each detection day, in order to highlight whether the detection data is data formed by electricity theft, it is necessary to generate detection indicators based on the detection data of each detection day. The detection indicators include the power consumption indicator to be detected, the line loss indicator to be detected, and the alarm indicator to be detected.

[0080] The power consumption index to be detected can be generated according to the power consumption in the detection data, the line loss index to be detected can be generated according to the line loss rate in the detection data, and the alarm index to be detected can be generated according to the number of alarms in the detection data.

[0081] S103: Generate an adversarial network model according to the target and generate benchmark indicators.

[0082] In this step, in order to detect electricity theft, the server also needs to generate an adversarial network model based on the target and generate benchmark indicators.

[0083] Specifically, the target generative adversarial network model includes a generator and a discriminator. The generator is used to generate power consumption indicators, line loss indicators and alarm indicators. The discriminator is used to determine the probability that the power consumption indicator, line loss indicator and alarm indicator are corresponding indicators when power theft occurs.

[0084] Since the target generative adversarial network model is a model trained based on the power data when electricity theft occurs, the power consumption index, line loss index and alarm index generated by the generator are closer to the corresponding indicators when electricity theft occurs. Therefore, the random noise vector can be input into the generator to obtain the benchmark indicators, which include the benchmark power consumption index, the benchmark line loss index and the benchmark alarm index.

[0085] It should be noted that step S103 can be executed after step S101-step S102, or before step S101-step S102, or simultaneously with step S101-step S102. The embodiment of the present application limits the execution order of step S103 and step S101-step S102, which can be determined according to actual conditions.

[0086] S104: Determine whether electricity theft occurs based on the detection index, the benchmark index and the preset index threshold.

[0087] In this step, after obtaining the detection index and the benchmark index, the server determines whether electricity theft occurs according to the detection index, the benchmark index and the preset index threshold.

[0088] Specifically, the preset indicator thresholds include a power consumption day threshold, a line loss threshold, and an alarm threshold.

[0089] A first detection result is generated according to the detection index, the benchmark index, the preset power consumption weight, the preset line loss weight, and the preset alarm weight. The first detection result is used to indicate whether electricity theft occurs.

[0090] That is, first according to the formula , calculate the electricity theft similarity score, where, represents the electricity theft similarity score, Indicates the preset power consumption weight, Indicates the power consumption index to be tested. Indicates the benchmark electricity consumption index, Indicates the preset line loss weight, Indicates the line loss index to be detected, represents the baseline line loss index, Indicates the preset alarm weight. Indicates the alarm indicator to be detected. Indicates the baseline alarm indicator.

[0091] Then, it is determined whether the electricity theft similarity score is greater than a preset similarity threshold. If the electricity theft similarity score is greater than the preset similarity threshold, it means that the detection index and the reference index have a high similarity, and the detection index is very likely to be the corresponding index when the electricity theft behavior occurs, and a first detection result indicating that the electricity theft behavior has occurred is generated. If the electricity theft similarity score is less than or equal to the preset similarity threshold, it means that the detection index and the reference index have a low similarity, and the detection index is very likely to be the corresponding index when the electricity theft behavior has not occurred, and a first detection result indicating that the electricity theft behavior has not occurred is generated.

[0092] It should be noted that and is a vector, express and The magnitude of the vector difference of ; and is a numerical value, express and The absolute value of the difference; and is a numerical value, express and The absolute value of the difference.

[0093] It should be noted that the preset power consumption weight, preset line loss weight, and preset alarm weight can be 0.1, 0.2, 0.5, 0.7, 0.8, 0.9, etc., and the preset similarity threshold can be 5, 20, 50, 70, etc. The embodiment of the present application does not limit the preset power consumption weight, preset line loss weight, preset alarm weight, and preset similarity threshold, and can be determined according to actual conditions.

[0094] A second detection result is generated according to the detection index, the power consumption day threshold, the line loss threshold and the alarm threshold, and the second detection result is used to indicate whether the power theft behavior occurs.

[0095] Since the power consumption index, line loss index and alarm index will all be abnormal when electricity theft occurs, it can be determined whether the power consumption index to be detected is greater than the power consumption day threshold, whether the line loss index to be detected is greater than the line loss threshold, and whether the alarm index to be detected is greater than the alarm threshold.

[0096] If the power consumption index to be detected is greater than the power consumption days threshold, or the line loss index to be detected is greater than the line loss threshold, or the alarm index to be detected is greater than the alarm threshold, a second detection result indicating that power theft has occurred is generated.

[0097] If the power consumption index to be detected is less than or equal to the power consumption days threshold, the line loss index to be detected is less than or equal to the line loss threshold, and the alarm index to be detected is less than or equal to the alarm threshold, a second detection result indicating that power theft has not occurred is generated.

[0098] It should be noted that the electricity consumption day threshold can be 3, 5, 7, etc., the line loss threshold can be 2, 5, 8, etc., and the alarm threshold can be 5, 10, 20, etc. The embodiment of the present application does not limit the electricity consumption day threshold, line loss threshold and alarm threshold, and can be determined according to actual conditions.

[0099] It should be noted that, since the line loss index to be detected is a vector, the method for judging whether the power consumption index to be detected is greater than the power consumption days threshold is: judging whether the modulus of the power consumption index to be detected is greater than the power consumption days threshold.

[0100] After obtaining the first detection result and the second detection result, the server determines whether electricity theft occurs according to the first detection result and the second detection result.

[0101] That is, if both the first detection result and the second detection result indicate that electricity theft has occurred, it is determined that electricity theft has occurred. If one of the first detection result and the second detection result indicates that electricity theft has not occurred, it is determined that electricity theft has not occurred.

[0102] It should be noted that the server can also input the detection index into the discriminator in the target generative adversarial network model to obtain the probability of electricity theft. If the probability of electricity theft is greater than the preset probability threshold, a third detection result indicating that the electricity theft behavior has occurred is generated; if the probability of electricity theft is less than or equal to the preset probability threshold, a third detection result indicating that the electricity theft behavior has not occurred is generated.

[0103] Then, at least two detection results are selected from the first detection result, the second detection result and the third detection result. If the selected detection results all indicate that electricity theft has occurred, it is determined that electricity theft has occurred. If there is a detection result indicating that electricity theft has not occurred among the selected detection results, it is determined that electricity theft has not occurred.

[0104] The preset probability threshold may be 0.5, 0.7, 0.9, etc. The embodiment of the present application does not limit the preset probability threshold and may be determined according to actual conditions.

[0105] It should be noted that the server can also upload detection data, detection indicators, benchmark indicators, and the results of determining whether electricity theft has occurred to the chain, using the tamper-proof nature of blockchain to ensure the integrity and authenticity of data in all detection processes.

[0106] The electricity theft detection method provided in this embodiment generates detection indicators based on the detection data after acquiring the detection data of the first preset number of consecutive detection days. Then, an adversarial network model is generated according to the target to generate a benchmark indicator. Finally, it is determined whether the electricity theft behavior occurs based on the detection indicator, the benchmark indicator and the preset indicator threshold. Compared with the manual detection in the prior art, this solution performs electricity theft detection through detection indicators, benchmark indicators and preset indicator thresholds, which effectively improves the detection accuracy.

[0107] Figure 2 This is a flow chart of the second embodiment of the electricity theft detection method provided by the present application. Based on the above embodiment, the present embodiment of the present application describes the situation where the server generates detection indicators based on detection data. Figure 2 As shown, the electricity theft detection method specifically includes the following steps:

[0108] S201: Determine the power consumption index to be tested based on the power consumption of the second preset number of days before each test day, the power consumption of the second preset number of days after each test day, and the power consumption of each test day.

[0109] In this step, after obtaining the detection data, the server determines the power consumption index to be detected based on the power consumption of the second preset number of days before each detection day, the power consumption of the second preset number of days after each detection day, and the power consumption of each detection day.

[0110] Specifically, for each detection day, the power consumption slope of the detection day is calculated based on the power consumption of the second preset number of days before the detection day, the power consumption of the second preset number of days after the detection day, and the power consumption on the detection day.

[0111] According to the formula Calculate the power consumption slope, where, , , i represents the i-th detection day among n detection days; represents the power consumption slope on the i-th detection day; n represents the first preset number; m represents the second preset number; im represents the m-th day before the i-th detection day; i+m represents the m-th day after the i-th detection day; , hour, represents the first Daily electricity consumption; , hour, represents the number of days after the i-th detection day. Daily electricity consumption; hour, Represents the electricity consumption on the i-th detection day.

[0112] For each test day, if the electricity consumption slope of the day before the test day exists, the electricity consumption decreasing flag of the test day is determined according to the electricity consumption slope of the test day and the electricity consumption slope of the day before the test day. The electricity consumption decreasing flag is 1 or 0. When the electricity consumption decreasing flag is 1, it indicates that the electricity consumption is decreasing. When the electricity consumption decreasing flag is 0, it indicates that the electricity consumption is not decreasing.

[0113] Specifically, according to the formula , determine the decreasing mark of electricity consumption. Where i represents the i-th test day among n test days; represents the power consumption slope on the i-th detection day; n represents the first preset number; Indicates the decreasing mark of electricity consumption on the i-th detection day.

[0114] For each test day, if the power consumption slope of the day before the test day does not exist, it means that the test day is the first test day, and the power consumption decrease mark of the test day is determined to be 0.

[0115] The decreasing mark of electricity consumption on each test day is added together to obtain the electricity consumption index to be tested.

[0116] S202: Determine the line loss index to be detected according to the line loss rate of the second preset number of days before each detection day, the line loss rate of the second preset number of days after each detection day, and the line loss rate of each detection day.

[0117] In this step, after obtaining the detection data, the server determines the line loss index to be detected based on the line loss rate of the second preset number of days before each detection day, the line loss rate of the second preset number of days after each detection day, and the line loss rate of each detection day.

[0118] Specifically, for each test day, the first line loss rate average value for the test day is calculated based on the line loss rates of the second preset number of days before the test day and the line loss rate of the test day. That is, the average value of the line loss rates of the test day and the second preset number of days before the test day is taken as the first line loss rate average value for the test day.

[0119] For each test day, the average value of the second line loss rate for that test day is calculated based on the line loss rates of the second preset number of days after the test day and the line loss rate on the test day; that is, the average value of the line loss rates on the test day and the second preset number of days thereafter is taken as the average value of the second line loss rate for that test day.

[0120] For each test day, the sub-line loss index of the test day is determined according to the first line loss rate average value and the second line loss rate average value of the test day.

[0121] According to the formula , determine the sub-line loss index. Where i represents the i-th test day among n test days; represents the sub-line loss index on the i-th detection day; n represents the first preset number; represents the average first line loss rate on the i-th detection day; represents the average value of the second line loss rate on the i-th detection day; T represents the preset line loss average threshold.

[0122] It should be noted that the preset line loss average threshold may be 1%, 2%, 5%, etc. The embodiment of the present application is not limited by the preset line loss average threshold and may be determined according to actual conditions.

[0123] The line loss index to be detected is generated according to the sub-line loss index of each detection day. That is, the sub-line loss index of each detection day is arranged in order from morning to night to form a vector, and then the line loss index to be detected is obtained.

[0124] S203: Add the number of alarms on each detection day to obtain an alarm index to be detected.

[0125] In this step, after obtaining the detection data, the server adds up the number of alarms on each detection day to obtain the alarm index to be detected.

[0126] It should be noted that the execution order of step S201, step S202 and step S203 can be: execute step S201 first, then execute step S202, and finally execute step S203; it can also be: execute step S202 first, then execute step S201, and finally execute step S203; it can also be: execute step S203 first, then execute step S202, and finally execute step S201; it can also be: step S201, step S202 and step S203 are executed simultaneously, etc. The embodiment of the present application does not limit the execution order of step S201, step S202 and step S203, which can be determined according to actual conditions.

[0127] The electricity theft detection method provided in this embodiment generates an electricity consumption index to be detected by detecting the electricity consumption in the data, generates a line loss index to be detected according to the line loss rate in the detection data, and generates an alarm index to be detected according to the number of alarms in the detection data, so that the three indicators can better highlight whether electricity theft occurs, thereby improving the accuracy of electricity theft detection.

[0128] The following describes a third embodiment of the electricity theft detection method provided by the present application, in which a server handles electricity theft according to the number of electricity thefts and whether the electricity theft occurs.

[0129] After the server determines whether electricity theft occurs, in order to process the electricity theft, it needs to obtain the number of electricity thefts, the time between the current moment and the first electricity theft moment, and the time between the current moment and the second electricity theft moment.

[0130] The first electricity-stealing time is the time when the electricity-stealing time is updated to 1 o'clock, and the second electricity-stealing time is the time when the electricity-stealing time is updated to 2 o'clock.

[0131] If the server determines that the number of electricity thefts is 0 and that electricity theft has occurred, it issues an alarm, increases the number of electricity thefts by one, and records the current first electricity theft time.

[0132] The alarm may be issued by sending an alarm message to the user's terminal device.

[0133] If the server determines that the number of electricity thefts is 0 and that no electricity theft has occurred, no action will be taken.

[0134] If the server determines that the number of electricity thefts is 1 and that no electricity theft has occurred, and the time between the current time and the first electricity theft time is less than or equal to the first preset number of days, indicating that the user did not steal electricity on one day within the first preset number of days of the first electricity theft, the number of electricity thefts is updated to 0.

[0135] If the server determines that the number of electricity thefts is 1 and that electricity theft has occurred, and the time between the current time and the first electricity theft time is equal to the first preset number of days, it means that the user has committed electricity theft every day within the first preset number of days of the first electricity theft. Therefore, regardless of whether electricity theft occurs on that day, power restriction is implemented, the number of electricity thefts is increased by one, and the current second electricity theft time is recorded.

[0136] If the server determines that the number of electricity thefts is 1 and that electricity theft has occurred, and the time between the current time and the first electricity theft time is less than the first preset number of days, in order to determine whether the user has a day without electricity theft within the first preset number of days after the first electricity theft, no processing is performed at this time.

[0137] It should be noted that the first preset number of days can be 2, 3, 4, etc. The embodiment of the present application does not limit the first preset number of days and can be determined according to actual conditions.

[0138] If the server determines that the number of electricity thefts is 2 and that no electricity theft has occurred, and the time between the current time and the second electricity theft time is less than or equal to the second preset number of days, it means that the user did not steal electricity on one day within the second preset number of days of the second electricity theft, and the number of electricity thefts is updated to 0.

[0139] If the server determines that the number of electricity thefts is 2 and that electricity theft has occurred, and the time between the current time and the second electricity theft time is equal to the second preset number of days, it means that the user has stolen electricity every day within the second preset number of days of the second electricity theft. Therefore, regardless of whether electricity theft occurs on that day, the server will report it and update the number of electricity thefts to 0.

[0140] If the server determines that the number of electricity thefts is 2 and that electricity theft has occurred, and the time between the current moment and the second electricity theft moment is less than the second preset number of days, in order to determine whether the user has a day without electricity theft within the second preset number of days after the second electricity theft, no processing is performed at this time.

[0141] It should be noted that the second preset number of days can be 4, 5, 6, etc. The embodiment of the present application does not limit the second preset number of days and can be determined according to actual conditions.

[0142] For example, Figure 3 The flowchart of the server processing electricity theft provided by this application is as follows: Figure 3 As shown, before the first electricity theft detection, the server determines the number of electricity thefts to be 0. If it is determined that no electricity theft has occurred during the subsequent electricity theft detection, the number of electricity thefts will not be modified, that is, the number of electricity thefts will be kept at 0. If it is determined that electricity theft has occurred, the number of electricity thefts will be increased by one, that is, the number of electricity thefts will be determined to be 1, and an alarm will be issued.

[0143] The number of electricity thefts is determined to be 1, and after an alarm is issued, electricity theft detection is performed within the first preset number of days. If it is determined that no electricity theft has occurred within the first preset number of days, the number of electricity thefts is determined to be 0. If it is determined that no electricity theft has occurred within the first preset number of days, that is, the user has committed electricity theft every day within the first preset number of days of the first electricity theft, the number of electricity thefts is increased by one, that is, the number of electricity thefts is determined to be 2, and power restriction is performed.

[0144] After the number of electricity thefts is determined to be 2 and power is restricted, electricity theft detection is performed within the second preset number of days. If it is determined that no electricity theft has occurred within the second preset number of days, the number of electricity thefts is determined to be 0. If it is determined that no electricity theft has occurred within the second preset number of days, that is, the user has committed electricity theft every day within the second preset number of days of the second electricity theft, the number of electricity thefts is determined to be 0 and reported.

[0145] The electricity theft detection method provided in this embodiment can effectively reduce the occurrence of electricity theft and improve electricity safety through multiple processing methods for electricity theft.

[0146] The following describes the training process of the target generative adversarial network model through the fourth embodiment of the electricity theft detection method provided by the present application.

[0147] First, a plurality of initial training data are obtained, each of which is power data corresponding to when the electricity theft occurs, including detection data of a first preset number of consecutive detection days.

[0148] Then, for each initial training data, target training data corresponding to the initial training data is generated according to the detection data in the initial training data, and the target training data includes a target power consumption index, a target line loss index and a target alarm index.

[0149] It should be noted that the process of generating target training data is similar to that of the second embodiment and will not be further described here.

[0150] A generator in the initial generative adversarial network model is used to generate multiple simulated training data, and the simulated training data includes simulated power consumption indicators, simulated line loss indicators, and simulated alarm indicators.

[0151] Some target training data will be selected as true data, and each simulated training data will be used as false data to train the discriminator in the initial generative adversarial network model. After the true data and false data are used up, the training of the discriminator will be stopped.

[0152] Then the generator is trained. The training process is as follows: the random noise vector is input into the generator to obtain the power consumption index to be identified, the line loss index to be identified, and the alarm index to be identified. The power consumption index to be identified, the line loss index to be identified, and the alarm index to be identified are input into the discriminator to obtain the recognition probability. The loss value is calculated according to the recognition probability, and the generator is updated according to the loss value.

[0153] After the generator is trained for the first preset number of times, the training of the generator is stopped.

[0154] The overall training times are updated, that is, the overall training times are increased by one, and it is determined whether the updated overall training times are equal to the second preset training times. The initial value of the overall training times is 0.

[0155] If the updated overall training number is less than the second preset training number, re-use the generator to generate multiple simulated training data, re-select some target training data as true data, and use each simulated training data as false data to train the discriminator; after the true data and false data are used up, stop training the discriminator. Re-train the generator for the first preset training number, and re-update the overall training number. If the updated overall training number is less than the second preset training number, repeat this process until the updated overall training number is equal to the second preset training number, and obtain the target generative adversarial network model.

[0156] It should be noted that the first preset number of training times and the second preset number of training times can be 5000, 10000, 1000000, etc. The embodiment of the present application does not limit the first preset number of training times and the second preset number of training times, which can be determined according to actual conditions.

[0157] The electricity theft detection method provided in this embodiment generates target training data by using the power data when the electricity theft occurs, and then performs model training based on the target training data to obtain a target generative adversarial network model, which effectively improves the similarity between the indicators generated by the generator in the target generative adversarial network model and the indicators when the electricity theft occurs, and also improves the discrimination accuracy of the discriminator.

[0158] The following is an embodiment of the device of the present application, which can be used to execute the embodiment of the method of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method of the present application.

[0159] Figure 4 A schematic diagram of the structure of an embodiment of the electricity theft detection device provided in this application; Figure 4 As shown, the electricity theft detection device 40 includes:

[0160] An acquisition module 41 is used to acquire detection data of a first preset number of consecutive detection days, wherein the detection data of each detection day includes power consumption and line loss rate of a second preset number of days before the detection day, power consumption and line loss rate of the second preset number of days after the detection day, and power consumption, line loss rate and number of alarms on the detection day;

[0161] The processing module 42 is used to:

[0162] Generate detection indicators based on the detection data of each detection day, wherein the detection indicators include a power consumption indicator to be detected, a line loss indicator to be detected, and an alarm indicator to be detected;

[0163] Generate benchmark indicators according to the target generation adversarial network model, wherein the benchmark indicators include benchmark power consumption indicators, benchmark line loss indicators and benchmark alarm indicators; the target generation adversarial network model is a model trained based on power data when power theft occurs;

[0164] The detection module 43 is used to determine whether electricity theft occurs according to the detection index, the benchmark index and a preset index threshold.

[0165] Furthermore, the processing module 42 is specifically configured to:

[0166] Determine the power consumption index to be detected according to the power consumption of the second preset number of days before each of the detection days, the power consumption of the second preset number of days after each of the detection days, and the power consumption of each of the detection days;

[0167] Determining the line loss index to be detected according to the line loss rate of the second preset number of days before each of the detection days, the line loss rate of the second preset number of days after each of the detection days, and the line loss rate of each of the detection days;

[0168] The number of alarms on each of the detection days is added together to obtain the alarm index to be detected.

[0169] Furthermore, the processing module 42 is further configured to:

[0170] For each of the detection days, the power consumption slope of the detection day is calculated according to the power consumption of the second preset number of days before the detection day, the power consumption of the second preset number of days after the detection day, and the power consumption of the detection day;

[0171] For each of the detection days, if the power consumption slope of the day before the detection day exists, determine the power consumption decreasing mark of the detection day according to the power consumption slope of the detection day and the power consumption slope of the day before the detection day, the power consumption decreasing mark is 1 or 0, when the power consumption decreasing mark is 1, it indicates that the power consumption is decreasing, and when the power consumption decreasing mark is 0, it indicates that the power consumption is not decreasing;

[0172] For each of the detection days, if the power consumption slope of the day before the detection day does not exist, the power consumption decreasing mark of the detection day is determined to be 0;

[0173] The power consumption decreasing mark of each detection day is added together to obtain the power consumption index to be detected.

[0174] Furthermore, the processing module 42 is further configured to:

[0175] For each of the detection days, calculating an average value of the first line loss rate of the detection day according to the line loss rates of the second preset number of days before the detection day and the line loss rate of the detection day;

[0176] For each of the detection days, calculating a second average value of the line loss rate for the detection day according to the line loss rates of a second preset number of days after the detection day and the line loss rate of the detection day;

[0177] For each of the detection days, determining a sub-line loss index for the detection day according to a first line loss rate average value and a second line loss rate average value of the detection day;

[0178] The line loss index to be detected is generated according to the sub-line loss index of each detection day.

[0179] Furthermore, the preset indicator thresholds include a power consumption day threshold, a line loss threshold and an alarm threshold. The detection module 43 is specifically used to:

[0180] Generate a first detection result according to the detection index, the benchmark index, the preset power consumption weight, the preset line loss weight, and the preset alarm weight, wherein the first detection result is used to indicate whether electricity theft occurs;

[0181] Generate a second detection result according to the detection index, the power consumption day threshold, the line loss threshold and the alarm threshold, wherein the second detection result is used to indicate whether power theft occurs;

[0182] Whether electricity theft occurs is determined according to the first detection result and the second detection result.

[0183] Furthermore, the processing module 42 is also used for:

[0184] If the number of electricity thefts is 0 and it is determined that electricity theft has occurred, an alarm is issued, the number of electricity thefts is increased by one, and the current first electricity theft time is recorded.

[0185] Furthermore, the processing module 42 is also used for:

[0186] If the number of electricity thefts is 1, and it is determined that the electricity theft has not occurred, and the time between the current moment and the first electricity theft moment is less than or equal to a first preset number of days, the number of electricity thefts is updated to 0;

[0187] If the number of electricity thefts is 1, and it is determined that electricity theft has occurred, and the time between the current time and the first electricity theft time is equal to the first preset number of days, power restriction is performed, the number of electricity thefts is increased by one, and the current second electricity theft time is recorded.

[0188] Furthermore, the processing module 42 is also used for:

[0189] If the number of electricity thefts is 2, and it is determined that no electricity theft has occurred, and the time between the current moment and the second electricity theft moment is less than or equal to the second preset number of days, the number of electricity thefts is updated to 0;

[0190] If the number of electricity thefts is 2, and it is determined that electricity theft has occurred, and the time between the current moment and the second electricity theft moment is equal to the second preset number of days, a reporting process is performed and the number of electricity thefts is updated to 0.

[0191] The electricity theft detection device provided in this embodiment is used to implement the technical solution in any of the above method embodiments. Its implementation principle and technical effects are similar and will not be described in detail here.

[0192] Figure 5 This is a schematic diagram of the structure of an electronic device provided in this application. Figure 5 As shown, the electronic device 50 includes:

[0193] Processor 51, memory 52, and communication interface 53;

[0194] The memory 52 is used to store executable instructions of the processor 51;

[0195] The processor 51 is configured to execute the technical solution in any of the aforementioned method embodiments by executing the executable instructions.

[0196] Optionally, the memory 52 may be independent or integrated with the processor 51 .

[0197] Optionally, when the memory 52 is a device independent of the processor 51, the electronic device 50 may further include:

[0198] The bus 54 , the memory 52 and the communication interface 53 are connected to the processor 51 via the bus 54 and communicate with each other. The communication interface 53 is used to communicate with other devices.

[0199] Optionally, the communication interface 53 may be implemented by a transceiver. The communication interface is used to implement communication between the database access device and other devices (such as a client, a read-write library, and a read-only library). The memory may include a random access memory (RAM) and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage.

[0200] The bus 54 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0201] The above-mentioned processor can be a general-purpose processor, including a central processing unit CPU, a network processor (NP), etc.; it can also be a digital signal processor DSP, an application-specific integrated circuit ASIC, a field programmable gate array FPGA or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0202] The electronic device is used to execute the technical solution in any of the aforementioned method embodiments, and its implementation principle and technical effect are similar and will not be repeated here.

[0203] An embodiment of the present application also provides a readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the technical solution provided by any of the aforementioned method embodiments is implemented.

[0204] An embodiment of the present application also provides a computer program product, including a computer program, which is used to implement the technical solution provided by any of the aforementioned method embodiments when executed by a processor.

[0205] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

[0206] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for detecting electricity theft, characterized in that: include: Acquire detection data of a first preset number of consecutive detection days, the detection data of each detection day including power consumption and line loss rate of a second preset number of days before the detection day, power consumption and line loss rate of the second preset number of days after the detection day, and power consumption, line loss rate and number of alarms on the detection day; Generate detection indicators based on the detection data of each detection day, wherein the detection indicators include a power consumption indicator to be detected, a line loss indicator to be detected, and an alarm indicator to be detected; Generate an adversarial network model according to the target, generate benchmark indicators, and the benchmark indicators include a benchmark power consumption indicator, a benchmark line loss indicator, and a benchmark alarm indicator; The target generative adversarial network model is a model trained based on the power data when the power theft occurs; Whether electricity theft occurs is determined based on the detection index, the benchmark index and a preset index threshold.

2. The method according to claim 1, characterized in that Generating detection indicators according to the detection data of each detection day includes: Determine the power consumption index to be detected according to the power consumption of the second preset number of days before each of the detection days, the power consumption of the second preset number of days after each of the detection days, and the power consumption of each of the detection days; Determining the line loss index to be detected according to the line loss rate of the second preset number of days before each of the detection days, the line loss rate of the second preset number of days after each of the detection days, and the line loss rate of each of the detection days; The number of alarms on each of the detection days is added together to obtain the alarm index to be detected.

3. The method according to claim 2, characterized in that Determining the power consumption index to be detected according to the power consumption of the second preset number of days before each of the detection days, the power consumption of the second preset number of days after each of the detection days, and the power consumption of each of the detection days includes: For each of the detection days, the power consumption slope of the detection day is calculated according to the power consumption of the second preset number of days before the detection day, the power consumption of the second preset number of days after the detection day, and the power consumption of the detection day; For each of the detection days, if the power consumption slope of the day before the detection day exists, determine the power consumption decreasing mark of the detection day according to the power consumption slope of the detection day and the power consumption slope of the day before the detection day, the power consumption decreasing mark is 1 or 0, when the power consumption decreasing mark is 1, it indicates that the power consumption is decreasing, and when the power consumption decreasing mark is 0, it indicates that the power consumption is not decreasing; For each of the detection days, if the power consumption slope of the day before the detection day does not exist, the power consumption decreasing mark of the detection day is determined to be 0; The power consumption decreasing mark of each detection day is added together to obtain the power consumption index to be detected.

4. The method according to claim 2, characterized in that Determining the line loss index to be detected according to the line loss rate of the second preset number of days before each of the detection days, the line loss rate of the second preset number of days after each of the detection days, and the line loss rate of each of the detection days includes: For each of the detection days, calculating an average value of the first line loss rate of the detection day according to the line loss rates of the second preset number of days before the detection day and the line loss rate of the detection day; For each of the detection days, calculating a second average value of the line loss rate for the detection day according to the line loss rates of a second preset number of days after the detection day and the line loss rate of the detection day; For each of the detection days, determining a sub-line loss index for the detection day according to a first line loss rate average value and a second line loss rate average value of the detection day; The line loss index to be detected is generated according to the sub-line loss index of each detection day.

5. The method according to claim 1, characterized in that The preset indicator thresholds include a power consumption day threshold, a line loss threshold, and an alarm threshold. The determining whether power theft occurs according to the detection indicator, the benchmark indicator, and the preset indicator thresholds includes: Generate a first detection result according to the detection index, the benchmark index, the preset power consumption weight, the preset line loss weight, and the preset alarm weight, wherein the first detection result is used to indicate whether electricity theft occurs; Generate a second detection result according to the detection index, the power consumption day threshold, the line loss threshold and the alarm threshold, wherein the second detection result is used to indicate whether power theft occurs; Whether electricity theft occurs is determined according to the first detection result and the second detection result.

6. The method according to any one of claims 1 to 5, characterized in that: The method further comprises: If the number of electricity thefts is 0 and it is determined that electricity theft has occurred, an alarm is issued, the number of electricity thefts is increased by one, and the current first electricity theft time is recorded.

7. The method according to claim 6, characterized in that The method further comprises: If the number of electricity thefts is 1, and it is determined that the electricity theft has not occurred, and the time between the current moment and the first electricity theft moment is less than or equal to a first preset number of days, the number of electricity thefts is updated to 0; If the number of electricity thefts is 1, and it is determined that electricity theft has occurred, and the time between the current time and the first electricity theft time is equal to the first preset number of days, power restriction is performed, the number of electricity thefts is increased by one, and the current second electricity theft time is recorded.

8. The method according to claim 7, characterized in that The method further comprises: If the number of electricity thefts is 2, and it is determined that no electricity theft has occurred, and the time between the current moment and the second electricity theft moment is less than or equal to the second preset number of days, the number of electricity thefts is updated to 0; If the number of electricity thefts is 2, and it is determined that electricity theft has occurred, and the time between the current moment and the second electricity theft moment is equal to the second preset number of days, a reporting process is performed and the number of electricity thefts is updated to 0.

9. An electricity theft detection device, characterized in that: include: An acquisition module, used to acquire detection data of a first preset number of consecutive detection days, the detection data of each detection day including power consumption and line loss rate of a second preset number of days before the detection day, power consumption and line loss rate of the second preset number of days after the detection day, and power consumption, line loss rate and number of alarms on the detection day; Processing modules for: Generate detection indicators based on the detection data of each detection day, wherein the detection indicators include a power consumption indicator to be detected, a line loss indicator to be detected, and an alarm indicator to be detected; Generate an adversarial network model according to the target, generate benchmark indicators, and the benchmark indicators include a benchmark power consumption indicator, a benchmark line loss indicator, and a benchmark alarm indicator; The target generative adversarial network model is a model trained based on the power data when the power theft occurs; The detection module is used to determine whether electricity theft occurs based on the detection index, the benchmark index and a preset index threshold.

10. An electronic device, characterized in that: include: Processor, memory, communication interface; The memory is used to store executable instructions of the processor; The processor is configured to execute the electricity theft detection method according to any one of claims 1 to 8 by executing the executable instructions.

11. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the electricity theft detection method according to any one of claims 1 to 8 is implemented.

12. A computer program product, characterized in that The invention comprises a computer program, wherein when the computer program is executed by a processor, the computer program is used to implement the electricity theft detection method according to any one of claims 1 to 8.

Images