Electricity stealing behavior identification method and device, electronic equipment and computer readable storage medium

Abstract

Embodiments of the present application provide a power stealing behavior identification method and device, electronic equipment and computer readable storage medium, which belong to the field of data processing. The method comprises: determining a user corresponding to collected information and a power consumption scene to which the user belongs according to an abnormal user identifier of the collected information, determining all collected information of the user in a power stealing interval from a cache as a target information set, determining a power stealing analysis model corresponding to the power consumption scene, analyzing power consumption of the user in the power stealing interval according to the power stealing analysis model and a feature quantity of each piece of collected information in the target information set, obtaining a power stealing analysis value of the user, and then judging whether power consumption of the user belongs to a power stealing behavior according to the power stealing analysis value, thereby realizing real-time monitoring and identification, using a power consumption scene-specific power stealing analysis model for power consumption analysis, realizing accurate analysis, and improving the accuracy of power stealing identification to improve the identification effect.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of data processing, in particular to a method and device for identifying electricity stealing behavior, an electronic device and a computer readable storage medium. BACKGROUND

[0002] Electricity stealing refers to the behavior of illegally occupying electric energy, not paying or paying less electricity fees, and using illegal means to not meter or meter less electricity. Due to the impact of electricity stealing behavior on the economic benefits and investment returns of power companies, identifying user electricity stealing behavior and implementing anti-electricity stealing work are urgent problems to be solved.

[0003] Traditional methods for identifying electricity stealing behavior include manual regular inspection and system identification. The manual regular inspection method is high in cost and low in efficiency. The system identification method identifies electricity stealing behavior by identifying data according to collected data, fraud events and working condition information, but still has the problem of poor identification effect. SUMMARY

[0004] Therefore, the purpose of the present application is to provide a method and device for identifying electricity stealing behavior, an electronic device and a computer readable storage medium, which can improve the problem of poor identification effect of the current method for identifying electricity stealing behavior.

[0005] To achieve the above purpose, the technical solutions adopted by the embodiments of the present application are as follows.

[0006] In a first aspect, the present application provides a method for identifying electricity stealing behavior, applied to an electronic device, wherein a relationship table is stored on the electronic device, and the relationship table records the corresponding relationship between a user identifier, a user and a user scenario. The method comprises the following steps:

[0007] When it is determined that the currently consumed collected information in the message queue is abnormal, the user corresponding to the collected information and the power consumption scenario to which the user belongs are determined according to the user identifier of the collected information, wherein the collected information is the electricity stealing related behavior information of a certain user monitored and collected by a monitoring device in real time;

[0008] All collected information of the user in the electricity stealing interval is determined from the cache as a target information set, and a electricity stealing analysis model corresponding to the power consumption scenario is determined;

[0009] According to the electricity stealing analysis model and the feature quantity of each piece of collected information in the target information set, the electricity consumption of the user in the electricity stealing interval is analyzed to obtain an electricity stealing analysis value of the user, wherein the feature quantity is used to represent the electricity stealing related behavior to which the collected information belongs.

[0010] According to the electricity stealing analysis value, it is determined whether the electricity consumption of the user belongs to electricity stealing behavior.

[0011] Further, the electricity stealing analysis model comprises a selected feature quantity combination;

[0012] The step of analyzing the electricity consumption of the user in the electricity stealing interval according to the electricity stealing analysis model and the feature quantity of each collected information in the target information set, to obtain the electricity stealing analysis value of the user, comprises:

[0013] Analyzing each collected information in the target information set to obtain the corresponding feature quantity of each collected information;

[0014] Extracting the feature quantity conforming to the selected feature quantity combination from all the feature quantities as a selected feature quantity, and calculating the electricity stealing weight of each selected feature quantity;

[0015] Based on all the selected feature quantities and the electricity stealing weights, analyzing the electricity consumption of the user in the electricity stealing interval to obtain the electricity stealing analysis value of the user.

[0016] Further, the electricity stealing analysis model further comprises a calculation weight of each selected feature quantity, and the step of calculating the electricity stealing weight of each selected feature quantity comprises:

[0017] Querying the weighting factor of each selected feature quantity from a preset weighting factor table;

[0018] For each selected feature quantity, the product of the calculation weight of the selected feature quantity and the weighting factor is taken as the electricity stealing weight of the selected feature quantity.

[0019] Further, the step of analyzing the electricity consumption of the user in the electricity stealing interval based on all the selected feature quantities and the electricity stealing weights to obtain the electricity stealing analysis value of the user comprises:

[0020] For each selected feature quantity, analyzing and obtaining the correlation coefficient between the selected feature quantity and each remaining selected feature quantity;

[0021] According to all the selected feature quantities and the correlation coefficients, calculating the weighted average value about all the selected feature quantities, and taking the weighted average value as the electricity stealing analysis value of the user.

[0022] Further, the electricity stealing analysis model further comprises an electricity stealing threshold value;

[0023] The step of judging whether the electricity consumption of the user belongs to electricity stealing behavior according to the electricity stealing analysis value comprises:

[0024] determining whether the electricity stealing analysis value is greater than or equal to the electricity stealing threshold value, if yes, confirming that the electricity consumption behavior of the user belongs to electricity stealing, and generating an electricity stealing work order of the electricity consumption;

[0025] The electricity stealing work order is used to prompt an operation and maintenance personnel to perform anti-electricity stealing processing on the user.

[0026] Further, the electricity stealing analysis model further comprises an observation threshold value and a suspicion threshold value;

[0027] The step of determining whether the electricity consumption of the user belongs to electricity stealing behavior according to the electricity stealing analysis value further comprises:

[0028] If the electricity stealing analysis value is less than the electricity stealing threshold value, it is determined whether the electricity stealing analysis value is less than the suspicion threshold value, if no, it is confirmed that the electricity consumption behavior of the user belongs to suspicious electricity stealing, and an alarm notification is generated;

[0029] If the electricity stealing analysis value is less than the suspicion threshold value, it is determined whether the electricity stealing analysis value is less than the observation threshold value, if no, the identification of the user is set as observation electricity stealing.

[0030] Further, the method further comprises a step of obtaining the weighting factor, which comprises:

[0031] For each electricity stealing related behavior, a unique corresponding feature quantity is configured for the electricity stealing related behavior;

[0032] According to historical data about electricity stealing, a normal distribution curve is established;

[0033] For each of the feature quantities, according to the distribution data of the feature quantity in the normal distribution curve, a weighting factor of the feature quantity is configured.

[0034] Further, the calculation formula of the correlation coefficient comprises:

[0035] wherein, R jk represents the correlation coefficient of the jth selected feature quantity and the kth selected feature quantity, Cov(j, k) represents the covariance of the jth selected feature quantity and the kth selected feature quantity, Var[j] represents the variance of the jth selected feature quantity, and Var[k] represents the variance of the kth selected feature quantity.

[0036] In a second aspect, an embodiment of the present application provides an electricity stealing behavior identification device applied to an electronic device, wherein a relationship table is stored on the electronic device, and the relationship table records the corresponding relationship between a user identification, a user and a user scene; the electricity stealing behavior identification device comprises a preprocessing module, a calculation module and an identification module.

[0037] The preprocessing module is configured to, when determining that the currently consumed collection information in the message queue is abnormal, determine a user corresponding to the collection information and a power consumption scene to which the user belongs according to a user identifier of the collection information, determine all collection information of the user in a power stealing interval from the cache as a target information set, and determine a power stealing analysis model corresponding to the power consumption scene.

[0038] The collection information is power stealing related behavior information of a user collected by a monitoring device in real time.

[0039] The calculation module is configured to analyze power consumption of the user in the power stealing interval according to the power stealing analysis model and a feature quantity of each piece of collection information in the target information set, to obtain a power stealing analysis value of the user, wherein the feature quantity is used to represent a power stealing related behavior to which the collection information belongs.

[0040] The identification module is configured to determine whether the power consumption of the user is a power stealing behavior according to the power stealing analysis value.

[0041] In a third aspect, an electronic device is provided, including a processor and a memory, the memory storing a computer program capable of being executed by the processor, and the processor can execute the computer program to implement the power stealing behavior identification method according to the first aspect.

[0042] In a fourth aspect, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the power stealing behavior identification method according to the first aspect.

[0043] The power stealing behavior identification method, device, electronic device and computer readable storage medium provided by the embodiments of the present application can put the collection information of the power stealing related behavior collected in real time into a message queue, and when determining that the currently consumed collection information in the message queue is abnormal, determine the user corresponding to the collection information and the power consumption scene to which the user belongs, so as to analyze the power stealing analysis value of the user by using the power stealing analysis model corresponding to the power consumption scene and the feature quantity of each piece of collection information of the user in the power stealing interval, and then determine whether the power consumption of the user is a power stealing behavior according to the power stealing analysis value, so as to realize real-time monitoring and identification, segment the power consumption scene, analyze the power consumption by using the power stealing analysis model exclusive to the power consumption scene, realize accurate analysis, and improve the accuracy of power stealing identification and the identification effect.

[0044] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the following preferred embodiments are described in detail below, and the accompanying drawings are described as follows. BRIEF DESCRIPTION OF DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some of the embodiments of the present application, and therefore should not be regarded as a limitation to the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0046] Figure 1 The block schematic diagram of the electricity stealing behavior identification system provided by the embodiments of the present application is shown.

[0047] Figure 2 The flowchart of one of the electricity stealing behavior identification methods provided by the embodiments of the present application is shown.

[0048] Figure 3 The flowchart of the part of the sub-steps in step S15 is shown. Figure 2

[0049] The flowchart of the part of the sub-steps in step S152 is shown. Figure 4 Figure 3 The flowchart of the part of the sub-steps in step S153 is shown.

[0050] Figure 5 The flowchart of the other of the electricity stealing behavior identification methods provided by the embodiments of the present application is shown.

[0051] Figure 6 Figure 3 The flowchart of the part of the sub-steps in step S17 is shown.

[0052] Figure 7 The flowchart of the part of the sub-steps in step S17 is shown. Figure 2

[0053] The block schematic diagram of the electricity stealing behavior identification device provided by the embodiments of the present application is shown. Figure 8

[0054] The block schematic diagram of the electronic device provided by the embodiments of the present application is shown. Figure 9

[0055] Icon: 100-electricity stealing behavior identification system; 110-electronic device; 120-monitoring device; 130-electric meter; 140-concentrator; 150-electricity stealing behavior identification device; 160-preprocessing module; 170-computing module; 180-identification module. DETAILED DESCRIPTION

[0056] ​​​The technical solutions in the embodiments of the present application will be clearly and completely described with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations.

[0057] Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0058] It should be noted that the relational terms such as "first" and "second" and the like are used only to distinguish one entity or operation from another, and do not necessarily require or imply that these entities or operations exist in any such actual relationship or order. Moreover, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or apparatus that includes a list of elements not only includes those elements, but also includes other elements not explicitly listed, or other elements inherent in such a process, method, article or apparatus. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article or apparatus that includes the element.

[0059] Due to the influence of electricity stealing behavior on the economic benefits and investment returns of power enterprises, it is an urgent problem to identify electricity stealing behavior of users and implement anti-electricity stealing work.

[0060] Traditional electricity stealing behavior identification methods include manual regular inspection and system identification.

[0061] The manual regular inspection method is a process of collecting data, manually screening, checking working condition information and fraud events, and identifying electricity stealing behavior by business personnel observing power consumption data, working condition data, and system collected electricity meter fraud event data to analyze electricity stealing user suspects. Manual identification is highly subjective and has large data volume, so the error rate is high. At the same time, it is time-consuming, costly, and inefficient.

[0062] The system identification method is to analyze collected data, fraud events, and working condition information and other data within a period of time by the system on the basis of manual identification to identify electricity stealing behavior. This method has large data volume, consumes a large amount of computing resources, and still has the problem of poor identification effect.

[0063] Based on the above considerations, the embodiment of the present application provides a power stealing behavior identification scheme, which can improve the problems of large amount of calculation, high error rate and poor identification effect of the current power stealing behavior identification method. The scheme will be described in detail below.

[0064] The power stealing behavior identification method provided by the present application can be applied to a power stealing behavior identification system 100 as shown in the figure, which comprises an electronic device 110 and a plurality of monitoring devices 120. Figure 1 The electronic device 110 and the plurality of monitoring devices 120 are connected in communication by wired or wireless mode.

[0065] Each monitoring device 120 is used to perform at least one monitoring and collecting task to monitor the concentrator 140 or the electric meter 130 of a certain user.

[0066] It should be understood that the concentrator 140 is connected to a plurality of electric meters 130. The concentrator 140 is the central management device and control device of the remote centralized meter reading system, which is responsible for functions such as timing reading of terminal (i.e. electric meter 130) data, command transmission of the system, data communication, network management, event recording, horizontal transmission of data, etc.

[0067] Each monitoring and collecting task includes a target object, a sampling period, a number of the monitoring and collecting task, a monitoring event and data of the target collection.

[0068] The monitoring device 120 performs each monitoring and collecting task to monitor the target object and collects the target data every interval of the sampling period to obtain collection information. The number is used as a characteristic identifier of the collection information, and the collection information with the identifier is uploaded to the electronic device 110.

[0069] It should be understood that the collection information also contains the user identifier of the target object. The user can be any power consumption unit such as individual family user, organization or group, company, school, etc.

[0070] The electronic device 110 is used to receive the collection information of any monitoring device 120, and determine a characteristic quantity used to represent the power stealing related behavior to which the collection information belongs according to the characteristic identifier of the collection information, and cache the collection information to a message queue according to the receiving order.

[0071] The power stealing related behavior includes but is not limited to: consumption of 0, abnormal wiring, opening the cover of the electric meter 130, etc. The characteristic quantity is pre-prepared and is a kind of representation quantity used to represent the power stealing related behavior to which the collection information belongs.

[0072] The electronic device 110 is also used to consume the collection information in the message queue according to the arrangement order to determine whether the collection information is abnormal.

[0073] In detail, one determination condition can be set for each feature quantity, and the electronic device 110 determines the determination condition according to the feature identifier of the collection information, and determines whether the collection information is abnormal according to the determination condition. For example, if the collection information is the consumption power of the power meter 130, and the determination condition is that the consumption power is 0, when the consumption power in the collection information is 0, it is determined that the collection information is abnormal.

[0074] When the collection information itself is prompt information triggered after an abnormal event, it is determined that the collection information is abnormal when it is determined that the collection information corresponds to an electricity stealing related behavior abnormal event. For example, when it is determined that the collection information is prompt information triggered by abnormal wiring of the power meter 130, it is determined that the collection information is abnormal.

[0075] It should be understood that the collection information after consumption can be deleted from the message queue and stored in the cache, or the collection information after consumption can be added with a consumed identifier but not deleted.

[0076] The electronic device 110 also stores a relationship table, and the relationship table records the corresponding relationship between the user identifier, the user and the user scene, and the electronic device 110 is also used to implement the electricity stealing behavior identification method.

[0077] Further, the electronic device 110 also has a feature quantity library, and the feature quantity library records the corresponding relationship between the feature identifier and the feature quantity, i.e., the corresponding relationship between the number and the feature quantity.

[0078] The electronic device 110 can be a server, a server cluster or a terminal.

[0079] In order to introduce the electricity stealing phase identification scheme in more detail, in one embodiment, referring to Figure 2 , an electricity stealing behavior identification method is provided, including the following steps. In this embodiment, the electricity stealing behavior identification method is applied to the electronic device 110 as shown in Figure 1 .

[0080] S11, when it is determined that the collection information currently consumed in the message queue is abnormal, the user corresponding to the collection information and the power consumption scene to which the user belongs are determined according to the user identifier of the collection information.

[0081] When the electronic device 110 determines that the collection information currently consumed in the message queue is abnormal, the relationship table is queried to query the relationship between the user identifier, the user and the power consumption scene, so as to determine the user corresponding to the collection information and the power consumption scene to which the user belongs.

[0082] The collection information is the electricity stealing related behavior information of a certain user monitored and collected by the monitoring device 120 in real time. When the collection information is abnormal data or abnormal prompt information, etc., it can be determined that the collection information is abnormal.

[0083] In the present application, the user refers to the power consumption unit. If the power consumption unit is a company, the user is a company. If the power consumption unit is a household user, the user is a household user. Each power consumption unit corresponds to at least one electric meter 130, and thus the user based on the collection information obtained from the electric meter 130 is the power consumption unit.

[0084] S13, all collection information of the user in the electricity stealing interval is determined from the cache as a target information set, and an electricity stealing analysis model corresponding to the power consumption scenario is determined.

[0085] The electricity stealing interval is a time interval determined by the analysis duration. For example, if the analysis duration is one day and the collection time of the collection information is 14:00 on April 20, the electricity stealing interval is from 14:00 on April 19 to 15:00 on April 20. The analysis duration can be adjusted or modified according to the needs.

[0086] The electronic device 110 caches collection information at multiple time points, and thus after the user of the abnormal collection information is determined, all collection information of the user in the electricity stealing interval can be extracted from the cache of the electronic device 110.

[0087] Considering the differences in power consumption and power consumption standards, in order to improve the accuracy of electricity stealing behavior identification, multiple power consumption scenarios can be obtained by scene segmentation according to the differences in power consumption and the like. For example, industrial and commercial power consumption, heavy power consumption, household power consumption, high-voltage power consumption, low-voltage power consumption, and the like can be divided. Then, according to the differences in power consumption scenarios, exclusive electricity stealing analysis models are developed for each power consumption scenario. Different electricity stealing analysis models have different analysis focuses and analysis standards, that is, the electricity stealing analysis model and the power consumption scenario are one-to-one. The electronic device 110 is also configured with a model library for storing electricity stealing analysis models of each power consumption scenario.

[0088] It should be understood that there can be different scene division standards according to different actual applications, actual needs or actual purposes, etc.

[0089] The power consumption scenario to which each user belongs can be pre-set, and thus after the power consumption scenario to which the user of the abnormal collection information belongs is determined, the electricity stealing analysis model can be determined by querying the model library.

[0090] S15, according to the electricity stealing analysis model and the feature quantity of each collection information in the target information set, the power consumption of the user in the electricity stealing interval is analyzed to obtain the electricity stealing analysis value of the user.

[0091] The feature quantity is used to represent the electricity stealing related behavior to which the collection information belongs. Moreover, the feature quantity can be a numerical value.

[0092] For example, when the electricity stealing related behavior to which the collection information belongs is that the consumed electricity is 0, the characteristic quantity of the collection information can be 3, and at this time, the characteristic quantity 3 represents the behavior that the consumed electricity is 0. Or, when the collection information is the prompt information of the abnormal connection, the electricity stealing related behavior to which the collection information belongs is the abnormal connection, and at this time, the characteristic quantity of the collection information can be 2, and 2 represents the abnormal connection of the electric meter 130.

[0093] Each characteristic quantity is a value exclusive to the corresponding electricity stealing related behavior, and the specific numerical value of the characteristic quantity can be set according to historical experience or analysis of historical data.

[0094] The characteristic quantity library of the electronic device 110 stores the corresponding relationship between the characteristic identifier and the characteristic quantity, and the collection information has the characteristic identifier, so that the characteristic quantity of the collection information can be queried according to the characteristic identifier.

[0095] S17, according to the electricity stealing analysis value, judging whether the user's electricity use is a electricity stealing behavior.

[0096] In the above electricity stealing behavior identification method, the collection information of the electricity stealing related behavior monitored and collected in real time is put into the message queue, and when it is determined that the collection information currently consumed by the message queue is abnormal, the user corresponding to the collection information and the electricity use scene to which the user belongs are determined, so that the electricity stealing analysis model corresponding to the electricity use scene and the characteristic quantity of each collection information of the user in the electricity stealing interval are used to analyze the electricity stealing analysis value of the user, and then the electricity stealing analysis value is used to judge whether the user's electricity use is a electricity stealing behavior, realizing real-time monitoring and identification, and dividing the electricity use scene, using the electricity use scene exclusive electricity stealing analysis model for electricity analysis, realizing accurate analysis, so as to improve the accuracy of electricity stealing identification, and improve the identification effect.

[0097] In order to improve the accuracy of electricity stealing behavior identification, the electricity stealing analysis model can include a selected characteristic quantity combination, the calculation weight of each selected characteristic quantity, and the determination threshold. The selected characteristic quantity combination represents the electricity stealing related behavior that the electricity use scene corresponding to the electricity stealing analysis model pays attention to. It should be noted that the selected characteristic quantity, the calculation weight and the determination threshold of the electricity stealing analysis model of different electricity use scenes are different.

[0098] In order to reduce the calculation amount and the calculation complexity, and at the same time improve the accuracy of electricity stealing analysis, in one embodiment, with reference to Figure 3 The above step S15 can be realized by the following substeps.

[0099] S151, analyzing each collection information in the target information set to obtain the characteristic quantity corresponding to each collection information.

[0100] In detail, for each piece of collected information, the characteristic quantity of the piece of collected information is obtained by querying a characteristic table pre-stored on the electronic device 110 according to the characteristic identifier of the piece of collected information, where the characteristic table records the correspondence between characteristic identifiers and characteristic quantities.

[0101] S152, extracting the characteristic quantity that meets the selected characteristic quantity combination from all characteristic quantities as the selected characteristic quantity, and calculating the electricity stealing weight of each selected characteristic quantity.

[0102] The electronic device 110 extracts the characteristic quantity belonging to the selected characteristic quantity combination as the selected characteristic quantity based on the selected characteristic quantity combination in the electricity stealing analysis model, filters other characteristic quantities, and further calculates the electricity stealing weight of each selected characteristic quantity.

[0103] The size of the electricity stealing weight of the selected characteristic quantity can be understood as: the possible electricity stealing value of the electricity stealing related behavior corresponding to the selected characteristic quantity in the electricity stealing interval.

[0104] S153, based on all selected characteristic quantities and electricity stealing weights, analyzing the electricity consumption of the user in the electricity stealing interval to obtain the electricity stealing analysis value of the user.

[0105] Through the above steps S151-S153, the electricity stealing related behavior of the user is analyzed accurately to obtain a more accurate electricity stealing analysis value.

[0106] The calculation method of the electricity stealing weight of each selected characteristic quantity can be flexibly set, for example, it can be calculated according to a preset rule, or it can be obtained by using a machine algorithm.

[0107] In one possible implementation, referring to Figure 4 The calculation weight of each selected characteristic quantity can be calculated through the following sub-steps.

[0108] S1521, querying the weighting factor of each selected characteristic quantity from the preset weighting factor table.

[0109] The weighting factor of each characteristic quantity can be set according to historical experience, or it can be calculated by using a preset rule, and the embodiment is not limited specifically.

[0110] After the characteristic quantity is formulated, the corresponding weighting factor is formulated for each characteristic quantity, and the correspondence between the characteristic quantity and the weighting factor is recorded in the weighting factor table, and the weighting factor table is stored in the electronic device 110. Thus, the electronic device 110 can query the weighting factor of each selected characteristic quantity from the weighting factor table after the selected characteristic quantity is determined.

[0111] The weighting factor can be used to represent the probability that the electricity stealing related behavior corresponding to the characteristic quantity constitutes the electricity stealing behavior in historical data or experience.

[0112] S1522, for each selected feature quantity, multiplying the calculation weight of the selected feature quantity and the weighting factor of the selected feature quantity as the electricity stealing weight of the selected feature quantity.

[0113] The calculation weight is specified by the electricity stealing analysis model of the electricity consumption scene corresponding to the user, and the calculation weight of the selected feature quantity represents the probability of the electricity stealing related behavior corresponding to the selected feature quantity constituting the electricity stealing behavior in the electricity consumption scene.

[0114] W represents the calculation weight, and Wf represents the weighting factor. The electricity stealing weight P of the selected feature quantity is: P = W x Wf.

[0115] Through the above steps S1521-S1523, a more accurate electricity stealing weight closer to the actual situation can be obtained.

[0116] In order to make the weighting factor of each feature quantity as close as possible to the situation in the actual electricity consumption scene, therefore, in an embodiment, the electricity stealing behavior identification method provided by the application further comprises the step of obtaining the weighting factor. Referring to Figure 5 , the step can be realized by the following steps.

[0117] S21, for each electricity stealing related behavior, configuring a unique corresponding feature quantity for the electricity stealing related behavior.

[0118] The feature quantity in the feature quantity library of the electronic device 110 is also obtained in the manner of S21.

[0119] S22, according to the historical data about electricity stealing, a normal distribution curve is established.

[0120] The historical data is various data of the user's electric meter 130 collected after the user steals electricity, or the prompt information of the event. That is, the historical data is the collection information corresponding to each feature quantity collected after the user steals electricity.

[0121] S23, for each feature quantity, according to the distribution data of the feature quantity in the normal distribution curve, configuring the weighting factor of the feature quantity.

[0122] In detail, the proportion of the distribution of the feature quantity in the normal distribution curve can be taken as the weighting factor of the feature quantity.

[0123] In actual calculation, the normal distribution curve about electricity stealing can be established by region, and then the weighting factor matched with the region is determined in the manner of S23.

[0124] After obtaining the electricity stealing weight, the calculation manner of the electricity stealing analysis value can be flexibly set. For example, the sum of the weight characteristic values of the selected characteristic quantities can be taken as the electricity stealing analysis value, or machine learning can be used to fit the electricity stealing weight of each selected characteristic quantity to obtain the electricity stealing analysis value. In this embodiment, no specific limitation is made.

[0125] In a possible implementation, with reference to Figure 6 The step S153 can also be implemented by the following steps.

[0126] S1531, for each selected characteristic quantity, analyze and obtain the correlation coefficient between the selected characteristic quantity and each remaining selected characteristic quantity.

[0127] The calculation manner of the correlation coefficient can be flexibly set. For example, machine learning or neural network fitting can be used, or the correlation coefficient can be calculated according to a preset rule. In this embodiment, no unique limitation is made.

[0128] In a possible implementation, the correlation coefficient can be calculated by using a calculation formula. The calculation formula of the correlation coefficient can include:

[0129] wherein R jk represents the correlation coefficient between the jth selected characteristic quantity and the kth selected characteristic quantity, Cov(j, k) represents the covariance between the jth selected characteristic quantity and the kth selected characteristic quantity, Var[j] represents the variance of the jth selected characteristic quantity, and Var[k] represents the variance of the kth selected characteristic quantity.

[0130] The greater the correlation coefficient between two characteristic quantities, the greater the possibility that both constitute electricity stealing behavior.

[0131] S1532, according to all the selected characteristic quantities and the correlation coefficients, calculate the weighted average value of all the selected characteristic quantities, and take the weighted average value as the electricity stealing analysis value of the user.

[0132] The calculation formula of the weighted average value can include: wherein E represents the electricity stealing analysis value, P i represents the electricity stealing weight of the ith selected characteristic quantity, and n represents the number of selected characteristic quantities.

[0133] Through the above S1531-S1532, the correlation between the selected characteristic quantities is considered in the process of calculating the electricity stealing weight, so that the electricity stealing analysis value is more accurate.

[0134] The determination threshold in the electricity stealing analysis model can include an observation threshold, a suspicion threshold and an electricity stealing threshold. For different electricity consumption scenarios, the observation threshold, the suspicion threshold and the electricity stealing threshold in the corresponding electricity stealing analysis model are also different.

[0135] On this basis, referring to Figure 7 The step S17 can be implemented by the following sub-steps.

[0136] S171, determining whether the electricity stealing analysis value is greater than or equal to the electricity stealing threshold. If not, step S172 is executed, and if yes, S177 is executed.

[0137] S172, determining whether the electricity stealing analysis value is less than the suspicious threshold. If yes, step S173 is executed, and if not, step S176 is executed.

[0138] S173, determining whether the electricity stealing analysis value is less than the observation threshold. If not, step S174 is executed. If yes, step S175 is executed.

[0139] S174, setting the identity of the user as observation electricity stealing. The identity of the observation electricity stealing is used to prompt the operation and maintenance personnel to further observe the electricity consumption of the user. At this time, the user is an observation user.

[0140] S175, determining that the electricity consumption of the user does not belong to electricity stealing behavior.

[0141] S176, confirming that the electricity consumption behavior of the user belongs to suspicious electricity stealing, and generating an alarm notification. The alarm notification is used to warn the user and remind the operation and maintenance personnel. At this time, the user is a suspicious user.

[0142] S177, confirming that the electricity consumption behavior of the user belongs to electricity stealing, and generating an electricity stealing work order. The electricity stealing work order is used to prompt the operation and maintenance personnel to perform anti-electricity stealing processing on the user. At this time, the user is an electricity stealing user.

[0143] Through the steps S171 to S176, it can be determined whether the electricity consumption of the user belongs to electricity stealing behavior.

[0144] The electricity stealing behavior identification method provided by the application pre-formulates exclusive characteristic quantities for electricity stealing related behaviors, trains weighting factors for each characteristic quantity, classifies users, and performs scene segmentation. A plurality of electricity consumption scenes are formulated for each electricity stealing analysis model obtained by scene segmentation. When the current consumption collection information in the message queue is abnormal, electricity stealing analysis is triggered. According to the electricity stealing analysis model to which the user belongs and all collection information of the user in the electricity stealing interval, electricity stealing analysis is performed to obtain an electricity stealing analysis value. Then, the electricity stealing behavior is identified according to the electricity stealing analysis value.

[0145] Through the above steps S11-S17 and the related sub-steps, the electricity stealing weight and the weighted average value (i.e., the electricity stealing analysis value) of each user are calculated in real time once the abnormal collection information occurs. Moreover, the electricity stealing analysis value is calculated based on the electricity stealing analysis model corresponding to the electricity scene to which the user belongs according to the electricity scene to which the user belongs. Based on the electricity stealing analysis model, invalid data can be filtered, the calculation amount and complexity are reduced, the pressure distribution is calculated, the pressure peak is reduced, and the performance bottleneck is relieved. According to the electricity stealing analysis value, the electricity stealing work order of the electricity stealing behavior is output in a hierarchical manner (observation user, suspected user, and electricity stealing user). The electricity stealing user type and the suspicion level are divided, and finally the object of electricity stealing inspection is locked, which can effectively reduce the labor cost.

[0146] Based on the above concept of the electricity stealing behavior recognition method, referring to Figure 8 In one embodiment, an electricity stealing behavior recognition device 150 is provided, which can be applied to Figure 1 the electronic device 110 as shown. The electricity stealing behavior recognition device 150 includes a preprocessing module 160, a calculation module 170, and a recognition model.

[0147] The preprocessing module 160 is configured to, when it is determined that the currently consumed collection information in the message queue is abnormal, determine the user corresponding to the collection information and the electricity scene to which the user belongs according to the user identifier of the collection information, determine all collection information of the user in the electricity stealing interval from the cache as a target information set, and determine the electricity stealing analysis model corresponding to the electricity scene.

[0148] The collection information is the electricity stealing related behavior information of a certain user collected by the monitoring device 120 in real time.

[0149] The calculation module 170 is configured to analyze the electricity stealing of the user in the electricity stealing interval according to the electricity stealing analysis model and the feature quantity of each collection information in the target information set, and obtain the electricity stealing analysis value of the user.

[0150] The feature quantity is used to represent the electricity stealing related behavior to which the collection information belongs.

[0151] The recognition module 180 is configured to determine whether the electricity of the user belongs to the electricity stealing behavior according to the electricity stealing analysis value.

[0152] The working principle of the above electricity stealing behavior recognition device 150 is that when the preprocessing module 160 determines that the currently consumed collection information in the message queue is abnormal, the user corresponding to the collection information and the electricity scene to which the user belongs are determined, so that the calculation module 170 uses the electricity stealing analysis model corresponding to the electricity scene and the feature quantity of each collection information of the user in the electricity stealing interval to analyze the electricity stealing analysis value of the user, and then the recognition module 180 determines whether the electricity of the user is the electricity stealing behavior according to the electricity stealing analysis value.

[0153] The electricity stealing behavior recognition device 150 can realize real-time monitoring and recognition, and segment electricity consumption scenarios, use electricity consumption scenario-specific electricity stealing analysis models for electricity consumption analysis, realize accurate analysis, thereby improving the accuracy of electricity stealing recognition, and improving the recognition effect.

[0154] For specific limitations of the electricity stealing behavior recognition device 150, refer to the limitations of the electricity stealing behavior recognition method in the foregoing, which will not be repeated here. Each module in the above electricity stealing behavior recognition device 150 can be realized by software, hardware, and combinations thereof, in whole or in part. The above modules can be embedded in or independent of the processor in the electronic device 110 in hardware form, or can be stored in the memory in the electronic device 110 in software form, so as to be called and executed by the processor to perform the operations corresponding to each of the above modules.

[0155] In one embodiment, an electronic device 110, which can be a server, can have an internal structure diagram as shown in Figure 9 The electronic device 110 includes a processor, a memory, and a network interface connected by a system bus. The processor of the electronic device 110 is used to provide computing and control capabilities. The memory of the electronic device 110 includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The database of the computer device is used to store data generated in the execution process of the electricity stealing behavior recognition method. The network interface of the computer device is used to communicate with external terminals through network connection. The computer program is executed by the processor to implement an electricity stealing behavior recognition method.

[0156] Those skilled in the art can understand that Figure 9 the structure shown in the foregoing is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the foregoing, or combine certain components, or have a different arrangement of components. Figure 9

[0157] In one embodiment, the electricity stealing behavior recognition device 150 provided by the present application can be realized in the form of a computer program, which can run on an electronic device 110 as shown in Figure 9 The memory of the electronic device 110 can store various program modules constituting the electricity stealing behavior recognition device 150, such as Figure 8 ​The pre-processing module 160, the calculation module 170, and the identification module 180 are shown. The computer program composed of various program modules enables the processor to perform the steps in the electricity stealing behavior identification method described in the specification.

[0158] For example, Figure 9 The electronic device 110 can perform steps S11-S13 through the pre-processing module 160 in the electricity stealing behavior identification apparatus 150 as shown. Figure 8 The pre-processing module 160 in the electricity stealing behavior identification apparatus 150 performs steps S11-S13. The electronic device 110 can perform step S15 through the calculation module 170. The electronic device 110 can perform step S17 through the identification module 180.

[0159] In one embodiment, an electronic device 110 is provided, comprising a memory and a processor, the memory storing a computer program, and the processor implementing the following steps when executing the computer program: when it is determined that the currently consumed collection information in the message queue is abnormal, determining the user corresponding to the collection information and the power consumption scene to which the user belongs according to the user identifier of the collection information, wherein the collection information is the electricity stealing related behavior information of a certain user collected by the monitoring device 120 in real time; determining all collection information of the user in the electricity stealing interval from the cache as a target information set, and determining the electricity stealing analysis model corresponding to the power consumption scene; analyzing the power consumption of the user in the electricity stealing interval according to the electricity stealing analysis model and the feature quantity of each collection information in the target information set, to obtain the electricity stealing analysis value of the user, wherein the feature quantity is used to represent the electricity stealing related behavior to which the collection information belongs; and judging whether the power consumption of the user belongs to the electricity stealing behavior according to the electricity stealing analysis value.

[0160] In one embodiment, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the following steps: when it is determined that the currently consumed collection information in the message queue is abnormal, determining the user corresponding to the collection information and the power consumption scene to which the user belongs according to the user identifier of the collection information, wherein the collection information is the electricity stealing related behavior information of a certain user collected by the monitoring device 120 in real time; determining all collection information of the user in the electricity stealing interval from the cache as a target information set, and determining the electricity stealing analysis model corresponding to the power consumption scene; analyzing the power consumption of the user in the electricity stealing interval according to the electricity stealing analysis model and the feature quantity of each collection information in the target information set, to obtain the electricity stealing analysis value of the user, wherein the feature quantity is used to represent the electricity stealing related behavior to which the collection information belongs; and judging whether the power consumption of the user belongs to the electricity stealing behavior according to the electricity stealing analysis value.

[0161] In several embodiments provided in the present application, it should be understood that the disclosed apparatus and method can also be implemented by other manners. The apparatus embodiments described above are merely illustrative, for example, the flowcharts and block diagrams in the drawings show the possible implementation architecture, function and operation of the apparatus, method and computer program product according to the embodiments of the present application. In this regard, each block in the flowcharts or block diagrams can represent a module, a program segment or a part of code, which contains one or more executable instructions for implementing the specified logic function. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in different orders from those noted in the drawings. For example, two consecutive blocks can actually be executed substantially in parallel, and they can also be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and the combination of blocks in the block diagrams and / or flowcharts, can be implemented by a dedicated hardware-based system for executing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0162] In addition, each functional module in the embodiments of the present application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0163] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present application. The aforementioned storage medium includes various media that can store program codes, such as a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0164] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A power stealing behavior identification method, characterized in that, The method is applied to an electronic device, and a relationship table is stored on the electronic device, wherein a corresponding relationship among a user identifier, a user, and a user scenario is recorded on the relationship table. When it is determined that the currently consumed collection information in the message queue is abnormal, a user corresponding to the collection information and a power consumption scenario to which the user belongs are determined according to a user identifier of the collection information, wherein the collection information is information about power stealing related behaviors of a certain user, which is collected by a monitoring device in real time. All collection information of the user in a power stealing interval is determined from the cache as a target information set, and a power stealing analysis model corresponding to the power consumption scenario is determined. According to the power stealing analysis model and a feature quantity of each piece of collection information in the target information set, power consumption of the user in the power stealing interval is analyzed to obtain a power stealing analysis value of the user, wherein the feature quantity is used to represent power stealing related behaviors to which the collection information belongs. Whether the power consumption of the user belongs to power stealing is determined according to the power stealing analysis value. The power stealing analysis model includes a selected feature quantity combination, and the selected feature quantity combination represents power stealing related behaviors that the power consumption scenario corresponding to the power stealing analysis model focuses on. The step of obtaining the power stealing analysis value of the user includes: Each piece of collection information in the target information set is analyzed to obtain a feature quantity corresponding to each piece of collection information. Feature quantities that meet the selected feature quantity combination are extracted from all the feature quantities as selected feature quantities, and a power stealing weight of each selected feature quantity is calculated. Based on all the selected feature quantities and the power stealing weights, power consumption of the user in the power stealing interval is analyzed to obtain the power stealing analysis value of the user.

2. The electricity stealing behavior identification method of claim 1, wherein, The power stealing analysis model further includes a calculation weight of each selected feature quantity, and the step of calculating the power stealing weight of each selected feature quantity includes: A weighting factor of each selected feature quantity is queried from a preset weighting factor table. For each selected feature quantity, a product of the calculation weight of the selected feature quantity and the weighting factor is taken as the power stealing weight of the selected feature quantity.

3. The electricity stealing behavior identification method of claim 1, wherein, The step of analyzing power consumption of the user in the power stealing interval based on all the selected feature quantities and the power stealing weights to obtain the power stealing analysis value of the user includes: For each selected feature quantity, a correlation coefficient between the selected feature quantity and each remaining selected feature quantity is analyzed and obtained. According to all the selected feature quantities and the correlation coefficients, a weighted average value about all the selected feature quantities is calculated, and the weighted average value is taken as the power stealing analysis value of the user.

4. The electricity stealing behavior identification method of claim 1, wherein, The power stealing analysis model further includes a power stealing threshold value. The step of determining whether the power consumption of the user belongs to power stealing according to the power stealing analysis value includes: It is determined whether the power stealing analysis value is greater than or equal to the power stealing threshold value, and if yes, it is confirmed that the power consumption of the user belongs to power stealing, and a power stealing work order of the power consumption is generated. The power stealing work order is used to prompt an operation and maintenance personnel to perform anti-power stealing processing on the user.

5. The electricity stealing behavior identification method of claim 4, wherein, The power stealing analysis model further includes an observation threshold value and a suspicion threshold value. The step of judging whether the electricity use of the user belongs to electricity larceny behavior according to the electricity larceny analysis value further comprises: If the electricity larceny analysis value is less than the electricity larceny threshold value, whether the electricity larceny analysis value is less than the suspicion threshold value is judged, if not, it is confirmed that the electricity use of the user belongs to suspicious electricity larceny, and an alarm notification is generated; If the electricity larceny analysis value is less than the suspicion threshold value, whether the electricity larceny analysis value is less than the observation threshold value is judged, if not, the identification of the user is set as observation electricity larceny.

6. The electricity stealing behavior identification method of claim 2, wherein, The method further comprises a step of obtaining the weighting factor, which comprises: For each electricity larceny related behavior, a unique corresponding feature quantity is configured for the electricity larceny related behavior; According to historical data about electricity larceny, a normal distribution curve is established; For each of the feature quantities, a weighting factor of the feature quantity is configured according to the distribution data of the feature quantity in the normal distribution curve.

7. An electricity theft behavior identification apparatus, characterized by, The electricity larceny behavior identification device is applied to an electronic device, and a relationship table is stored on the electronic device, wherein the relationship table records the corresponding relationship among user identification, users and user scenarios; the electricity larceny behavior identification device comprises a preprocessing module, a calculation module and an identification module; The preprocessing module is configured to, when it is determined that the currently consumed collection information in the message queue is abnormal, determine the user corresponding to the collection information and the power use scenario to which the user belongs according to the user identification of the collection information, determine all collection information of the user in the electricity larceny interval from the cache as a target information set, and determine the electricity larceny analysis model corresponding to the power use scenario; The collection information is the electricity larceny related behavior information of a certain user collected by a monitoring device in real time; The calculation module is configured to analyze the electricity use of the user in the electricity larceny interval according to the electricity larceny analysis model and the feature quantity of each piece of collection information in the target information set, and obtain the electricity larceny analysis value of the user, wherein the feature quantity is used to represent the electricity larceny related behavior to which the collection information belongs; When the electricity larceny analysis model comprises a selected feature quantity combination, the selected feature quantity combination represents the electricity larceny related behavior focused on by the power use scenario corresponding to the electricity larceny analysis model, the calculation module is further configured to analyze each piece of collection information in the target information set to obtain the feature quantity corresponding to each piece of collection information, extract the feature quantity conforming to the selected feature quantity combination from all the feature quantities as a selected feature quantity, and calculate the electricity larceny weight of each selected feature quantity; based on all the selected feature quantities and the electricity larceny weight, the electricity use of the user in the electricity larceny interval is analyzed to obtain the electricity larceny analysis value of the user; The identification module is configured to judge whether the electricity use of the user belongs to electricity larceny behavior according to the electricity larceny analysis value.

8. An electronic device, comprising: The computer program is executed by the processor to implement the electricity larceny behavior identification method according to any one of claims 1 to 6.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the electricity larceny behavior identification method according to any one of claims 1 to 6.

Images