Abnormal power data identification method and device based on power edge terminal

Abstract

The embodiment of the invention discloses an abnormal power data identification method and device based on a power edge terminal. A specific embodiment of the method comprises the following steps: in response to a received abnormal power data identification request sent by a power edge terminal, obtaining to-be-detected power information sent by the power edge terminal; performing equipment abnormity identification on each preprocessed power equipment image in the preprocessed power equipment image set; inputting the to-be-detected power consumption data set into a pre-trained power consumption abnormity identification model to obtain a power consumption abnormity factor information set; on the basis of the power utilization abnormal factor information set, updating each to-be-detected power utilization user identifier to obtain a target power utilization user identifier group; and inputting the user electricity consumption monitoring data group into a pre-trained electricity consumption abnormity type detection model to obtain a user electricity consumption abnormity type group. According to the embodiment, the efficiency and accuracy of power data detection are improved.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the computer field, and in particular to a method and device for identifying abnormal power data based on a power edge terminal. Background Art

[0002] With the continuous growth of global energy demand and the increasingly severe environmental problems, smart grid technology, as a key technology to achieve efficient energy utilization and optimal configuration, has become an important direction for the development of global power systems. Smart grids integrate advanced information and communication technologies, automation control technologies, and power electronics technologies to achieve intelligent, interactive, and clean power grids. In this context, as an important part of smart grids, the research on wireless access scenarios of power edge access smart devices has important practical significance and broad application prospects.

[0003] With the popularization of Internet of Things technology, the number of intelligent devices in the power system has increased dramatically, generating massive amounts of data. How to efficiently process and analyze this data and extract valuable information is important for optimizing power grid operation, improving energy efficiency, and achieving demand-side management.

[0004] At present, the method usually adopted for detecting the data collected by the power edge terminal is: sending different types of power data to be detected to the corresponding power data processing terminal for abnormal analysis. However, this method reduces the efficiency of power abnormality analysis and is not conducive to the maintenance of the power grid.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the inventive concept and therefore it may contain information that does not form the prior art that is already known in this country to a person of ordinary skill in the art. Summary of the invention

[0006] The content of this disclosure is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this disclosure is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0007] Some embodiments of the present disclosure propose an abnormal power data identification method, device, electronic device and computer-readable medium based on a power edge terminal to solve one or more of the technical problems mentioned in the above background technology section.

[0008] In a first aspect, some embodiments of the present disclosure provide a method for identifying abnormal power data based on a power edge terminal, the method comprising: in response to receiving an abnormal power data identification request sent by the power edge terminal, obtaining the power information to be detected sent by the above-mentioned power edge terminal, wherein the above-mentioned power information to be detected comprises: a set of images of power equipment to be detected, a set of power consumption data to be detected, the power consumption data to be detected in the power consumption data set to be detected corresponds to power users, and the power equipment images to be detected in the set of power equipment images to be detected correspond to the power equipment to be detected; preprocessing each power equipment image to be detected in the above-mentioned set of power equipment images to be detected to generate a preprocessed power equipment image to obtain a set of preprocessed power equipment images; performing device abnormality identification on each preprocessed power equipment image in the above-mentioned set of preprocessed power equipment images to generate a device abnormality identification result to obtain a set of device abnormality identification results; inputting the above-mentioned power consumption data set to be detected into a pre-trained power consumption abnormality identification system; In the identification model, an abnormal power consumption factor information set is obtained, wherein the power consumption data to be detected in the above-mentioned power consumption data set to be detected corresponds to the abnormal power consumption factor information in the above-mentioned abnormal power consumption factor information set; based on the above-mentioned abnormal power consumption factor information set, each power user identification to be detected is updated to obtain a target power user identification group, wherein one power user identification to be detected corresponds to one power user; the user power consumption monitoring data corresponding to each target power user identification in the above-mentioned target power user identification group is obtained to obtain a user power consumption monitoring data group; the above-mentioned user power consumption monitoring data group is input into a pre-trained power consumption abnormality type detection model to obtain a user power consumption abnormality type group, wherein one user power consumption monitoring data corresponds to one user power consumption abnormality type; the above-mentioned user power consumption abnormality type group and the corresponding user identifications are sent to the associated user power consumption abnormality management terminal, and the above-mentioned equipment abnormality identification result set is sent to the associated power equipment abnormality management terminal.

[0009] In a second aspect, some embodiments of the present disclosure provide an abnormal power data identification device based on a power edge terminal, the device comprising: a first acquisition unit, configured to, in response to receiving an abnormal power data identification request sent by the power edge terminal, obtain the power information to be detected sent by the above-mentioned power edge terminal, wherein the above-mentioned power information to be detected comprises: a set of power equipment images to be detected, a set of power consumption data to be detected, the power consumption data to be detected in the power consumption data set to be detected corresponds to the power user, and the power equipment images to be detected in the set of power equipment images to be detected correspond to the power equipment to be detected; a preprocessing unit, configured to preprocess each power equipment image to be detected in the above-mentioned set of power equipment images to be detected, so as to generate a preprocessed power equipment image and obtain a set of preprocessed power equipment images; an identification unit, configured to perform device abnormality identification on each preprocessed power equipment image in the above-mentioned set of preprocessed power equipment images, so as to generate a device abnormality identification result and obtain a set of device abnormality identification results; a first input unit, configured to input the above-mentioned power consumption data set to be detected into a pre-trained power consumption abnormality identification system. A power consumption anomaly factor information set is obtained in the identification model, wherein the power consumption data to be detected in the above-mentioned power consumption data set to be detected corresponds to the power consumption anomaly factor information in the above-mentioned power consumption anomaly factor information set; an updating unit is configured to update each power user identifier to be detected based on the above-mentioned power consumption anomaly factor information set to obtain a target power user identifier group, wherein one power user identifier to be detected corresponds to one power user; a second acquisition unit is configured to obtain user power consumption monitoring data corresponding to each target power user identifier in the above-mentioned target power user identifier group to obtain a user power consumption monitoring data group; a second input unit is configured to input the above-mentioned user power consumption monitoring data group into a pre-trained power consumption anomaly type detection model to obtain a user power consumption anomaly type group, wherein one user power consumption monitoring data corresponds to one user power consumption anomaly type; a sending unit is configured to send the above-mentioned user power consumption anomaly type group and the corresponding each user identifier to the associated user power consumption anomaly management terminal, and send the above-mentioned equipment anomaly identification result set to the associated power equipment anomaly management terminal.

[0010] In a third aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner of the above-mentioned first aspect.

[0011] In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in any implementation manner of the above-mentioned first aspect is implemented.

[0012] The above-mentioned embodiments of the present disclosure have the following beneficial effects: through the abnormal power data identification method based on the power edge terminal of some embodiments of the present disclosure, the efficiency and accuracy of power data detection are improved. First, in response to receiving the abnormal power data identification request sent by the power edge terminal, the power information to be detected sent by the above-mentioned power edge terminal is obtained. Among them, the above-mentioned power information to be detected includes: a power equipment image set to be detected, a power consumption data set to be detected, the power consumption data to be detected in the power consumption data set to be detected corresponds to the power user, and the power equipment image to be detected in the power equipment image set to be detected corresponds to the power equipment to be detected. Thus, it is convenient to detect each power consumption data to be detected. Secondly, each power equipment image to be detected in the above-mentioned power equipment image set to be detected is preprocessed to generate a preprocessed power equipment image and obtain a preprocessed power equipment image set. Each preprocessed power equipment image in the above-mentioned preprocessed power equipment image set is subjected to equipment abnormality identification to generate an equipment abnormality identification result and obtain an equipment abnormality identification result set. Thus, the appearance of the power equipment can be detected. Then, the above-mentioned power consumption data set to be detected is input into the pre-trained power consumption abnormality identification model to obtain a power consumption abnormality factor information set. Among them, the power data to be detected in the above-mentioned power data set to be detected corresponds to the power abnormality factor information in the above-mentioned power abnormality factor information set. Based on the above-mentioned power abnormality factor information set, each power user identifier to be detected is updated to obtain a target power user identifier group. Among them, one power user identifier to be detected corresponds to one power user. Then, the user power monitoring data corresponding to each target power user identifier in the above-mentioned target power user identifier group is obtained to obtain a user power monitoring data group. The above-mentioned user power monitoring data group is input into the pre-trained power abnormality type detection model to obtain a user power abnormality type group. Among them, one user power monitoring data corresponds to one user power abnormality type. Therefore, the power abnormality type detection model can predict each user power abnormality type of each user power monitoring data corresponding to the power user identifier to be detected with a larger power abnormality factor, thereby reducing the amount of data processing in the model prediction process. Finally, the above-mentioned user power abnormality type group and the corresponding user identifiers are sent to the associated user power abnormality management terminal, and the above-mentioned device abnormality identification result set is sent to the associated power device abnormality management terminal. As a result, the efficiency and accuracy of power data detection are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.

[0014] Figure 1 is a flow chart of some embodiments of the abnormal power data identification method based on the power edge terminal according to the present disclosure; Figure 2 is a structural schematic diagram of some embodiments of an abnormal power data identification device based on a power edge terminal according to the present disclosure; Figure 3 It is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0015] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0016] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure can be combined with each other.

[0017] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0018] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0019] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0020] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0021] Figure 1 1 is a flowchart of some embodiments of the abnormal power data identification method based on the power edge terminal according to the present disclosure. 100 of some embodiments of the abnormal power data identification method based on the power edge terminal according to the present disclosure is shown. The abnormal power data identification method based on the power edge terminal includes the following steps: Step 101, in response to receiving an abnormal power data identification request sent by a power edge terminal, obtaining the power information to be detected sent by the power edge terminal.

[0022] In some embodiments, the execution subject (for example, a computing device) of the abnormal power data identification method based on the power edge terminal can obtain the power information to be detected sent by the power edge terminal in response to receiving the abnormal power data identification request sent by the power edge terminal. Among them, the above-mentioned power information to be detected includes: a power device image set to be detected, a power consumption data set to be detected, the power consumption data to be detected in the power consumption data set to be detected corresponds to the power user, and the power device image to be detected in the power device image set to be detected corresponds to the power device to be detected. The power edge terminal can be a terminal for collecting various power data. For example, the power edge terminal can include: a power device image acquisition terminal and a user power consumption data acquisition terminal. The power device image acquisition terminal can be a camera device. The user power consumption data acquisition terminal can be a power data acquisition terminal (for example, a smart meter) for monitoring the power consumption data of each user in the area. The image of the power device to be detected corresponds to one power device. The power equipment can include but is not limited to: transformers, circuit breakers, distribution boxes and other equipment. The image of the power device to be detected can represent the appearance image of the power device to be detected. One power consumption data to be detected corresponds to one power user. The abnormal power data identification request may represent a request for abnormal identification of the power information to be detected. The power consumption data to be detected may include: user identification, average monthly power consumption, regional power supply information (for example, the total power supply of the community / village where the user is located), regional line loss rate, and abnormal power consumption record information. The average monthly power consumption may represent the average monthly power consumption of the user within a preset time period. The abnormal power consumption record information may indicate whether the user has had abnormal power consumption behavior within a preset time period. Each of the above-mentioned power consumption data to be detected may include an indicator for analyzing abnormal power consumption factors. The above-mentioned abnormal power consumption factor may be an identification indicating abnormal power consumption. For example, the above-mentioned abnormal power consumption identification may be: 0, 1. The above-mentioned 0 may indicate normal power consumption. The above-mentioned 1 may indicate abnormal power consumption.

[0023] Step 102 , preprocessing each of the electric power equipment images to be detected in the above-mentioned electric power equipment image set to generate a preprocessed electric power equipment image, thereby obtaining a preprocessed electric power equipment image set.

[0024] In some embodiments, the execution subject may preprocess each image of the power equipment to be detected in the above-mentioned image set of the power equipment to be detected to generate a preprocessed power equipment image and obtain a preprocessed power equipment image set. For example, each image of the power equipment to be detected in the above-mentioned image set of the power equipment to be detected may be preprocessed by an image denoising algorithm. The image denoising algorithm may be: a filter-based method, a model-based method. The filter-based method may be: a median filter, a mean filter, a Wiener filter. The model-based method may be: a statistical model-based method, a partial differential equation-based method.

[0025] Step 103 , performing device abnormality recognition on each pre-processed power equipment image in the pre-processed power equipment image set to generate a device abnormality recognition result, thereby obtaining a device abnormality recognition result set.

[0026] In some embodiments, the execution subject may perform equipment abnormality recognition on each pre-processed power equipment image in the pre-processed power equipment image set to generate an equipment abnormality recognition result and obtain an equipment abnormality recognition result set. For example, a pre-trained image surface detection model may be used to perform equipment abnormality recognition on each pre-processed power equipment image in the pre-processed power equipment image set to generate an equipment abnormality recognition result and obtain an equipment abnormality recognition result set. For example, the image surface detection model may be an ‌R-CNN series model (R-CNN, SPP-Net, Fast R-CNN, Faster R-CNN), ‌YOLO model, SSD (Single Shot Multibox Detector) model, ‌FPN (Feature PyramidNetwork) model. The equipment abnormality recognition result may indicate whether the appearance of the power equipment in the pre-processed power equipment image is abnormal. For example, the equipment abnormality recognition result may indicate whether the appearance of the power equipment is damaged / rusted / deformed, etc.

[0027] In practice, the execution entity may input the pre-processed power equipment image into a pre-trained power equipment image anomaly recognition model to obtain an equipment anomaly recognition result.

[0028] The power equipment image anomaly recognition model can be trained through the following steps: The first step is to obtain a set of power equipment image samples, wherein one power equipment image sample corresponds to one power equipment type.

[0029] The second step is to cluster the above-mentioned power equipment image sample set to obtain a power equipment image sample group set. For example, each power equipment image sample of the same power equipment type can be clustered into one category.

[0030] The third step is to determine the initial power equipment image anomaly recognition model. The initial power equipment image anomaly recognition model includes: an initial power equipment image anomaly recognition network group, and one initial power equipment image anomaly recognition network corresponds to one power equipment type. The initial power equipment image anomaly recognition network can be an untrained network model for anomaly recognition of power equipment images. For example, the initial power equipment image anomaly recognition network can be an R-CNN series model, a YOLO model, an SSD (Single Shot Multibox Detector) model, or an FPN (Feature Pyramid Network) model.

[0031] In the fourth step, for each initial power equipment image anomaly recognition network in the initial power equipment image anomaly recognition network group, the following training steps are performed: 1. Select a target power equipment image sample from the power equipment image sample group corresponding to the initial power equipment image anomaly recognition network. For example, one power equipment image sample can be selected from the power equipment image sample group as the target power equipment image sample.

[0032] 2. Input the sample power equipment images included in the target power equipment image samples into the initial power equipment image anomaly recognition network to obtain the initial power equipment image anomaly recognition results.

[0033] 3. Determine the recognition difference value between the above-mentioned initial power equipment image abnormality recognition result and the corresponding sample label. For example, the recognition difference value between the above-mentioned initial power equipment image abnormality recognition result and the corresponding sample label can be determined by a preset loss function. The loss function can be a cross entropy loss function or a hinge loss function.

[0034] 4. In response to determining that the above-mentioned recognition difference value is less than or equal to the preset recognition difference value, the initial power equipment image abnormality recognition network is determined as the trained power equipment image abnormality recognition network.

[0035] The fifth step is to merge the trained power equipment image anomaly recognition networks into a power equipment image anomaly recognition model.

[0036] Step 104: input the above-mentioned power consumption data set to be detected into a pre-trained power consumption anomaly recognition model to obtain a power consumption anomaly factor information set.

[0037] In some embodiments, the execution subject may input the power consumption data set to be detected into a pre-trained power consumption anomaly recognition model to obtain a power consumption anomaly factor information set, wherein the power consumption data to be detected in the power consumption data set to be detected corresponds to the power consumption anomaly factor information in the power consumption anomaly factor information set.

[0038] In practice, the execution entity may input the power consumption data set to be detected into the pre-trained power consumption anomaly recognition model through the following steps: In the first step, the above-mentioned power consumption data set to be detected is input into the feature extraction layer included in the above-mentioned power consumption anomaly recognition model to obtain a power consumption data feature information set. Among them, one power consumption data to be detected corresponds to one power consumption data feature information. Among them, the above-mentioned feature extraction layer can be a convolution layer with the power consumption data to be detected as input data and the power consumption data feature information as output data. The above-mentioned power consumption data feature information can be a feature vector that characterizes the power consumption data to be detected.

[0039] In the second step, the power consumption data feature information set is input into the hidden layer included in the power consumption anomaly recognition model to obtain a nonlinear power consumption data feature information set. One power consumption data feature information corresponds to one nonlinear power consumption data feature information. The nonlinear power consumption data feature information can be a feature vector obtained by performing a nonlinear transformation on the power consumption data feature information.

[0040] In the third step, the nonlinear power consumption data feature information set is input into the output layer of the power consumption anomaly recognition model to obtain the power consumption anomaly factor information set. The output layer may be a softmax function layer that takes the nonlinear power consumption data feature information as input and the power consumption anomaly factor information as output. One power consumption anomaly factor information corresponds to one nonlinear power consumption data feature information.

[0041] Step 105: Based on the above power consumption abnormality factor information set, update processing is performed on each power user identification to be detected to obtain a target power user identification group.

[0042] In some embodiments, the execution subject may update each to-be-detected power user identification based on the power consumption abnormality factor information set to obtain a target power user identification group, wherein one to-be-detected power user identification corresponds to one power user.

[0043] In practice, the above execution entity can update the identification of each electricity user to be detected through the following steps: In the first step, each power consumption abnormal factor information satisfying a target condition in the power consumption abnormal factor information set is determined as a target power consumption abnormal factor information group. The target condition may be that the power consumption abnormal factor information is less than a preset value. For example, the preset value may be 1.

[0044] In the second step, the above-mentioned electricity user identities to be detected are determined as a set of electricity user identities to be detected.

[0045] In the third step, each of the to-be-detected power user identifiers corresponding to each of the target power consumption abnormality factor information in the target power consumption abnormality factor information group is determined as each to-be-detected power user identifier to be updated.

[0046] In the fourth step, the above-mentioned to-be-eliminated to-be-detected electricity user identifications are deleted from the above-mentioned to-be-detected electricity user identification set, so as to update the to-be-detected electricity user identification set.

[0047] In the fifth step, the updated set of to-be-detected electricity user identifications is determined as the target electricity user identification group.

[0048] Step 106: Acquire the user power consumption monitoring data corresponding to each target power user identification in the target power user identification group to obtain a user power consumption monitoring data group.

[0049] In some embodiments, the execution subject may obtain the user electricity consumption monitoring data corresponding to each target electricity user identification in the target electricity user identification group to obtain a user electricity consumption monitoring data group.

[0050] Step 107: input the above user electricity consumption monitoring data group into a pre-trained electricity consumption abnormality type detection model to obtain a user electricity consumption abnormality type group.

[0051] In some embodiments, the execution subject may input the user power consumption monitoring data group into a pre-trained power consumption abnormality type detection model to obtain a user power consumption abnormality type group. One user power consumption monitoring data corresponds to one user power consumption abnormality type. The user power consumption abnormality type may represent the type of abnormal power consumption behavior.

[0052] The power consumption anomaly type detection model can be trained through the following steps: The first step is to obtain a historical electricity consumption monitoring data set corresponding to each electricity user identification to be detected. Among them, the historical electricity consumption monitoring data includes: a historical electricity consumption monitoring sub-data group. The historical electricity consumption monitoring data can be data on user power consumption collected by the electricity monitoring device within a preset time period. The electricity monitoring device includes: a smart meter and a power load collector. The above-mentioned historical electricity consumption monitoring data can be data used to determine the type identification of abnormal electricity consumption. The historical electricity consumption monitoring sub-data can be: an electricity consumption data sequence, an electricity consumption correlation value sequence. The above-mentioned electricity consumption monitoring sub-data has corresponding task detection information. The above-mentioned task detection information can represent a pre-packaged task processing method. The above-mentioned electricity consumption data sequence can correspond to each total electricity consumption within a preset time. The above-mentioned electricity consumption correlation value sequence can correspond to the above-mentioned preset time. The above-mentioned electricity consumption correlation value sequence can be a sequence of electricity consumption correlation values ​​arranged in time (date) order within the above-mentioned preset time. For example, the electricity consumption correlation value sequence can be {0.6, 0.8, 0.2, 0.1}. The electricity consumption correlation value can be a Pearson correlation coefficient between the characterization of electricity consumption.

[0053] The second step is to identify the abnormal power usage type of each historical power usage monitoring data in the historical power usage monitoring data set to obtain a historical power usage abnormal type identification set. For example, the historical power usage abnormal type identification can be a name indicating the type of abnormal power usage behavior. The power usage abnormal type identification can be: power usage abnormality due to meter tampering or fraud, and power usage abnormality due to coded code.

[0054] The second step may include the following sub-steps: The first sub-step is to perform the following processing steps on each historical electricity consumption monitoring sub-data in each historical electricity consumption monitoring data: 1. Obtain task detection information corresponding to the historical power consumption monitoring sub-data. For example, the task detection information may be a programming language code for performing power consumption anomaly detection on the historical power consumption monitoring sub-data.

[0055] 2. Execute the detection task corresponding to the above-mentioned task detection information to perform power consumption anomaly detection on the above-mentioned historical power consumption monitoring sub-data to generate historical power consumption anomaly type information. As an example, when the above-mentioned task detection information is a numerical sequence related to power consumption, the above-mentioned execution subject may execute the detection task corresponding to the above-mentioned numerical sequence related to power consumption to perform power consumption anomaly detection on the above-mentioned historical power consumption monitoring sub-data to obtain the historical power consumption anomaly type information corresponding to the above-mentioned historical power consumption monitoring sub-data. The above-mentioned historical power consumption anomaly type information may be a characterization of "power consumption anomaly due to meter tampering or fraudulent power consumption" or a characterization of "power consumption anomaly due to the absence of meter tampering or fraudulent power consumption".

[0056] In the second sub-step, the historical power usage anomaly type information that meets the preset conditions in each historical power usage anomaly type information is determined as the historical power usage anomaly type identifier to obtain a historical power usage anomaly type identifier set. The preset condition may be that the historical power usage anomaly type information represents meter tampering or fraudulent power usage anomaly.

[0057] The third step is to train the initial power consumption anomaly type detection model based on the above historical power consumption monitoring data set and the above historical power consumption anomaly type identification set to obtain a trained power consumption anomaly type detection model.

[0058] The third step may include the following sub-steps: The first sub-step is to perform the following processing steps for each historical electricity consumption monitoring data in the above historical electricity consumption monitoring data set: 1. The historical power consumption abnormality type identifications corresponding to the historical power consumption monitoring data are concentrated and determined as reference historical power consumption abnormality type identifications.

[0059] 2. Combine the above historical electricity consumption monitoring data and the above reference historical electricity consumption anomaly type identification into electricity consumption monitoring data training samples.

[0060] The second sub-step is to select a target power consumption monitoring data training sample from each power consumption monitoring data training sample. For example, one power consumption monitoring data training sample can be randomly selected from each power consumption monitoring data training sample as the target power consumption monitoring data training sample.

[0061] In a third sub-step, the historical electricity consumption monitoring data included in the target electricity consumption monitoring data training sample is input into the initial electricity consumption abnormality type detection model to obtain the initial user electricity consumption abnormality type detection result.

[0062] The fourth sub-step is to determine the loss value between the above-mentioned initial user power consumption abnormality type detection result and the corresponding reference historical power consumption abnormality type identification. For example, the loss value between the above-mentioned initial user power consumption abnormality type detection result and the corresponding reference historical power consumption abnormality type identification can be determined by a preset damage function.

[0063] The fifth sub-step, in response to determining that the above-mentioned loss value is less than or equal to the preset loss value, determines the initial power consumption abnormality type detection model as the trained power consumption abnormality type detection model.

[0064] Therefore, the power consumption anomaly type detection model can be used to predict the power consumption anomaly type of each user in each power consumption monitoring data corresponding to the power user identification to be detected with a larger power consumption anomaly factor, thereby reducing the amount of data processing in the model prediction process and reducing the waste of computer computing resources.

[0065] Step 108: sending the user power usage anomaly type group and the corresponding user identifiers to the associated user power usage anomaly management terminal, and sending the device anomaly identification result set to the associated power device anomaly management terminal.

[0066] In some embodiments, the execution subject may send the user power usage anomaly type group and the corresponding user identifiers to the associated user power usage anomaly management terminal, and send the device anomaly identification result set to the associated power equipment anomaly management terminal. The user power usage anomaly management terminal may be a terminal for managing user power usage anomalies. The power equipment anomaly management terminal may be a terminal for managing power equipment surface anomalies.

[0067] Further references Figure 2 As an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of an abnormal power data identification device based on a power edge terminal. These embodiments of the abnormal power data identification device based on a power edge terminal are similar to Figure 1 Corresponding to the method embodiments shown, the abnormal power data identification device based on the power edge terminal can be specifically applied to various electronic devices.

[0068] like Figure 2As shown, some embodiments of the abnormal power data identification device 200 based on the power edge terminal include: a first acquisition unit 201, a preprocessing unit 202, an identification unit 203, a first input unit 204, an update unit 205, a second acquisition unit 206, a second input unit 207 and a sending unit 208. Among them, the first acquisition unit 201 is configured to obtain the power information to be detected sent by the above-mentioned power edge terminal in response to receiving the abnormal power data identification request sent by the above-mentioned power edge terminal, wherein the above-mentioned power information to be detected includes: a set of power equipment images to be detected, a power consumption data set to be detected, the power consumption data to be detected in the power consumption data set to be detected corresponds to the power user, and the power equipment images to be detected in the power equipment image set to be detected correspond to the power equipment to be detected; the preprocessing unit 202 is configured to preprocess each power equipment image to be detected in the above-mentioned power equipment image set to generate a preprocessed power equipment image and obtain a preprocessed power equipment image set; the identification unit 203 is configured to perform device abnormality identification on each preprocessed power equipment image in the above-mentioned preprocessed power equipment image set to generate a device abnormality identification result and obtain a device abnormality identification result set; the first input unit 204 is configured to input the above-mentioned power consumption data set to be detected into a pre-trained power consumption abnormality identification model to obtain a power consumption abnormality factor information set, wherein the above The power data to be detected in the power data set to be detected corresponds to the power abnormality factor information in the power abnormality factor information set; the updating unit 205 is configured to update each power user identifier to be detected based on the power abnormality factor information set to obtain a target power user identifier group, wherein one power user identifier to be detected corresponds to one power user; the second acquisition unit 206 is configured to obtain the user power monitoring data corresponding to each target power user identifier in the target power user identifier group to obtain a user power monitoring data group; the second input unit 207 is configured to input the user power monitoring data group into a pre-trained power abnormality type detection model to obtain a user power abnormality type group, wherein one user power monitoring data corresponds to one user power abnormality type; the sending unit 208 is configured to send the user power abnormality type group and the corresponding user identifiers to the associated user power abnormality management terminal, and send the device abnormality identification result set to the associated power equipment abnormality management terminal.

[0069] It can be understood that the units recorded in the abnormal power data identification device 200 based on the power edge terminal are similar to the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the method are also applicable to the abnormal power data identification device 200 based on the power edge terminal and the units contained therein, and will not be repeated here.

[0070] Reference below Figure 3 , which shows a schematic diagram of the structure of an electronic device 300 (e.g., a computing device) suitable for implementing some embodiments of the present disclosure. The electronic devices in some embodiments of the present disclosure may include, but are not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0071] like Figure 3 As shown, the electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 to a random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0072] Typically, the following devices may be connected to the I / O interface 305: an input device 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 308 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 309. The communication device 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0073] In particular, according to some embodiments of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from the network through the communication device 309, or installed from the storage device 308, or installed from the ROM 302. When the computer program is executed by the processing device 301, the above-mentioned functions defined in the method of some embodiments of the present disclosure are executed.

[0074] It should be noted that the computer-readable medium recorded in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In some embodiments of the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device. In some embodiments of the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer readable signal media may also be any computer readable medium other than computer readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0075] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0076] The above-mentioned computer-readable medium may be included in the above-mentioned electronic device; or it may exist independently without being assembled into the electronic device. The above-mentioned computer-readable medium carries one or more programs. When the above-mentioned one or more programs are executed by the electronic device, the electronic device: in response to receiving an abnormal power data identification request sent by the power edge terminal, obtains the power information to be detected sent by the above-mentioned power edge terminal, wherein the above-mentioned power information to be detected includes: a set of power equipment images to be detected, a set of power consumption data to be detected, the power consumption data to be detected in the power consumption data set to be detected corresponds to the power user, and the power equipment images to be detected in the set of power equipment images to be detected correspond to the power equipment to be detected; preprocessing each power equipment image to be detected in the above-mentioned set of power equipment images to be detected to generate a preprocessed power equipment image and obtain a set of preprocessed power equipment images; performing device abnormality identification on each preprocessed power equipment image in the above-mentioned set of preprocessed power equipment images to generate a device abnormality identification result and obtain a set of device abnormality identification results; inputting the above-mentioned power consumption data set to be detected into the pre-trained power consumption abnormality identification system; In the common recognition model, an abnormal power consumption factor information set is obtained, wherein the power consumption data to be detected in the above-mentioned power consumption data set to be detected corresponds to the abnormal power consumption factor information in the above-mentioned abnormal power consumption factor information set; based on the above-mentioned abnormal power consumption factor information set, each power user identification to be detected is updated to obtain a target power user identification group, wherein one power user identification to be detected corresponds to one power user; the user power consumption monitoring data corresponding to each target power user identification in the above-mentioned target power user identification group is obtained to obtain a user power consumption monitoring data group; the above-mentioned user power consumption monitoring data group is input into a pre-trained power consumption abnormality type detection model to obtain a user power consumption abnormality type group, wherein one user power consumption monitoring data corresponds to one user power consumption abnormality type; the above-mentioned user power consumption abnormality type group and the corresponding user identifications are sent to the associated user power consumption abnormality management terminal, and the above-mentioned equipment abnormality recognition result set is sent to the associated power equipment abnormality management terminal.

[0077] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0078] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0079] The units described in some embodiments of the present disclosure may be implemented by software or by hardware. The described units may also be set in a processor, for example, may be described as: a processor comprising: a first acquisition unit, a preprocessing unit, an identification unit, a first input unit, an update unit, a second acquisition unit, a second input unit and a sending unit. Among them, the names of these units do not constitute a limitation on the unit itself under certain circumstances. For example, the sending unit may also be described as "a unit that sends the above-mentioned user power consumption abnormality type group and the corresponding user identifiers to the associated user power consumption abnormality management terminal, and sends the above-mentioned device abnormality identification result set to the associated power equipment abnormality management terminal".

[0080] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0081] The above descriptions are only some preferred embodiments of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the embodiments of the present disclosure.

Claims

1. A method for identifying abnormal power data based on a power edge terminal, comprising: In response to receiving an abnormal power data identification request sent by a power edge terminal, obtaining power information to be detected sent by the power edge terminal, wherein the power information to be detected includes: a power device image set to be detected and a power consumption data set to be detected, the power consumption data to be detected in the power consumption data set to be detected corresponds to the power user, and the power device image to be detected in the power device image set to be detected corresponds to the power device to be detected; Preprocessing each of the to-be-detected electric power equipment images in the to-be-detected electric power equipment image set to generate a preprocessed electric power equipment image, thereby obtaining a preprocessed electric power equipment image set; Performing device abnormality recognition on each pre-processed power equipment image in the pre-processed power equipment image set to generate a device abnormality recognition result, thereby obtaining a device abnormality recognition result set; Inputting the to-be-detected power consumption data set into a pre-trained power consumption anomaly recognition model to obtain a power consumption anomaly factor information set, wherein the to-be-detected power consumption data in the to-be-detected power consumption data set corresponds to the power consumption anomaly factor information in the power consumption anomaly factor information set; Based on the power consumption abnormality factor information set, each power user identification to be detected is updated to obtain a target power user identification group, wherein one power user identification to be detected corresponds to one power user; Acquire user electricity consumption monitoring data corresponding to each target electricity user identification in the target electricity user identification group to obtain a user electricity consumption monitoring data group; Inputting the user power consumption monitoring data group into a pre-trained power consumption abnormality type detection model to obtain a user power consumption abnormality type group, wherein one user power consumption monitoring data corresponds to one user power consumption abnormality type; The user power usage anomaly type group and the corresponding user identifiers are sent to the associated user power usage anomaly management terminal, and the device anomaly identification result set is sent to the associated power device anomaly management terminal.

2. The method according to claim 1, wherein: Before inputting the user power consumption monitoring data group into a pre-trained power consumption abnormality type detection model to obtain the user power consumption abnormality type group, the method further includes: Acquire a historical electricity consumption monitoring data set corresponding to each electricity user identifier to be detected, wherein the historical electricity consumption monitoring data includes: a historical electricity consumption monitoring sub-data set; Performing power consumption anomaly type identification processing on each historical power consumption monitoring data in the historical power consumption monitoring data set to obtain a historical power consumption anomaly type identification set; The initial power consumption anomaly type detection model is trained according to the historical power consumption monitoring data set and the historical power consumption anomaly type identification set to obtain a trained power consumption anomaly type detection model.

3. The method according to claim 2, wherein: The performing power consumption abnormality type identification processing on each historical power consumption monitoring data in the historical power consumption monitoring data set to obtain a historical power consumption abnormality type identification set includes: For each historical electricity consumption monitoring sub-data in each historical electricity consumption monitoring data, the following processing steps are performed: Obtaining task detection information corresponding to the historical electricity consumption monitoring sub-data; Executing the detection task corresponding to the task detection information to perform power consumption anomaly detection on the historical power consumption monitoring sub-data to generate historical power consumption anomaly type information; The historical power consumption abnormality type information that meets the preset conditions in each historical power consumption abnormality type information is determined as the historical power consumption abnormality type identifier, and a historical power consumption abnormality type identifier set is obtained.

4. The method according to claim 1, wherein: The step of inputting the power consumption data set to be detected into a pre-trained power consumption anomaly recognition model to obtain a power consumption anomaly factor information set includes: Inputting the power consumption data set to be detected into the feature extraction layer included in the power consumption anomaly recognition model to obtain a power consumption data feature information set, wherein one power consumption data to be detected corresponds to one power consumption data feature information; Inputting the power consumption data feature information set into the hidden layer included in the power consumption anomaly recognition model to obtain a nonlinear power consumption data feature information set, wherein one power consumption data feature information corresponds to one nonlinear power consumption data feature information; The nonlinear power consumption data feature information set is input into the output layer of the power consumption anomaly recognition model to obtain a power consumption anomaly factor information set.

5. The method according to claim 1, wherein: The performing device abnormality recognition on each pre-processed power equipment image in the pre-processed power equipment image set to generate a device abnormality recognition result includes: The pre-processed electric power equipment image is input into a pre-trained electric power equipment image anomaly recognition model to obtain an equipment anomaly recognition result.

6. An abnormal power data identification device based on a power edge terminal, comprising: A first acquisition unit is configured to, in response to receiving an abnormal power data identification request sent by a power edge terminal, acquire power information to be detected sent by the power edge terminal, wherein the power information to be detected includes: a power device image set to be detected and a power consumption data set to be detected, the power consumption data to be detected in the power consumption data set to be detected corresponds to the power user, and the power device image to be detected in the power device image set to be detected corresponds to the power device to be detected; A preprocessing unit is configured to preprocess each of the to-be-detected power equipment images in the to-be-detected power equipment image set to generate a preprocessed power equipment image, thereby obtaining a preprocessed power equipment image set; an identification unit configured to perform device abnormality identification on each pre-processed power equipment image in the pre-processed power equipment image set to generate a device abnormality identification result and obtain a device abnormality identification result set; A first input unit is configured to input the power consumption data set to be detected into a pre-trained power consumption anomaly recognition model to obtain a power consumption anomaly factor information set, wherein the power consumption data to be detected in the power consumption data set to be detected corresponds to the power consumption anomaly factor information in the power consumption anomaly factor information set; An updating unit is configured to update each to-be-detected power user identification based on the power consumption abnormality factor information set to obtain a target power user identification group, wherein one to-be-detected power user identification corresponds to one power user; A second acquisition unit is configured to acquire user electricity consumption monitoring data corresponding to each target electricity user identification in the target electricity user identification group to obtain a user electricity consumption monitoring data group; A second input unit is configured to input the user power consumption monitoring data group into a pre-trained power consumption abnormality type detection model to obtain a user power consumption abnormality type group, wherein one user power consumption monitoring data corresponds to one user power consumption abnormality type; The sending unit is configured to send the user power consumption abnormality type group and the corresponding user identifications to the associated user power consumption abnormality management terminal, and send the device abnormality identification result set to the associated power device abnormality management terminal.

7. An electronic device comprising: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 5.

8. A computer readable medium having a computer program stored thereon, wherein: When the program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images