Power consumer archive information anomaly detection method and device and terminal equipment

Abstract

The invention provides a power consumer archive information anomaly detection method and device and terminal equipment, and is suitable for the technical field of data processing, and the method comprises the steps: obtaining power consumer archive information; respectively extracting the power consumer archive information to obtain power consumption information of a plurality of users; based on preset power user archive sample information and a preset user archive anomaly detection model, performing prediction calculation on the plurality of pieces of user power consumption information to obtain a plurality of power consumption anomaly probability values; and when the power consumption abnormity probability value is greater than a preset abnormity probability threshold value, generating user file abnormity work order information according to the power user file information. According to the method, the power consumer archive can be automatically detected, so that the detection efficiency of the power consumer archive is improved, the detection accuracy of the abnormal power consumption condition is improved, and the normal operation of a power network is maintained.

Description

Technical Field

[0001] The present application belongs to the field of data processing technology, and in particular, relates to a method, device and terminal equipment for detecting abnormalities in archive information of power users. Background Art

[0002] In the process of electricity business processing, it is necessary to calculate the quantity and fee of user file information collected from the user end. Among them, the electricity quantity and fee review link is the most complex and time-consuming in the entire quantity and fee cycle, and it is very easy to cause errors in electricity fee calculation due to abnormal conditions in user file information that are not promptly checked.

[0003] In the prior art, the electricity charge audit method only relies on manual verification and statistical analysis to identify abnormal situations in the electricity user files. When abnormal situations are found in the electricity user files, an abnormal work order for the electricity file is manually generated according to the abnormal situation of the user. The generated abnormal work order still needs to be manually verified by the business personnel on the other side of the power grid. After the verification, a non-policy refund process is initiated for the electricity user.

[0004] However, in the current process of digital development of power business, the number of staff responsible for calculating electricity bills has been greatly reduced. The method of relying solely on manual investigation of abnormal conditions in user file information and statistics on quantity and fees is inefficient and the accuracy cannot be guaranteed, which can easily cause losses to users and power operators. Summary of the invention

[0005] In view of this, the embodiments of the present application provide a method, apparatus and terminal device for detecting anomalies in power user file information, which are used to realize automatic detection of anomalies in power user file information, aiming to solve the problem that the prior art relies solely on manual anomaly detection of power user files, which is inefficient and has a high probability of false detection and missed detection, while also bringing complex workload and heavy business processing pressure to power operation workers.

[0006] A first aspect of an embodiment of the present application provides a method for detecting abnormality in electric power user profile information, comprising:

[0007] Obtaining power user profile information;

[0008] Extracting the power user file information separately to obtain power consumption information of multiple users;

[0009] Based on the preset power user file sample information and the preset user file anomaly detection model, predict and calculate the power consumption information of multiple users respectively to obtain multiple power consumption anomaly probability values;

[0010] When the power consumption abnormality probability value is greater than a preset abnormality probability threshold, user file abnormality work order information is generated according to the power user file information.

[0011] A second aspect of an embodiment of the present application provides a device for detecting abnormality in electric power user profile information, comprising:

[0012] The power user file information acquisition module is used to obtain the power user file information;

[0013] A user electricity consumption information extraction module, used to extract the power user file information separately to obtain multiple user electricity consumption information;

[0014] A power consumption anomaly probability value calculation module is used to predict and calculate the power consumption information of multiple users based on the preset power user file sample information and the preset user file anomaly detection model, so as to obtain multiple power consumption anomaly probability values;

[0015] The user profile abnormal work order information generation module is used to generate user profile abnormal work order information according to the power user profile information when the power consumption abnormal probability value is greater than a preset abnormal probability threshold.

[0016] A third aspect of an embodiment of the present application provides a terminal device, which includes a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, the steps of the method for detecting anomalies in power user profile information as described in any one of the first aspects above are implemented.

[0017] A fourth aspect of an embodiment of the present application provides a computer-readable storage medium, comprising: storing a computer program, characterized in that when the computer program is executed by a processor, the steps of the method for detecting anomalies in power user profile information as described in any one of the first aspects above are implemented.

[0018] Compared with the prior art, the embodiments of the present application have the following beneficial effects: by extracting information from the power user file information, multiple user electricity consumption information is automatically generated, and then the user electricity consumption information is predicted and calculated through the user file anomaly detection model, and the calculated power consumption anomaly probability value is used to determine whether there is an abnormality in the power user file. When it is determined that the power user file is abnormal, a user file abnormality work order is automatically generated, thereby improving the accuracy and efficiency of detecting anomalies in the power user file information, shortening the review cycle of electricity fee accounting, and improving the quality of power network operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application, 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 application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0020] Figure 1 It is a schematic diagram of the implementation process of the power user profile information anomaly detection method provided in Example 1 of the present application;

[0021] Figure 2 This is a schematic diagram of the implementation process of the power user profile information anomaly detection method provided in Example 2 of the present application;

[0022] Figure 3 It is a schematic diagram of the implementation process of the power user profile information anomaly detection method provided in Example 3 of the present application;

[0023] Figure 4 This is a schematic diagram of the implementation process of the power user profile information anomaly detection method provided in the fourth embodiment of the present application;

[0024] Figure 5 This is a schematic diagram of the implementation process of the power user profile information anomaly detection method provided in Example 5 of the present application;

[0025] Figure 6 This is a schematic diagram of the implementation process of the power user profile information anomaly detection method provided in Example 5 of the present application;

[0026] Figure 7 It is a structural schematic diagram of a device for detecting abnormality in electric power user profile information provided in an embodiment of the present application;

[0027] Figure 8 It is a schematic diagram of a terminal device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0028] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0029] In order to illustrate the technical solution described in this application, a specific embodiment is provided below for illustration.

[0030] Figure 1The following is a flowchart of the method for detecting abnormality in the power user profile information provided in the first embodiment of the present application, which is described in detail as follows:

[0031] Step S101, obtaining power user profile information.

[0032] In this embodiment, the power user profile information can be obtained from the marketing system of the power network. The power user profile information may include customer snapshots, installation point snapshots, service location snapshots, customer agreement snapshots, customer pipeline snapshots, voltage regulation equipment snapshots, and connection snapshots.

[0033] Step S102: extract the power user profile information separately to obtain power consumption information of multiple users.

[0034] In this embodiment, the power user file information may be parsed, irrelevant data may be eliminated, and then the electricity bill settlement related data may be extracted, and the extracted information may be generated into multiple wide table files according to different contents as the user's electricity consumption information.

[0035] In this embodiment, preferably, the user's electricity consumption information includes electricity power factor information, demand electricity charge information, basic electricity charge information, transformer power loss information and power grid line loss information. The electricity power factor information includes the metering point number, metering point level, user electricity category, user operating capacity, power factor assessment method, power factor assessment standard, electricity category of the metering point electricity price, electricity price industry classification of the metering point electricity price, metering point capacity, total forward reactive power and total reverse reactive power. The demand electricity charge information includes the user's electricity category, user operating capacity, user account establishment date, pricing strategy, metering point electricity price, electricity price industry classification of the metering point electricity price, basic electricity charge calculation method (basic electricity charge calculated according to demand), metering point area number, transformer area number, transformer operating status, power supply nature, user contract capacity, and demand verification value. Basic electricity fee information includes user electricity consumption category, user operating capacity, user account establishment date, pricing strategy, metering point execution price, metering point execution price industry classification, basic electricity fee calculation method (basic electricity fee calculated by capacity), metering point area number, transformer area number. Transformer power loss information includes metering point level, metering point metering method, transformer loss sharing mark, transformer loss billing mark, transformer transformer loss calculation method, transformer loss sharing agreement value, transformer loss number, transformer area number, metering point area number, area status, area power supply unit, transformer public and private mark. Power grid line loss information includes metering point line number, line relationship, line loss calculation mark, line loss sharing mark, line loss calculation method code, line loss sharing agreement value, line power supply unit, line status, unit length line reactance, unit length line resistance, conductor length, public and private line mark, total forward reactive power, and total reverse reactive power. Electricity power factor information, demand charge information, basic charge information, transformer power loss information, and power grid line loss information can all be presented in the form of a wide table.

[0036] Step S103, based on the preset power user file sample information and the preset user file anomaly detection model, predictive calculations are performed on the power consumption information of the plurality of users respectively to obtain a plurality of power consumption anomaly probability values.

[0037] In this embodiment, the preset user profile anomaly detection model can be a neural network model, specifically, a random forest model. It is understandable that the user's electricity consumption information needs to be pre-processed before it can be input into the user profile anomaly detection model for calculation. The pre-processing includes cleaning the data, filling in the missing data items in the data, marking and removing the outliers in the data, and then inserting it into the user profile anomaly detection model to continue data processing after processing such as feature scaling and normalization. The preset power user profile sample information can be past power user profiles collected from the marketing system, which is used to train the neural network so that the trained neural network can meet the prediction requirements for user electricity consumption information. When the neural network training is completed, the currently collected user electricity consumption information is input into the trained neural network for calculation, and the calculation error result is the probability of abnormal electricity consumption, where the error result can be the mean square error.

[0038] Step S104, determining whether the power consumption abnormality probability value is greater than a preset abnormal probability threshold; if so, generating user file abnormal work order information according to the power user file information; if not, generating user file normal work order information according to the power user file information.

[0039] In this embodiment, the preset abnormal probability threshold can be set manually. When the abnormal power consumption probability value is greater than the abnormal probability threshold, it indicates that there is an abnormal situation in the power user file corresponding to the abnormal power consumption probability value, so a user file abnormal work order can be generated through the power user file, and then the work order information is sent to the power user terminal after manual review. When the abnormal power consumption probability value is less than or equal to the abnormal probability threshold, it indicates that there is no abnormal situation in the power user file corresponding to the abnormal power consumption probability value, and then the user file normal work order information is generated to indicate that the power user file is normal, and there is no need to generate an abnormal work order and notify the power user.

[0040] The method for detecting anomalies in electric power user file information provided in the embodiment of the present application automatically generates multiple user electricity usage information by extracting information from the electric power user file information, and then predicts and calculates the user electricity usage information through a user file anomaly detection model, and judges whether the electric power user file is very likely to have an anomaly through the calculated electricity usage anomaly probability value. When it is determined that the electric power user file is abnormal, a user file anomaly work order is automatically generated, thereby improving the accuracy and efficiency of detecting anomalies in the electric power user file information, shortening the review cycle of electricity fee accounting, and improving the quality of power network operation.

[0041] Figure 2The flowchart of the implementation of the power user profile information anomaly detection method provided in the second embodiment of the present application is shown. The difference between the second embodiment and the first embodiment is that the preset power user profile sample information includes a plurality of preset user power consumption sample information; the step S103 specifically includes:

[0042] Step S201, extracting parameters of a preset user profile anomaly detection model to obtain a user profile anomaly detection model parameter set.

[0043] In this embodiment, it can be understood that the preset user profile anomaly detection model can be a random forest model, and the extracted parameters of the random forest model can be the number of trees, the maximum depth, the minimum number of sample splits, etc. The extracted parameters are aggregated into a set to generate a user profile anomaly detection model parameter set.

[0044] Step S202: Generate parameter list information according to the user profile anomaly detection model parameter set and the preset parameter candidate value range.

[0045] In this embodiment, the preset parameter candidate value range can be manually set. By adding and subtracting the user profile anomaly detection model parameters from the preset parameter candidate value range, the upper limit value, lower limit value and multiple intermediate values ​​of the user profile anomaly detection model parameters are obtained respectively, and a list is generated according to the user profile anomaly detection model parameters and their upper limit value, lower limit value and intermediate values ​​for subsequent training of the user profile anomaly detection model, that is, generating parameter list information.

[0046] Step S203, obtaining a trained user profile anomaly detection model according to the parameter list information, multiple preset user electricity usage sample information and a preset user profile anomaly detection model.

[0047] In this embodiment, the preset user power usage sample information and parameter list information may be input into the preset user profile anomaly detection model to train the user profile anomaly detection model. When the mean square error value calculated by the model is less than the preset mean square error threshold, it indicates that the model training is completed, and the trained user profile anomaly detection model is obtained. The preset mean square error threshold may be 90%.

[0048] In this embodiment, it is understandable that different user profile anomaly detection models are trained for different user electricity usage information to ensure the accuracy of detecting different user electricity usage information. For example, for detecting electricity power factor information, it is necessary to train the preset user profile anomaly detection model with the preset electricity power factor sample information, and after training, a model is obtained that is only used to predict the occurrence of electricity power factor; for detecting demand electricity fee information, it is necessary to train the preset user profile anomaly detection model with the preset demand electricity fee sample information, and after training, a model is obtained that is only used to predict the demand electricity fee expenditure.

[0049] Step S204, predicting and calculating the power usage information of the plurality of users according to the trained user profile anomaly detection model to obtain a plurality of power usage anomaly probability values.

[0050] In this embodiment, the electricity usage information of multiple users is respectively input into multiple trained user profile anomaly detection models, the user electricity usage information is predicted and calculated using the model parameters adjusted during the training process, and the electricity usage anomaly probability value is calculated using the calculated user electricity usage information prediction value and the user electricity usage information true value, which is used to subsequently determine whether there is any abnormality in the electricity user profile.

[0051] The method for detecting anomalies in electric power user profile information provided in the embodiment of the present application extracts parameters of a user profile anomaly detection model, summarizes the extracted parameters to generate a parameter list as the initial parameters of the model, and trains the user profile anomaly detection model through pre-collected user electricity usage samples, thereby adjusting the initial parameters of the model, so that the user profile anomaly detection model has high accuracy in prediction and decision-making, thereby improving the efficiency and effect of anomaly detection on user profile information, so as to ensure the normal operation of the power network.

[0052] Figure 3 The flowchart of the implementation of the power user profile information anomaly detection method provided in the third embodiment of the present application is shown. The difference between the third embodiment and the second embodiment is that the step S203 specifically includes:

[0053] Step S301, performing parameter combination extraction on the parameter list information to obtain first parameter combination information and second parameter combination information.

[0054] In this embodiment, it can be understood that the parameter list information can be a plurality of value combinations of a plurality of parameters of the user profile anomaly detection model, and different parameter value combinations and different user electricity usage samples are used to perform calculations for evaluating the model performance. The parameter combinations in the parameter list information are extracted to obtain a plurality of parameter combinations, including the first parameter combination information and the second parameter combination information.

[0055] Step S302: divide a plurality of preset user power usage sample information to obtain first user power usage sample information and second user power usage sample information.

[0056] In this embodiment, multiple preset user electricity usage sample information are divided to divide the user electricity usage samples into a training set and a test set, wherein 80% of the user electricity usage sample information can be divided into the first user electricity usage sample information as the training set, and the remaining 20% ​​can be divided into the second user electricity usage sample information as the test set.

[0057] Step S303: obtaining first user profile detection information according to the first user's electricity usage sample information, the first parameter combination information, and a preset user profile anomaly detection model.

[0058] In this embodiment, the parameters in the first parameter combination information are used as parameters of the user profile anomaly detection model, and the first user's electricity usage sample information is input into the user profile anomaly detection model having the parameters in the first parameter combination information for calculation, and the calculation result is the first user profile detection information.

[0059] Step S304: obtaining second user profile detection information according to the second user's electricity usage sample information, the first parameter combination information, and a preset user profile anomaly detection model.

[0060] In this embodiment, the parameters in the first parameter combination information are used as parameters of the user profile anomaly detection model, and the second user's electricity usage sample information is input into the user profile anomaly detection model with the parameters in the first parameter combination information for calculation, and the calculation result is the second user profile detection information.

[0061] Step S305: obtaining third user profile detection information according to the first user power usage sample information, the second parameter combination information and a preset user profile anomaly detection model.

[0062] In this embodiment, the parameters in the second parameter combination information are used as parameters of the user profile anomaly detection model, and the first user's electricity usage sample information is input into the user profile anomaly detection model with the parameters in the second parameter combination information for calculation, and the calculation result is the third user profile detection information.

[0063] Step S306: Obtain fourth user profile detection information according to the second user's electricity usage sample information, the second parameter combination information, and a preset user profile anomaly detection model.

[0064] In this embodiment, the parameters in the second parameter combination information are used as parameters of the user profile anomaly detection model, and the second user's electricity usage sample information is input into the user profile anomaly detection model with the parameters in the second parameter combination information for calculation, and the calculation result is the fourth user profile detection information.

[0065] Step S307: obtaining a trained user profile anomaly detection model according to the first user profile detection information, the second user profile detection information, the third user profile detection information, the fourth user profile detection information and a preset user profile anomaly detection model.

[0066] In this embodiment, the accuracy of the user profile anomaly detection model corresponding to the first parameter combination information may be calculated by using the first user profile detection information and the second user profile detection information, and the accuracy of the user profile anomaly detection model corresponding to the second parameter combination information may be calculated by using the third user profile detection information and the fourth user profile detection information. By comparing the two accuracy rates, the user profile anomaly detection model corresponding to a set of parameters with a high accuracy rate is used as the trained user profile anomaly detection model.

[0067] In this embodiment, the performance of the trained model may also be evaluated by calculating the recall rate and F1 score of the user profile anomaly detection model to ensure the stability and consistency of the user profile anomaly detection model on different data sets.

[0068] The method for detecting anomalies in power user profile information provided in the embodiment of the present application calculates user profile detection information by selecting multiple parameter combination information, and is used to evaluate the calculation accuracy of the user profile anomaly detection model during the training process through the calculated user profile detection information, and selects the parameter combination that optimizes the prediction performance of the user profile anomaly detection model from multiple groups of parameter combination information, thereby improving the stability and reliability of the trained user profile anomaly detection model.

[0069] Figure 4 The flowchart of the method for detecting abnormality in the power user profile information provided in the fourth embodiment of the present application is shown. The difference between the method and the third embodiment is that the step S307 specifically includes:

[0070] Step S401: Calculate the first user profile detection accuracy rate according to the first user profile detection information and the second user profile detection information.

[0071] In this embodiment, the first user profile detection information may be divided by the second user profile detection information to obtain the first user profile detection accuracy rate.

[0072] Step S402: Calculate the second user profile detection accuracy rate based on the third user profile detection information and the fourth user profile detection information.

[0073] In this embodiment, the second user profile detection accuracy rate may be obtained by dividing the third user profile detection information by the fourth user profile detection information.

[0074] Step S403, determining whether the first user profile detection accuracy is greater than the second user profile detection accuracy; if so, proceeding to step S404; if not, proceeding to step S405.

[0075] In this embodiment, when the first user profile detection accuracy is greater than the second user profile detection accuracy, it means that when the user profile anomaly detection model uses the first parameter combination information, the performance of the user profile anomaly detection model is better than when the second parameter combination information is used for the user profile anomaly detection model, and the use of the first parameter combination information can make the prediction result of the user's electricity usage information more accurate and reliable. When the first user profile detection accuracy is less than or equal to the second user profile detection accuracy, it means that when the user profile anomaly detection model uses the second parameter combination information, the performance of the user profile anomaly detection model is better than when the first parameter combination information is used for the user profile anomaly detection model, and the use of the second parameter combination information can make the prediction result of the user's electricity usage information more accurate and reliable.

[0076] Step S404: Using the first parameter combination information as a parameter combination of a preset user profile anomaly detection model to obtain a trained user profile anomaly detection model.

[0077] In this embodiment, when the first user profile detection accuracy is greater than the second user profile detection accuracy, it means that when the user profile anomaly detection model uses the first parameter combination information, the performance of the user profile anomaly detection model is better than when the second parameter combination information is used for the user profile anomaly detection model. Using the first parameter combination information can make the prediction result of the user's electricity consumption information more accurate and reliable. Therefore, the first parameter combination information is used as the model parameter of the user profile anomaly detection model. At this point, the training is completed, and the trained user profile anomaly detection model is obtained.

[0078] Step S405: Using the second parameter combination information as a parameter combination of a preset user profile anomaly detection model to obtain a trained user profile anomaly detection model.

[0079] In this embodiment, when the first user profile detection accuracy is less than or equal to the second user profile detection accuracy, it means that when the user profile anomaly detection model uses the second parameter combination information, the performance of the user profile anomaly detection model is better than when the first parameter combination information is used for the user profile anomaly detection model. The use of the second parameter combination information can make the prediction results of the user's electricity consumption information more accurate and reliable. Therefore, the second parameter combination information is used as the model parameter of the user profile anomaly detection model. At this point, the training is completed, and the trained user profile anomaly detection model is obtained.

[0080] In this embodiment, the trained user profile anomaly detection model can be encapsulated and integrated into the marketing system of the power network through a RESTful API interface, thereby realizing the automated anomaly detection and review process of power user profile information.

[0081] The method for detecting anomalies in power user profile information provided in the embodiment of the present application calculates the prediction accuracy of the user profile anomaly detection model corresponding to different model parameters, and determines a set of parameters with the best performance as the parameters after model training, thereby ensuring the prediction accuracy of the user profile anomaly detection model, and is used to timely identify abnormal situations in user profiles, reduce the workload of staff, and promptly notify power service providers and users to take measures to deal with abnormal situations, so as to ensure the normal operation of the power network.

[0082] Figure 5 The flowchart of the implementation of the power user profile information anomaly detection method provided in the fifth embodiment of the present application is shown. The difference between the fifth embodiment and the second embodiment is that the step S204 specifically includes:

[0083] Step S501: extracting features from a plurality of pieces of user electricity usage information to obtain user electricity usage feature information.

[0084] In this embodiment, an autoencoder method may be used to extract features from user electricity usage information, and the extracted features are used as user electricity usage feature information for subsequent detection, analysis and calculation.

[0085] Step S502, performing multiple sampling processes on the user power consumption characteristic information according to a preset number of characteristic sampling times to obtain multiple sets of user power consumption sampling information; the user power consumption sampling information sets include multiple power consumption sampling information.

[0086] In this embodiment, the preset feature sampling times can be manually set. The user power consumption feature information is sampled to calculate the scattered values ​​of the user power consumption information, and the sampled data is aggregated into a user power consumption sampling information set, which is used to evaluate the feature importance of each user power consumption information in the future.

[0087] Step S503 , calculating the power consumption information gain corresponding to each user power consumption sampling information set according to the preset feature sampling times, the number of power consumption sampling information in the user power consumption sampling information set, and multiple power consumption sampling information.

[0088] In this embodiment, the preset feature sampling times can be manually set. It can be done by dividing the preset feature sampling times and the number of power sampling information, taking the result of the division as a coefficient, and then performing a logarithmic operation on the coefficient to obtain a logarithmic calculation result, and then multiplying the coefficient with the logarithmic calculation result to obtain the power information gain corresponding to the user power sampling information set. It can be understood that the greater the information entropy of a data feature, the greater the importance of the feature information contained in the data feature.

[0089] Step S504, determining whether the power consumption information gain is greater than a preset power consumption information gain threshold; if so, proceeding to step S505; if not, not using the user power consumption sampling information set corresponding to the power consumption information gain as user power consumption prediction information.

[0090] In this embodiment, the preset power consumption information gain threshold can be set manually. When the power consumption information gain is greater than the preset power consumption information gain threshold, it means that the user power consumption sampling information set corresponding to the power consumption information gain has a more important data feature, and has a greater reference value for abnormal situations of user power consumption information, and the user power consumption sampling information set corresponding to the power consumption information gain is used as user power consumption prediction information. When the power consumption information gain is less than or equal to the preset power consumption information gain threshold, it means that the data feature of the user power consumption sampling information set corresponding to the power consumption information gain is not of high importance, and has a small reference value for abnormal situations of user power consumption information, and the user power consumption sampling information set corresponding to the power consumption information gain is not used as user power consumption prediction information.

[0091] Step S505: taking the user power consumption sampling information set corresponding to the power consumption information gain as user power consumption prediction information.

[0092] In this embodiment, when the electricity consumption information gain is greater than the preset electricity consumption information gain threshold, it means that the user electricity consumption sampling information set corresponding to the electricity consumption information gain has a more important data feature and is of great reference value for abnormal situations of user electricity consumption information. Therefore, the user electricity consumption sampling information set corresponding to the electricity consumption information gain is used as user electricity consumption prediction information.

[0093] Step S506, calculating the power consumption anomaly probability value according to the user power consumption prediction information, the user power consumption information and the trained user profile anomaly detection model.

[0094] In this embodiment, the trained user profile anomaly detection model may be a random forest model, and the power consumption anomaly probability value may be a mean square error. The power consumption information of the user may be input into the trained user profile anomaly detection model for prediction calculation, and the prediction calculation result and the previously calculated user power consumption prediction information may be used to calculate the mean square error to obtain the power consumption anomaly probability value.

[0095] The method for detecting anomalies in power user profile information provided in the embodiment of the present application performs multiple samplings on the data features of the user's power usage information, thereby dispersing the overall user power usage information into various local information and performing separate data feature importance assessments to ensure the effectiveness and accuracy of the output results of the user profile anomaly detection model.

[0096] Figure 6 The flowchart of the implementation of the power user profile information anomaly detection method provided in the sixth embodiment of the present application is shown. The difference between the sixth embodiment and the fifth embodiment is that the step S503 specifically includes:

[0097] Step S601, calculating the user's power consumption experience entropy information according to a preset feature sampling number and the number of power consumption sampling information in the user's power consumption sampling information set.

[0098] In this embodiment, the preset feature sampling times can be manually set. The feature sampling times can be divided by the number of power consumption sampling information in the user power consumption sampling information set to calculate the user power consumption experience entropy information, so as to prepare for the subsequent calculation of power consumption information gain.

[0099] Step S602: Calculate the user's power usage experience conditional entropy information based on the user's power usage experience entropy information and the power usage sampling information.

[0100] In this embodiment, the user's power consumption experience entropy information may be subjected to a logarithmic operation, and then the result of the logarithmic operation is multiplied by the power consumption sampling information to obtain the user's power consumption experience conditional entropy information.

[0101] Step S603: Calculate the power consumption information gain corresponding to each user power consumption sampling information set according to the user power consumption experience entropy information and the user power consumption experience conditional entropy information.

[0102] In this embodiment, the power consumption information gain corresponding to the user power consumption sampling information set may be obtained by dividing the user power consumption experience entropy information by the user power consumption experience conditional entropy information.

[0103] The method for detecting anomalies in power user file information provided in the embodiment of the present application quantifies the importance and reliability of multiple local user power usage information by gradually calculating the power usage information gain for multiple local user power usage information obtained by sampling, thereby ensuring the accuracy of the output results of the user file anomaly detection model, helping staff to quickly identify and lock in anomalies in user files, and ensuring the safety and stability of power network operations.

[0104] Corresponding to the method of the above embodiment, Figure 7 A structural block diagram of an apparatus for detecting anomalies in profile information of electric power users provided in an embodiment of the present application is shown. For ease of explanation, only the parts related to the embodiment of the present application are shown. Figure 7 The exemplary power user profile information anomaly detection device may be the execution subject of the power user profile information anomaly detection method provided in the aforementioned first embodiment.

[0105] Reference Figure 7 , the power user profile information anomaly detection device comprises:

[0106] The power user profile information acquisition module 710 is used to acquire power user profile information;

[0107] The user power consumption information extraction module 720 is used to extract the power user file information separately to obtain power consumption information of multiple users;

[0108] The power consumption anomaly probability value calculation module 730 is used to perform prediction calculations on the power consumption information of a plurality of users based on the preset power user file sample information and the preset user file anomaly detection model to obtain a plurality of power consumption anomaly probability values;

[0109] The user profile abnormal work order information generation module 740 is used to generate user profile abnormal work order information according to the power user profile information when the power consumption abnormal probability value is greater than a preset abnormal probability threshold.

[0110] The process of each module in the power user profile information anomaly detection device provided in the embodiment of the present application realizing its own function can be specifically referred to the aforementioned Figure 1 The description of the first embodiment is not repeated here.

[0111] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0112] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0113] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0114] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0115] In addition, in the description of the present specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish descriptions, and cannot be understood as indicating or suggesting relative importance. It should also be understood that although the terms "first", "second", etc. are used to describe various elements in some embodiments of the present application in the text, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, the first table can be named as the second table, and similarly, the second table can be named as the first table without departing from the scope of the various described embodiments. The first table and the second table are both tables, but they are not the same table.

[0116] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0117] The power user profile information anomaly detection method provided in the embodiment of the present application can be applied to terminal devices such as mobile phones, tablet computers, wearable devices, vehicle-mounted devices, augmented reality (AR) / virtual reality (VR) devices, laptops, ultra-mobile personal computers (UMPC), netbooks, personal digital assistants (PDA), etc. The embodiment of the present application does not impose any restrictions on the specific type of terminal devices.

[0118] For example, the terminal device can be a station (STAION, ST) in a WLAN, a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA) device, a handheld device with wireless communication function, a computing device or other processing device connected to a wireless modem, a vehicle-mounted device, a vehicle networking terminal, a computer, a laptop computer, a handheld communication device, a handheld computing device, a satellite wireless device, a wireless modem card, a TV set-top box (settop box, STB), customer premises equipment (customer premises equipment, CPE) and / or other devices for communicating on a wireless system and a next-generation communication system, such as a mobile terminal in a 5G network or a mobile terminal in a future evolved Public Land Mobile Network (PLMN) network, etc.

[0119] As an example but not limitation, when the terminal device is a wearable device, the wearable device can also be a general term for wearable devices that are intelligently designed and developed using wearable technology for daily wear, such as glasses, gloves, watches, clothing and shoes. A wearable device is a portable device that is worn directly on the body or integrated into the user's clothes or accessories. Wearable devices are not just hardware devices, but also achieve powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable smart devices include those that are fully functional, large in size, and can achieve complete or partial functions without relying on smartphones, such as smart watches or smart glasses, as well as those that only focus on a certain type of application function and need to be used in conjunction with other devices such as smartphones, such as various smart bracelets and smart jewelry for vital sign monitoring.

[0120] Figure 8Schematic diagram of the structure of a terminal device provided by an embodiment of the present application. Figure 8 As shown, the terminal device 8 of this embodiment includes: at least one processor 80 ( Figure 8 Only one is shown in the figure), a memory 81, wherein the memory 81 stores a computer program 82 that can be run on the processor 80. When the processor 80 executes the computer program 82, the steps in the above-mentioned various power user profile information abnormality detection method embodiments are implemented, such as Figure 1 Alternatively, when the processor 70 executes the computer program 72, the functions of each module / unit in the above-mentioned device embodiments are realized, for example Figure 7 Functions of modules 710 to 740 are shown.

[0121] The terminal device 8 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The terminal device may include, but is not limited to, a processor 80 and a memory 81. Those skilled in the art will appreciate that Figure 8 It is only an example of the terminal device 8 and does not constitute a limitation of the terminal device 8. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal device may also include an input sending device, a network access device, a bus, etc.

[0122] The processor 80 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0123] In some embodiments, the memory 81 may be an internal storage unit of the terminal device 8, such as a hard disk or memory of the terminal device 8. The memory 81 may also be an external storage device of the terminal device 8, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the terminal device 8. Further, the memory 81 may also include both an internal storage unit of the terminal device 8 and an external storage device. The memory 81 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of the computer program, etc. The memory 81 may also be used to temporarily store data that has been sent or is to be sent.

[0124] In addition, each functional unit in each embodiment of the present application 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.

[0125] An embodiment of the present application also provides a terminal device, which includes at least one memory, at least one processor, and a computer program stored in the at least one memory and executable on the at least one processor, wherein when the processor executes the computer program, the terminal device implements the steps in any of the above-mentioned method embodiments.

[0126] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0127] An embodiment of the present application provides a computer program product. When the computer program product runs on a terminal device, the terminal device can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0128] If the integrated module / 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 present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0129] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0130] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0131] 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.

[0132] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person 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 may be replaced by equivalents. Such 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 application, and should all be included in the protection scope of the present application.

Claims

1. A method for detecting abnormality in power user profile information, characterized in that: include: Obtaining power user profile information; Extracting the power user file information separately to obtain power consumption information of multiple users; Based on the preset power user file sample information and the preset user file anomaly detection model, predict and calculate the power consumption information of multiple users respectively to obtain multiple power consumption anomaly probability values; When the power consumption abnormality probability value is greater than a preset abnormality probability threshold, user file abnormality work order information is generated according to the power user file information.

2. The method for detecting abnormality in electric power user profile information according to claim 1, characterized in that: The user electricity consumption information includes electricity power factor information, demand electricity fee information, basic electricity fee information, transformer power loss information and power grid line loss information.

3. The method for detecting abnormality in electric power user profile information according to claim 1, characterized in that: The preset power user file sample information includes a plurality of preset user power consumption sample information; The step of performing prediction calculations on the power consumption information of a plurality of users based on the preset power user profile sample information and the preset user profile anomaly detection model to obtain a plurality of power consumption anomaly probability values ​​specifically includes: Extract parameters of the preset user profile anomaly detection model to obtain a user profile anomaly detection model parameter set; Generate parameter list information according to the user profile anomaly detection model parameter set and the preset parameter candidate value range; Obtaining a trained user profile anomaly detection model according to the parameter list information, a plurality of preset user power usage sample information, and a preset user profile anomaly detection model; According to the trained user profile anomaly detection model, a plurality of the user's electricity usage information is predicted and calculated to obtain a plurality of power usage anomaly probability values.

4. The method for detecting abnormality in electric power user profile information according to claim 3, characterized in that: The step of obtaining a trained user profile anomaly detection model according to the parameter list information, a plurality of preset user power usage sample information and a preset user profile anomaly detection model specifically includes: Performing parameter combination extraction on the parameter list information to obtain first parameter combination information and second parameter combination information; Dividing a plurality of preset user power usage sample information to obtain first user power usage sample information and second user power usage sample information; Obtaining first user profile detection information according to the first user's electricity usage sample information, the first parameter combination information, and a preset user profile anomaly detection model; Obtaining second user profile detection information according to the second user's power usage sample information, the first parameter combination information, and a preset user profile anomaly detection model; Obtaining third user profile detection information according to the first user power usage sample information, the second parameter combination information, and a preset user profile anomaly detection model; Obtaining fourth user profile detection information according to the second user's electricity usage sample information, the second parameter combination information, and a preset user profile anomaly detection model; A trained user profile anomaly detection model is obtained according to the first user profile detection information, the second user profile detection information, the third user profile detection information, the fourth user profile detection information and a preset user profile anomaly detection model.

5. The method for detecting abnormality in electric power user profile information according to claim 4, characterized in that: The step of obtaining a trained user profile anomaly detection model according to the first user profile detection information, the second user profile detection information, the third user profile detection information, the fourth user profile detection information and a preset user profile anomaly detection model specifically includes: Calculating the first user profile detection accuracy rate according to the first user profile detection information and the second user profile detection information; Calculating the second user profile detection accuracy rate according to the third user profile detection information and the fourth user profile detection information; Determining whether the first user profile detection accuracy is greater than the second user profile detection accuracy; If yes, the first parameter combination information is used as a parameter combination of a preset user profile anomaly detection model to obtain a trained user profile anomaly detection model; If not, the second parameter combination information is used as the parameter combination of the preset user profile anomaly detection model to obtain a trained user profile anomaly detection model.

6. The method for detecting abnormality in electric power user profile information according to claim 3, characterized in that: The step of performing prediction calculation on the power consumption information of the plurality of users according to the trained user profile anomaly detection model to obtain a plurality of power consumption anomaly probability values ​​specifically includes: Extracting features from the plurality of user power usage information to obtain user power usage feature information; According to a preset number of characteristic sampling times, the user's power consumption characteristic information is sampled multiple times to obtain multiple sets of user power consumption sampling information; the user power consumption sampling information set includes multiple power consumption sampling information; Calculate the power consumption information gain corresponding to each user power consumption sampling information set according to the preset feature sampling times, the number of power consumption sampling information in the user power consumption sampling information set, and multiple power consumption sampling information; When the power consumption information gain is greater than a preset power consumption information gain threshold, a user power consumption sampling information set corresponding to the power consumption information gain is used as user power consumption prediction information; The power consumption anomaly probability value is calculated based on the user power consumption prediction information, the user power consumption information and the trained user profile anomaly detection model.

7. The method for detecting abnormality in electric power user profile information according to claim 6, characterized in that: The step of calculating the power consumption information gain corresponding to each user power consumption sampling information set according to the preset feature sampling times, the number of power consumption sampling information in the user power consumption sampling information set, and the multiple power consumption sampling information specifically includes: Calculate the user's power consumption experience entropy information according to the preset feature sampling times and the number of power consumption sampling information in the user's power consumption sampling information set; Calculating the user's power usage experience conditional entropy information based on the user's power usage experience entropy information and the power usage sampling information; The power consumption information gain corresponding to each user power consumption sampling information set is calculated according to the user power consumption experience entropy information and the user power consumption experience conditional entropy information.

8. A device for detecting abnormality in electric power user profile information, characterized in that: include: The power user file information acquisition module is used to obtain the power user file information; A user electricity consumption information extraction module, used to extract the power user file information separately to obtain multiple user electricity consumption information; A power consumption anomaly probability value calculation module is used to predict and calculate the power consumption information of multiple users based on the preset power user file sample information and the preset user file anomaly detection model, so as to obtain multiple power consumption anomaly probability values; The user profile abnormal work order information generation module is used to generate user profile abnormal work order information according to the power user profile information when the power consumption abnormal probability value is greater than a preset abnormal probability threshold.

9. A terminal device, characterized in that: The terminal device includes a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images