A method, device, equipment and storage medium for detecting abnormal power consumption
By obtaining data from the power consumption terminal, using the grouping model and abnormal analysis model generation process, combined with user priority, the problem of inaccurate diagnosis of power consumption abnormalities in the existing technology is solved, and accurate and efficient diagnosis of power consumption abnormalities is achieved. , adapt to the new situation in the power industry.
Patent Information
- Application Number
- CN202011465945.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-12-14
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2040-12-14
AI Technical Summary
The existing electricity consumption abnormality analysis model cannot accurately diagnose electricity consumption abnormalities and cannot adapt to the new situation of market opening and diversification of electricity consumption business in the power industry.
By obtaining electricity consumption behavior data from multiple power consumption terminals, using preset clustering models for clustering, generating exception levels, and determining the power consumption terminal to be checked based on user priorities, using exception filtering rules and abnormal analysis model generation process to improve Diagnostic accuracy.
It has achieved more accurate and efficient diagnosis of abnormal electricity consumption of power terminals, adapted to the new situation and demand of the power industry, and improved the innovation capabilities of power grid management.
Smart Images

Figure CN112560940B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power consumption anomaly detection, and in particular to a power consumption anomaly detection method, device, equipment and storage medium. Background Art
[0002] As the reform of the power system continues to advance, the power industry faces new situations such as market openness and diversified entities. Grid companies must actively carry out management model innovation and further "improve quality, increase efficiency, and reduce costs" to meet the changing market needs and better serve various market entities and customers. The successful experience of big data analysis in promoting management model innovation and new technology application is worth learning from.
[0003] Therefore, the development and promotion of new power grid technologies can support the construction of smart grids and enhance management innovation capabilities by leveraging the innovation-driven role of new technologies such as big data and mobile Internet.
[0004] However, with the diversified development of electricity consumption business, the methods of electricity theft have become more and more complicated. The original abnormal electricity consumption analysis model cannot adapt to the needs of the new situation and cannot accurately diagnose abnormal electricity consumption. Summary of the invention
[0005] The present invention provides a method, device, equipment and storage medium for detecting abnormal power consumption, which solves the technical problem that the abnormal power consumption analysis model in the prior art cannot accurately diagnose abnormal power consumption.
[0006] The present invention provides a method for detecting abnormal power consumption, involving multiple power consumption terminals, and the method comprises:
[0007] Acquire electricity consumption behavior data within a predetermined period from the plurality of electricity consumption terminals respectively;
[0008] Using a preset clustering model to classify the electricity consumption behavior data to obtain a variety of clustering data;
[0009] Input each type of clustering data into a corresponding power consumption anomaly analysis model, and output an anomaly level corresponding to the power consumption terminal; the power consumption anomaly analysis model is obtained through a preset anomaly analysis model generation process;
[0010] Based on the abnormality level and the preset user priority, determining the power-consuming terminal to be checked from the plurality of power-consuming terminals;
[0011] According to the result of checking the power-consuming terminal to be checked, it is determined whether the power-consuming terminal to be checked is in an abnormal power consumption state.
[0012] Optionally, the power consumption anomaly analysis model includes an anomaly filtering rule, the anomaly filtering rule is provided with an anomaly rule score, and the step of inputting each type of the clustering data into the corresponding power consumption anomaly analysis model and outputting the anomaly level corresponding to the power consumption terminal includes:
[0013] Screening each type of clustering data respectively according to the abnormal filtering rules;
[0014] Determine the data satisfying the abnormal filtering rule in each type of the clustered data as abnormal data;
[0015] Counting the number of occurrences of each type of abnormal data within the preset period;
[0016] The total abnormal score of the power consumption terminal corresponding to each type of abnormal data is calculated by using the number of occurrences of each type of abnormal data, the abnormal rule score and the preset risk coefficient corresponding to the power consumption terminal corresponding to each type of abnormal data;
[0017] Using the total score of each type of abnormality and the number of occurrences of each type of abnormal data, determine the average abnormality score corresponding to the power consumption terminal corresponding to each type of abnormal data;
[0018] Based on the proportion of the average score of each type of abnormality to the total score of abnormality, the abnormality level corresponding to the power terminal is output.
[0019] Optionally, the abnormal analysis model generation process includes:
[0020] Acquire historical electricity consumption behavior data within a predetermined period from the plurality of electricity consumption terminals respectively;
[0021] Using the preset grouping model to classify the historical electricity consumption behavior data to generate a variety of historical classification data;
[0022] A plurality of the historical classification data are used to respectively train a plurality of preset initial power consumption anomaly analysis models to obtain the power consumption anomaly analysis model.
[0023] Optionally, the step of determining whether the power terminal to be checked is in an abnormal power consumption state according to a result of checking the power terminal to be checked includes:
[0024] If the checking result is abnormal, determining that the power consumption terminal to be checked is in an abnormal power consumption state;
[0025] If the checking result is normal, it is determined that the power-consuming terminal to be checked is not in an abnormal power consumption state.
[0026] The present invention also provides a power consumption anomaly detection device, involving multiple power consumption terminals, the device comprising:
[0027] An electricity consumption behavior data acquisition module, used to respectively acquire electricity consumption behavior data within a predetermined period from the plurality of electricity consumption terminals;
[0028] A clustering module, used to classify the electricity consumption behavior data using a preset clustering model to obtain a variety of clustering data;
[0029] An abnormality level output module, used to input each type of clustering data into a corresponding power consumption abnormality analysis model, and output an abnormality level corresponding to the power consumption terminal; the power consumption abnormality analysis model is obtained by calling a preset abnormality analysis model generation module;
[0030] A module for determining a power terminal to be checked, configured to determine a power terminal to be checked from a plurality of power terminals based on the abnormality level and a preset user priority;
[0031] The power consumption abnormality state judgment module is used to determine whether the power consumption terminal to be checked is in the power consumption abnormality state according to the checking result of the power consumption terminal to be checked.
[0032] Optionally, the power consumption anomaly analysis model includes an anomaly filtering rule, and the anomaly level output module includes:
[0033] A screening submodule, used for screening each type of cluster data according to the abnormal filtering rules;
[0034] An abnormal data determination submodule, used for determining data satisfying the abnormal filtering rule in each type of clustered data as abnormal data;
[0035] An occurrence count submodule, used to count the occurrence times of each type of abnormal data within the preset period;
[0036] The abnormal total score calculation submodule is used to calculate the abnormal total score of the power terminal corresponding to each type of abnormal data by using the number of occurrences of each type of abnormal data, the abnormal rule score and the preset risk coefficient corresponding to the power terminal corresponding to each type of abnormal data;
[0037] The abnormal average score calculation submodule is used to determine the abnormal average score corresponding to the power consumption terminal corresponding to each type of abnormal data by using the total score of each type of abnormal data and the number of occurrences of each type of abnormal data;
[0038] The abnormality level determination submodule is used to output the abnormality level corresponding to the power terminal based on the proportion of the average score of each type of abnormality to the total abnormality score.
[0039] Optionally, the abnormal analysis model generation module includes:
[0040] A historical electricity consumption behavior data acquisition submodule, used to respectively acquire historical electricity consumption behavior data within a predetermined period from the plurality of electricity consumption terminals;
[0041] A historical classification data generation submodule, used to classify the historical electricity consumption behavior data using the preset grouping model to generate a variety of historical classification data;
[0042] The training submodule is used to respectively train a plurality of preset initial power consumption abnormality analysis models using a plurality of the historical classification data to obtain the power consumption abnormality analysis model.
[0043] Optionally, the abnormal power consumption state judgment module includes:
[0044] The abnormal state determination submodule is used to determine that the power consumption terminal to be checked is in an abnormal power consumption state if the checking result is abnormal;
[0045] The normal state determination submodule is used to determine that the power-consuming terminal to be checked is not in an abnormal power consumption state if the checking result is normal.
[0046] The present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of any of the above-mentioned power consumption anomaly detection methods.
[0047] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by the processor, implements any of the above-mentioned methods for detecting abnormal power consumption.
[0048] It can be seen from the above technical solutions that the present invention has the following advantages:
[0049] The present invention obtains power consumption behavior data within a predetermined period from multiple power terminals, clusters the power consumption behavior data using a clustering model to obtain a variety of clustering data; each type of clustering data is then input into a corresponding type of power consumption anomaly analysis model to obtain the abnormal level of each power terminal, and determines the power terminal to be checked in combination with the user priority; after receiving the inspection results of the power terminal to be checked, it is determined whether the power terminal to be checked is in an abnormal power consumption state. This method solves the technical problem that the power consumption anomaly analysis model in the prior art cannot accurately diagnose the power consumption anomaly, and diagnoses the power consumption anomaly of the power terminal more accurately and efficiently. BRIEF DESCRIPTION OF THE DRAWINGS
[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0051] Figure 1 A flowchart of a method for detecting abnormal power consumption provided by an embodiment of the present invention;
[0052] Figure 2 A flowchart of a method for detecting abnormal power consumption provided by an optional embodiment of the present invention;
[0053] Figure 3 A structural block diagram of a power consumption anomaly detection device provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0054] The existing power consumption anomaly analysis model usually only detects the characteristics of multiple users' power terminals based on a unified standard to determine the power consumption anomaly. However, due to the differences in power consumption habits in different regions, different industries or different users, if a unified standard is used to measure power consumption anomalies, the accuracy of determining power consumption anomalies may be reduced. To this end, the embodiments of the present invention provide a power consumption anomaly detection method, device, equipment and storage medium to solve the technical problem that the power consumption anomaly analysis model in the prior art cannot accurately diagnose power consumption anomalies.
[0055] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0056] See also Figure 1 , Figure 1 A flowchart of a method for detecting abnormal power consumption provided by an embodiment of the present invention.
[0057] The present invention provides a method for detecting abnormal power consumption, involving multiple power consumption terminals, and the method comprises:
[0058] Step 101, acquiring electricity consumption behavior data within a predetermined period from a plurality of electricity consumption terminals respectively;
[0059] In an embodiment of the present invention, electricity consumption behavior data within a predetermined period is obtained from multiple electricity consumption terminals respectively, where the predetermined period can be a time length in units of minutes, hours, days, months or years, and the electricity consumption behavior data includes current, voltage, power, phase angle data and other data.
[0060] Step 102, using a preset clustering model to classify the electricity usage behavior data to obtain a variety of clustering data;
[0061] After obtaining the electricity consumption behavior data, a preset grouping model is used to classify the electricity consumption behavior data of each electricity consumption terminal to facilitate independent processing of the subsequent electricity consumption anomaly analysis model.
[0062] Specifically, the electricity consumption behavior data can be divided into multiple clustering data based on current, voltage, power and phase angle.
[0063] Step 103, inputting each type of clustering data into a corresponding power consumption anomaly analysis model, and outputting anomaly levels corresponding to the power consumption terminals, respectively; the power consumption anomaly analysis model is obtained through a preset anomaly analysis model generation process;
[0064] In a specific implementation, when the method is implemented for the first time, an abnormal analysis model generation process can be used to generate multiple types of power consumption abnormal analysis models, and then each type of clustering data can be input into the corresponding type of power consumption abnormal analysis model to obtain the abnormal level corresponding to the power consumption terminal.
[0065] Step 104, determining a power-consuming terminal to be checked from the plurality of power-consuming terminals based on the abnormality level and the preset user priority;
[0066] After knowing the abnormality level of each power-consuming terminal, since users may have different priorities, for example, different power-consuming terminals have different power-consuming urgency, in order to efficiently verify whether the power-consuming terminal really has power-consuming abnormalities, the power-consuming terminals to be checked can be determined from multiple power-consuming terminals based on the abnormality level and the preset user priority.
[0067] Step 105: Determine whether the power-consuming terminal to be checked is in an abnormal power consumption state according to the result of checking the power-consuming terminal to be checked.
[0068] In a specific implementation, after determining the power-consuming terminal to be checked, the power-consuming terminal to be checked can be checked to obtain a checking result to verify whether the power-consuming terminal to be checked has abnormal power consumption, and finally, based on the checking result, it is determined whether the power-consuming terminal to be checked is in an abnormal power consumption state.
[0069] In an embodiment of the present invention, by acquiring power consumption behavior data within a predetermined period from multiple power terminals, clustering the power consumption behavior data using a clustering model to obtain a variety of clustering data; each type of clustering data is then input into a corresponding type of power consumption anomaly analysis model to obtain the abnormal level of each power terminal, and combined with the user priority, the power terminal to be checked is determined; after receiving the inspection results of the power terminal to be checked, it is determined whether the power terminal to be checked is in an abnormal power consumption state. This method solves the technical problem that the power consumption anomaly analysis model in the prior art cannot accurately diagnose the power consumption anomaly, and diagnoses the power consumption anomaly of the power terminal more accurately and efficiently.
[0070] See also Figure 2 , Figure 2 A flowchart of a method for detecting abnormal power consumption provided by an embodiment of the present invention.
[0071] The present invention provides a method for detecting abnormal power consumption, involving multiple power consumption terminals, and the method comprises:
[0072] Step 201, acquiring electricity consumption behavior data within a predetermined period from a plurality of electricity consumption terminals respectively;
[0073] Step 202, using a preset clustering model to classify the electricity usage behavior data to obtain a variety of clustering data;
[0074] In the embodiment of the present invention, the specific implementation process of steps 201-202 is similar to the above steps 101-102, which will not be repeated here.
[0075] In the embodiment of the present invention, by adopting big data and mobile Internet technology, guided by user demand perception, using power consumption abnormality data assets as energy, and driven by power consumption abnormality identification and cause labeling analysis, the seamless connection between the source system, the big backend, and the small frontend is promoted around the discovery and correction of power consumption abnormalities, so as to realize the transformation of power grid companies to an active and innovative modern power consumption management model.
[0076] Step 203, inputting each type of clustering data into a corresponding power consumption anomaly analysis model, and outputting an anomaly level corresponding to the power consumption terminal;
[0077] In a specific implementation, the power consumption anomaly analysis model includes an anomaly filtering rule, and the anomaly filtering rule is provided with an anomaly rule score. Step 203 may include the following sub-steps:
[0078] Screening each type of clustering data respectively according to the abnormal filtering rules;
[0079] Determine the data satisfying the abnormal filtering rule in each type of the clustered data as abnormal data;
[0080] Counting the number of occurrences of each type of abnormal data within the preset period;
[0081] The total abnormal score of the power consumption terminal corresponding to each type of abnormal data is calculated by using the number of occurrences of each type of abnormal data, the abnormal rule score and the preset risk coefficient corresponding to the power consumption terminal corresponding to each type of abnormal data;
[0082] Using the total score of each type of abnormality and the number of occurrences of each type of abnormal data, determine the average abnormality score corresponding to the power consumption terminal corresponding to each type of abnormal data;
[0083] Based on the proportion of the average score of each type of abnormality to the total score of abnormality, the abnormality level corresponding to the power terminal is output.
[0084] In an embodiment of the present invention, each type of clustering data is screened separately according to the abnormal filtering rules, for example, whether the current exceeds the current abnormal range, whether the voltage exceeds the voltage abnormal range, whether the power exceeds the power abnormal range, whether the phase angle exceeds the phase angle abnormal range, etc. If any parameter exceeds the corresponding abnormal range, the data is determined to be abnormal data. Then count the number of occurrences of each type of abnormal data within the preset period. Since the losses corresponding to each type of power abnormality may be different, a corresponding abnormal rule score can be configured for each type of power abnormality; then set the risk coefficient according to the scenario and industry in which the power terminal is located, and finally use the number of occurrences of each type of abnormal data, the abnormal rule score and the preset risk coefficient corresponding to each power terminal corresponding to the abnormal data to calculate the total abnormal score of the power terminal corresponding to each type of abnormal data. Specifically, it can be calculated by the following formula:
[0085] Total anomaly score = ∑(abnormal rule score * number of occurrences) * risk factor
[0086] The time range for summing is a preset period.
[0087] In order to reflect the frequency of abnormal power consumption in power consumption terminals, the quotient of the total score of each type of abnormality and the number of occurrences of the abnormal data can be used as the average score of the abnormality. Finally, based on the proportion of the average score of each type of abnormality to the total score of the abnormality, the abnormal situation of each type of abnormal data is determined, and then combined with the abnormal situation of all power consumption terminals, the abnormal level corresponding to the power consumption terminal is determined.
[0088] By analyzing the users with abnormal electricity consumption in the past, we can find that the data patterns presented by users of different types and in different regions for the same type of abnormality are completely different. For example, shunting is a common abnormal electricity consumption behavior. According to the statistics of the current imbalance rate in the three billing periods of peak, valley and flat, the study found that the current imbalance rate of individual users during peak hours is greater than 90%, and one phase current is greater than 0.3. When the three-phase current is short-circuited, the three-phase current is very small, and the imbalance rate of the three periods of peak, valley and flat is greater than 60%. When one phase is shunted, the current of the other phase is 3 times that of the phase current, and the imbalance rate of the three periods of peak, valley and flat is greater than 60%. When two phases are shunted, the current of the unshunted phase is higher, which can reach 5A, while the shunted phase is less than 0.1A. At this time, the current imbalance rate of the peak, valley and flat is also greater than 90%. In addition, in addition to the analysis of the meter data, if the load control has voltage and current data, it is still necessary to combine the data analysis of the load control. The user's installation and electricity usage time is also an analysis factor. For some users, a certain phase of current may be basically unused from the beginning of electricity installation. In addition to analyzing abnormal electricity usage according to the usual rules, the use of self-data comparison method is also an important analysis method to study and analyze the trajectory of changes in electricity usage behavior and find out the time point when electricity usage behavior changes suddenly.
[0089] Optionally, it is also possible to collect electricity consumption behavior data of multiple preset periods of an electricity terminal, classify the electricity consumption behavior data using a preset clustering model to obtain a variety of clustering data; and determine whether the clustering data is abnormal through the multiple clustering data and the electricity consumption anomaly analysis model.
[0090] In the embodiment of the present invention, the power consumption anomaly analysis model is obtained through a preset anomaly analysis model generation process;
[0091] Furthermore, the abnormal analysis model generation process includes the following sub-steps:
[0092] Acquire historical electricity consumption behavior data within a predetermined period from the plurality of electricity consumption terminals respectively;
[0093] Using the preset grouping model to classify the historical electricity consumption behavior data to generate a variety of historical classification data;
[0094] A plurality of the historical classification data are used to respectively train a plurality of preset initial power consumption anomaly analysis models to obtain the power consumption anomaly analysis model.
[0095] In an embodiment of the present invention, since the initial power consumption abnormality analysis model does not have multiple abnormal ranges corresponding to different power consumption terminals, a larger initial abnormal range corresponding to a variety of data can be set, and each initial abnormal range can be further adjusted through a variety of historical classification data to obtain a power consumption abnormality analysis model.
[0096] Step 204, determining a power-consuming terminal to be checked from the plurality of power-consuming terminals based on the abnormality level and the preset user priority;
[0097] In the embodiment of the present invention, the specific implementation process of step 204 is similar to the above step 104 and will not be repeated here.
[0098] Step 205: if the checking result is abnormal, determining that the power-consuming terminal to be checked is in an abnormal power consumption state;
[0099] Step 206: If the checking result is normal, determine that the power-consuming terminal to be checked is not in an abnormal power consumption state.
[0100] In the specific implementation, taking the current anomaly investigation as an example, the investigation process can be verified on site, using a current clamp to measure whether the current incoming line of the junction box is consistent with the current value displayed by the meter. For users with large electricity consumption, the meter should be calibrated on site regularly or irregularly, and the various functions of the meter should be checked to see if they are normal. If any of them is abnormal, it is determined that the power terminal to be investigated is in an abnormal power consumption state; if the results are all normal, it is determined that the power terminal to be investigated is not in an abnormal power consumption state.
[0101] In an embodiment of the present invention, by acquiring power consumption behavior data within a predetermined period from multiple power terminals, clustering the power consumption behavior data using a clustering model to obtain a variety of clustering data; each type of clustering data is then input into a corresponding type of power consumption anomaly analysis model to obtain the abnormal level of each power terminal, and combined with the user priority, the power terminal to be checked is determined; after receiving the inspection results of the power terminal to be checked, it is determined whether the power terminal to be checked is in an abnormal power consumption state. This method solves the technical problem that the power consumption anomaly analysis model in the prior art cannot accurately diagnose the power consumption anomaly, and diagnoses the power consumption anomaly of the power terminal more accurately and efficiently.
[0102] See also Figure 3 , Figure 3 A structural block diagram of a power consumption anomaly detection device provided in an embodiment of the present invention.
[0103] An embodiment of the present invention provides a power consumption anomaly detection device, which involves multiple power consumption terminals, and the device includes:
[0104] The power consumption behavior data acquisition module 301 is used to respectively acquire power consumption behavior data within a predetermined period from the plurality of power consumption terminals;
[0105] A clustering module 302 is used to classify the electricity usage behavior data using a preset clustering model to obtain a variety of clustering data;
[0106] The abnormality level output module 303 is used to input each type of the cluster data into the corresponding power consumption abnormality analysis model, and output the abnormality level corresponding to the power consumption terminal; the power consumption abnormality analysis model is obtained by calling a preset abnormality analysis model generation module;
[0107] A to-be-checked power-consuming terminal determining module 304 is used to determine a to-be-checked power-consuming terminal from the plurality of power-consuming terminals based on the abnormality level and a preset user priority;
[0108] The power consumption abnormality state judgment module 305 is used to determine whether the power consumption terminal to be checked is in the power consumption abnormality state according to the result of checking the power consumption terminal to be checked.
[0109] Optionally, the power consumption anomaly analysis model includes an anomaly filtering rule, and the anomaly level output module 303 includes:
[0110] A screening submodule, used for screening each type of cluster data according to the abnormal filtering rules;
[0111] An abnormal data determination submodule, used for determining data satisfying the abnormal filtering rule in each type of clustered data as abnormal data;
[0112] An occurrence count submodule, used to count the occurrence times of each type of abnormal data within the preset period;
[0113] The abnormal total score calculation submodule is used to calculate the abnormal total score of the power terminal corresponding to each type of abnormal data by using the number of occurrences of each type of abnormal data, the abnormal rule score and the preset risk coefficient corresponding to the power terminal corresponding to each type of abnormal data;
[0114] The abnormal average score calculation submodule is used to determine the abnormal average score corresponding to the power consumption terminal corresponding to each type of abnormal data by using the total score of each type of abnormal data and the number of occurrences of each type of abnormal data;
[0115] The abnormality level determination submodule is used to output the abnormality level corresponding to the power terminal based on the proportion of the average score of each type of abnormality to the total abnormality score.
[0116] Optionally, the abnormal analysis model generation module includes:
[0117] A historical electricity consumption behavior data acquisition submodule, used to respectively acquire historical electricity consumption behavior data within a predetermined period from the plurality of electricity consumption terminals;
[0118] A historical classification data generation submodule, used to classify the historical electricity consumption behavior data using the preset grouping model to generate a variety of historical classification data;
[0119] The training submodule is used to respectively train a plurality of preset initial power consumption abnormality analysis models using a plurality of the historical classification data to obtain the power consumption abnormality analysis model.
[0120] Optionally, the abnormal power consumption state judgment module 305 includes:
[0121] The abnormal state determination submodule is used to determine that the power consumption terminal to be checked is in an abnormal power consumption state if the checking result is abnormal;
[0122] The normal state determination submodule is used to determine that the power-consuming terminal to be checked is not in an abnormal power consumption state if the checking result is normal.
[0123] The present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of any of the above-mentioned power consumption anomaly detection methods.
[0124] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by the processor, implements any of the above-mentioned methods for detecting abnormal power consumption.
[0125] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.
[0126] In the several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0127] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0128] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0129] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.
[0130] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for detecting abnormal power consumption, characterized in that: Involving multiple power terminals, the method includes: Acquire electricity consumption behavior data within a predetermined period from the plurality of electricity consumption terminals respectively; Using a preset clustering model to classify the electricity consumption behavior data to obtain a variety of clustering data; Input each type of clustering data into a corresponding power consumption anomaly analysis model, and output an anomaly level corresponding to the power consumption terminal; the power consumption anomaly analysis model is obtained through a preset anomaly analysis model generation process; Based on the abnormality level and the preset user priority, determining the power-consuming terminal to be checked from the plurality of power-consuming terminals; Determining whether the power consumption terminal to be checked is in an abnormal power consumption state according to the result of checking the power consumption terminal to be checked; The power consumption anomaly analysis model includes an anomaly filtering rule, and the anomaly filtering rule is provided with an anomaly rule score. The step of inputting each type of the cluster data into the corresponding power consumption anomaly analysis model and outputting the anomaly level corresponding to the power consumption terminal includes: Screening each type of clustering data respectively according to the abnormal filtering rules; Determine the data satisfying the abnormal filtering rule in each type of the clustered data as abnormal data; Counting the number of occurrences of each type of abnormal data within the predetermined period; The number of occurrences of each type of abnormal data, the abnormal rule score and the preset risk coefficient corresponding to each power terminal corresponding to the abnormal data are used to calculate the total abnormal score of the power terminal corresponding to each type of abnormal data; the total abnormal score = ∑ (abnormal rule score * number of occurrences) * risk coefficient; wherein the time range of the summation is the preset period; Using the total score of each type of abnormality and the number of occurrences of each type of abnormal data, determine the average abnormality score corresponding to the power consumption terminal corresponding to each type of abnormal data; Based on the proportion of the average score of each type of abnormality to the total score of abnormality, the abnormality level corresponding to the power terminal is output.
2. The method according to claim 1, characterized in that The abnormal analysis model generation process includes: Acquire historical electricity consumption behavior data within a predetermined period from the plurality of electricity consumption terminals respectively; Using the preset grouping model to classify the historical electricity consumption behavior data to generate a variety of historical classification data; A plurality of the historical classification data are used to respectively train a plurality of preset initial power consumption anomaly analysis models to obtain the power consumption anomaly analysis model.
3. The method according to claim 1, characterized in that The step of determining whether the power terminal to be checked is in an abnormal power consumption state according to the result of checking the power terminal to be checked comprises: If the checking result is abnormal, determining that the power consumption terminal to be checked is in an abnormal power consumption state; If the checking result is normal, it is determined that the power-consuming terminal to be checked is not in an abnormal power consumption state.
4. A power consumption anomaly detection device, characterized in that: Involving multiple electricity terminals, the device includes: An electricity consumption behavior data acquisition module, used to respectively acquire electricity consumption behavior data within a predetermined period from the plurality of electricity consumption terminals; A clustering module, used to classify the electricity consumption behavior data using a preset clustering model to obtain a variety of clustering data; An abnormality level output module, used to input each type of clustering data into a corresponding power consumption abnormality analysis model, and output an abnormality level corresponding to the power consumption terminal; the power consumption abnormality analysis model is obtained by calling a preset abnormality analysis model generation module; A module for determining a power terminal to be checked, configured to determine a power terminal to be checked from a plurality of power terminals based on the abnormality level and a preset user priority; An abnormal power consumption state judgment module is used to determine whether the power consumption terminal to be checked is in an abnormal power consumption state according to the result of checking the power consumption terminal to be checked; The power consumption anomaly analysis model includes an anomaly filtering rule, the anomaly filtering rule is provided with an anomaly rule score, and the anomaly level output module includes: A screening submodule, used for screening each type of cluster data according to the abnormal filtering rules; An abnormal data determination submodule, used for determining data satisfying the abnormal filtering rule in each type of cluster data as abnormal data; An occurrence count submodule, used to count the occurrence times of each type of abnormal data within the predetermined period; The abnormal total score calculation submodule is used to calculate the abnormal total score of the power terminal corresponding to each type of abnormal data by using the number of occurrences of each type of abnormal data, the abnormal rule score and the preset risk coefficient corresponding to the power terminal corresponding to each type of abnormal data; abnormal total score = ∑ (abnormal rule score * number of occurrences) * risk coefficient; wherein the time range of the summation is a preset period; The abnormal average score calculation submodule is used to determine the abnormal average score corresponding to the power consumption terminal corresponding to each type of abnormal data by using the total score of each type of abnormal data and the number of occurrences of each type of abnormal data; The abnormality level determination submodule is used to output the abnormality level corresponding to the power terminal based on the proportion of the average score of each type of abnormality to the total abnormality score.
5. The device according to claim 4, characterized in that The abnormal analysis model generation module includes: A historical electricity consumption behavior data acquisition submodule, used to respectively acquire historical electricity consumption behavior data within a predetermined period from the plurality of electricity consumption terminals; A historical classification data generation submodule, used to classify the historical electricity consumption behavior data using the preset grouping model to generate a variety of historical classification data; The training submodule is used to respectively train a plurality of preset initial power consumption abnormality analysis models using a plurality of the historical classification data to obtain the power consumption abnormality analysis model.
6. The device according to claim 4, characterized in that The abnormal power consumption state judgment module includes: The abnormal state determination submodule is used to determine that the power consumption terminal to be checked is in an abnormal power consumption state if the checking result is abnormal; The normal state determination submodule is used to determine that the power-consuming terminal to be checked is not in an abnormal power consumption state if the checking result is normal.
7. An electronic device, characterized in that: It comprises a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the processor executes the steps of the power consumption anomaly detection method as described in any one of claims 1-3.
8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the power consumption abnormality detection method as described in any one of claims 1 to 3 is implemented.
