Electric power big data intelligent detection system, method, device and equipment

Abstract

The invention belongs to the technical field of power internet of things, and particularly relates to a power big data intelligent detection system, method, device and equipment. The method comprises the following steps: acquiring a power utilization scene video stream through a camera, and acquiring user power utilization information through power data acquisition equipment; face recognition is carried out on the video stream, if the person in the video stream is recognized as an electric power employee, whether dressing of the electric power employee accords with a dressing specification is further recognized, and a recognition result is reported to the cloud server; and judging whether the electricity utilization of the user is abnormal or not based on the electricity utilization information of the user, if so, reporting alarm information and the electricity utilization information of the user to the cloud server, and if not, reporting the electricity utilization information of the user to the cloud server. According to the electric power big data intelligent detection scheme provided by the invention, the system complexity is simplified, and meanwhile, intelligent real-time monitoring of the video monitoring data and the power utilization data can be realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric power Internet of Things, and specifically relates to an electric power big data intelligent detection system, method, device and equipment. Background Art

[0002] The existing power substation monitoring and management platform performs intelligent detection of power big data based on the traditional centralized cloud computing model, that is, big data files (such as video files) are transmitted to the cloud for centralized processing through high-speed power line communication (HPLC). However, due to the bandwidth limitation of HPLC technology, big data files often need to be transmitted to the cloud through wireless or optical fiber communication, which will require the establishment of a wireless / optical fiber communication system based on the HPLC system, which is time-consuming, labor-intensive, complex, and costly; in addition, the transmission, storage, and analysis of massive data will consume a lot of resources in the cloud computing center, so the centralized processing mode in the cloud is difficult to meet the demand for real-time processing of big data files. Summary of the invention

[0003] The embodiment of the present invention proposes an intelligent detection solution for power big data to solve the problem that the existing power substation monitoring and management platform based on the centralized cloud computing model needs to set up a wireless / optical fiber communication system for transmitting big data in addition to the HPLC transmission system, resulting in high system complexity and high cost, and it is difficult to meet the real-time processing requirements of big data files.

[0004] In a first aspect, a power big data intelligent detection system includes a cloud server and an edge server, wherein the edge server is connected to the cloud server network, wherein:

[0005] The edge server is connected to a camera network, the camera is used to capture a video stream of a power usage scene, the edge server is used to perform face recognition on the video stream using a deep learning model, is used to identify the power worker in the video stream, is used to further identify whether the power worker's attire complies with the dress code, and is used to report the recognition result to the cloud server;

[0006] The edge server is also connected to a power data collection device, which is used to collect user power usage information; the edge server is used to determine whether the user's power usage is abnormal based on the user's power usage information through a machine learning model, and to report alarm information and user power usage information to the cloud server when the power usage is abnormal, and to report user power usage information to the cloud server when there is no abnormality in the power usage.

[0007] Furthermore, it also includes a concentrator, a collector and a terminal device; the concentrator is connected to the cloud server network, a plurality of collectors are connected to the concentrator via a high-speed power line carrier communication network, the edge server is connected one-to-one with the collector via RS-485, and the terminal device includes the camera and the power data acquisition device.

[0008] In a second aspect, a method for intelligent detection of electric power big data is implemented based on the electric power big data intelligent detection system, comprising:

[0009] The video stream of the power consumption scene is obtained through the camera, and the user's power consumption information is obtained through the power data collection equipment;

[0010] Performing face recognition on the video stream, if the person in the video stream is identified as an electric power worker, further identifying whether the clothing of the electric power worker complies with the dress code, and reporting the recognition result to the cloud server;

[0011] Based on the user's electricity usage information, determine whether the user's electricity usage is abnormal. If there is an abnormality, report the alarm information and the user's electricity usage information to the cloud server. If there is no abnormality, report the user's electricity usage information to the cloud server.

[0012] Furthermore, the method for performing face recognition on the video stream includes:

[0013] Acquire a frame image from a video stream, and preprocess the frame image, wherein the preprocessing includes one or more of denoising, cropping, scaling, and grayscale conversion;

[0014] Performing face detection on the frame image using a deep learning model to determine whether a face exists in the frame image;

[0015] If there is a face, a convolutional neural network is used to extract facial features, and the facial features are matched with facial features in a preset face library, and corresponding operations are performed on the matching results according to preset rules.

[0016] Further, the method for identifying whether the attire of the power employee complies with the dress code includes:

[0017] Capture the image of the power worker in his or her uniform;

[0018] The power worker's attire image is input into the power worker's attire standard recognition model to identify whether the power worker's attire is standard.

[0019] Furthermore, the training method of the power worker dress code recognition model includes:

[0020] Obtaining standard image materials of electric power workers in uniform and marking them as standard, and non-standard image materials of electric power workers in uniform and marking them as non-standard, and pre-processing the image materials; wherein the pre-processing includes one or more of denoising, cropping, and scaling;

[0021] The preprocessed image material is used as a sample and the annotation is used as a label to train a convolutional neural network model to obtain a recognition model for the dress code of power workers.

[0022] Furthermore, the method for judging whether the user's electricity usage is abnormal based on the user's electricity usage information includes:

[0023] Obtain user electricity consumption information collected by power data collection equipment;

[0024] Inputting the user's electricity usage information into an anti-electricity theft model to determine whether the user's electricity usage is abnormal;

[0025] If there is any abnormality, an alarm message corresponding to the abnormality is triggered.

[0026] Furthermore, the training method of the anti-electricity theft model includes:

[0027] Collecting electricity consumption data of a preset number of users and cleaning and preprocessing the data; wherein the electricity consumption data includes the user's electricity consumption, electricity consumption time, voltage, and current, and the cleaning and preprocessing include removing outliers, processing missing data, and performing time series analysis;

[0028] Selecting features related to the user's electricity usage behavior from the electricity usage data to form a sample set, and marking the data in the sample set as a normal electricity usage mode or an abnormal electricity usage mode according to a specific task, wherein the features include electricity usage mode, peak and valley distribution of electricity consumption, daily / weekly / monthly electricity usage mode, and energy-saving behavior;

[0029] The data in the sample set is used as samples, and the annotations corresponding to the data are used as labels to train the machine learning model to obtain an anti-electricity theft model.

[0030] In a third aspect, a power big data intelligent detection device includes:

[0031] An acquisition module is used to acquire a video stream of a power consumption scene through a camera, and to acquire user power consumption information through a power data acquisition device;

[0032] a recognition module, for performing face recognition on the video stream, for identifying the power workers in the video stream, for further identifying whether the clothing of the power workers complies with the dress code, and for reporting the recognition result to the cloud server;

[0033] The reporting module is used to determine whether the user's electricity usage is abnormal based on the user's electricity usage information, and to report the alarm information and the user's electricity usage information to the cloud server when the electricity usage is abnormal, and to report the user's electricity usage information to the cloud server when there is no abnormality in the electricity usage.

[0034] In a fourth aspect, a computer device includes a memory and a processor:

[0035] The memory is used to store executable instructions;

[0036] The processor is used to execute the executable instructions to implement the detection method according to any one of claims 3-8.

[0037] To summarize, the power big data intelligent detection system, method, and device provided by the present invention, by setting an edge server on the terminal side of the power substation, the edge server reads the local video data and power consumption data in real time, and processes the big data files on the terminal side while ensuring the integrity, security, and stability of the data, thereby improving the overall processing speed of the power big data, reducing the transmission delay, and realizing intelligent real-time monitoring of the substation video surveillance data and power consumption data. At the same time, because there is no need to transmit normal video surveillance data to the cloud, there is no need to set up a wireless / optical fiber communication system for transmitting big data, which simplifies the complexity of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The features and advantages of the present invention will be more clearly understood by referring to the accompanying drawings, which are schematic and should not be construed as limiting the present invention in any way. In the accompanying drawings:

[0039] Figure 1 It is a schematic diagram of an electric power big data intelligent detection system applicable to the present invention;

[0040] Figure 2 is a flow chart of a method for intelligent detection of power big data according to some embodiments of the present invention;

[0041] Figure 3 is a flow chart of an algorithm for performing face recognition and dress code recognition on a video stream shown in one embodiment of the present invention;

[0042] Figure 4 is a flow chart of a user anti-electricity theft algorithm shown in one embodiment of the present invention;

[0043] Figure 5 is a flow chart of an algorithm for processing local video and power consumption data on a terminal side shown in one embodiment of the present invention;

[0044] Figure 6 is a schematic diagram of an intelligent detection device for power big data shown in one embodiment of the present invention;

[0045] Figure 7 is a schematic diagram of a computer device according to some embodiments of the present invention. DETAILED DESCRIPTION

[0046] In the detailed description below, many specific details of the present invention are set forth by way of example in order to provide a thorough understanding of the relevant disclosure. However, it will be apparent to one of ordinary skill in the art that the present invention can be implemented without these details. It should be understood that the use of the terms "system," "device," "unit," and / or "module" in the present invention is a method for distinguishing different parts, elements, parts, or assemblies at different levels in a sequential arrangement. However, these terms may be replaced by other expressions if other expressions can achieve the same purpose.

[0047] It should be understood that when a device, unit or module is referred to as being "on," "connected to," or "coupled to" another device, unit or module, it may be directly on, connected to, coupled to, or communicating with other devices, units or modules, or there may be intermediate devices, units or modules, unless the context clearly indicates an exception. For example, the term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0048] The terms used in the present invention are only for describing specific embodiments and are not intended to limit the scope of the present invention. As shown in the present specification and claims, unless the context clearly indicates an exception, the words "a", "an", "a kind" and / or "the" do not specifically refer to the singular, but may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of clearly identified features, wholes, steps, operations, elements and / or components, and such expressions do not constitute an exclusive list, and other features, wholes, steps, operations, elements and / or components may also be included.

[0049] These and other features and characteristics of the present invention, methods of operation, functions of related elements of structure, combinations of parts and economies of manufacture may be better understood with reference to the following description and accompanying drawings, which form a part of the specification. However, it is to be clearly understood that the drawings are for illustration and description purposes only and are not intended to limit the scope of protection of the present invention. It is to be understood that the drawings are not drawn to scale.

[0050] The present invention uses a variety of structural diagrams to illustrate various variations of the embodiments of the present invention. It should be understood that the above or below structures are not intended to limit the present invention. The scope of protection of the present invention shall be subject to the claims.

[0051] Figure 1 Schematic diagram of an intelligent detection system for power big data applicable to the present invention. Figure 1 The system shown includes a concentrator connected to a cloud server network, multiple collectors connected to the concentrator via a high-speed power line carrier communication network, an edge server connected one-to-one with the collector via RS-485, and a terminal device connected to the edge server, wherein the terminal device includes a camera and a power data acquisition device.

[0052] The edge server is connected to a camera network, the camera is used to capture a video stream of a power usage scene, and the edge server uses a deep learning model to perform face recognition on the video stream. If the person in the video stream is identified as an electric power worker, the edge server further identifies whether the dress of the electric power worker complies with the dress code, and reports the identification result to the cloud server;

[0053] The edge server is also connected to a power data collection device, which is used to collect user power usage information. The edge server determines whether there is any abnormality in the user's power usage based on the user's power usage information through a machine learning model. If there is an abnormality, the alarm information and the user's power usage information are reported to the cloud server. If there is no abnormality, the user's power usage information is reported to the cloud server.

[0054] Preferably, the method for the edge server to report information to the cloud server includes:

[0055] The edge server reports information to the collector via RS-485, the collector reports information to the concentrator via high-speed power line carrier communication, and the concentrator reports the information to the cloud server via the network.

[0056] Figure 2 is a flow chart of a method for intelligent detection of power big data according to some embodiments of the present invention. In some embodiments, the method for intelligent detection of power big data is based on Figure 1 The detection system shown is implemented, and the power big data intelligent detection method includes the following steps:

[0057] S210, obtaining a video stream of a power consumption scene through a camera, and obtaining user power consumption information through a power data collection device.

[0058] Specifically, a camera is set in the electricity usage scene, the camera captures video information, and the edge server receives the video stream shot by the camera through RTSP.

[0059] At the same time, through power data collection equipment, such as electric meters, the user's electricity consumption information is obtained, such as the user's electricity consumption, electricity consumption time, voltage, current, etc. The electric meter can be connected to the edge computing device through RS-485, or it can be connected to the network, or the edge computing device is embedded in the electric meter.

[0060] S220, performing face recognition on the video stream. If the person in the video stream is identified as an electric power worker, further identifying whether the clothing of the electric power worker complies with the dress code, and reporting the identification result to the cloud server.

[0061] Specifically, a face recognition library and a dress code recognition model are established, image information in the video stream is intercepted, and face recognition is performed on the image information. When the person in the image information is identified as an electric power worker, the dress code recognition of the electric power worker is performed, such as Figure 3 shown.

[0062] The edge server transmits the identification results to the collector via RS-485, the collector transmits the data to the concentrator via HPLC, and the concentrator uploads the data to the cloud server.

[0063] Among them, the specific steps of the face recognition algorithm are as follows:

[0064] Step 1: Install a camera with sufficient resolution and quality to support face recognition on the edge device to collect face images or video streams.

[0065] Step 2: Preprocess the acquired images or video streams, including denoising, cropping, scaling, grayscale, and other operations to prepare the data for subsequent processing.

[0066] Step 3: Use a deep learning model to perform face detection on the edge device. This can help determine whether there is a face in the image.

[0067] Step 4: Extract facial features. Use the convolutional neural network (CNN) in the deep learning model to extract facial features.

[0068] Step 5: Match the extracted facial features with known facial features to identify the face. Compare the face with the faces in the known face database.

[0069] Step 6: Take appropriate actions based on the recognition results, such as generating an alarm, recording the recognition results, triggering other devices, etc.

[0070] Step 7: Regularly monitor system performance and update the model based on demand and new data to maintain the accuracy and efficiency of face recognition.

[0071] The specific steps of the dress code recognition algorithm are as follows:

[0072] Step 1: Obtain image materials of power workers in standard attire and image materials of power workers in irregular attire;

[0073] Step 2: Preprocess the acquired image material, including denoising, image enhancement, image cropping and scaling, to prepare data for model processing.

[0074] Step 3: Select an appropriate deep learning model for the dress code recognition task. You can use a convolutional neural network (CNN) or other visual models, and choose the model based on the complexity of the problem and resource constraints. Train the selected model using a labeled dress image dataset. Make sure the model can learn the characteristics of both dress codes and non-codes.

[0075] Step 4: Capture the video stream through the camera and intercept the image of the power worker's clothing;

[0076] Step 5: Input the power worker's clothing image into the clothing recognition model in step 3. The model will classify the image and determine whether it meets the regulations. Take appropriate actions based on the recognition results, such as generating an alarm, recording dress code violations, and notifying relevant personnel.

[0077] Step 6: Regularly monitor system performance and update the model as needed to ensure accuracy and efficiency of dress code recognition.

[0078] S230, judging whether there is any abnormality in the user's electricity usage based on the user's electricity usage information, if there is any abnormality, reporting the alarm information and the user's electricity usage information to the cloud server, if there is no abnormality, reporting the user's electricity usage information to the cloud server.

[0079] Specifically, the user's electricity usage information, such as the user's electricity usage, electricity usage time, voltage, current and other information, is obtained through power data collection equipment, such as an electricity meter. The above user's electricity usage information is input into the anti-electricity theft model to determine whether the user's electricity usage is abnormal. If there is an abnormality, the corresponding alarm query information is generated, and the alarm information and the user's electricity usage information are reported to the cloud server. If there is no abnormality, the user's electricity usage information is reported to the cloud server.

[0080] Electricity usage (anti-electricity theft) monitoring algorithm such as Figure 4 As shown, the specific steps are as follows:

[0081] Step 1: Use past electricity usage datasets to train the machine learning model so that it can accurately detect power theft or anomalies;

[0082] Step 2: The terminal device sends the collected power consumption information to the edge server;

[0083] Step 3: The edge server uses a machine learning model to determine whether the user's electricity usage is abnormal;

[0084] Step 4: If there is an abnormality, the edge server sends the alarm information and power consumption information to the collector via RS-485; if there is no abnormality, the edge server sends the power consumption information to the collector via RS-485;

[0085] Step 5: The collector transmits the data to the concentrator via HPLC, and the concentrator uploads the data to the cloud;

[0086] In step 1, the training steps of the anti-electricity theft model based on machine learning are as follows:

[0087] Step 1: First, it is necessary to collect a large amount of electricity consumption data from users. The data should include information such as the amount of electricity used, the time of use, voltage, current, etc.

[0088] Step 2: Clean and preprocess the collected data, including removing outliers, processing missing data, and performing time series analysis to ensure data quality and consistency;

[0089] Step 3: Select features related to the user's electricity consumption behavior, such as electricity consumption pattern, peak and valley distribution of electricity consumption, daily / weekly / monthly electricity consumption pattern, energy-saving behavior, etc.

[0090] Step 4: Label the data according to the specific task to identify normal and abnormal power usage patterns;

[0091] Step 5: Select an appropriate machine learning model. For anomaly detection problems, an outlier detection algorithm can be used. Use a labeled dataset to train the selected model so that the model can learn the patterns of user electricity usage characteristics;

[0092] Step 6: Use cross-validation or reserve a portion of the data as a test set to evaluate the performance of the model. According to the performance in the actual application, optimize and adjust the model to improve the prediction accuracy.

[0093] The present invention sets an edge server on the terminal side of the power substation, and the edge server reads the local video data and power consumption data in real time, and processes the big data files on the terminal side, thereby improving the overall processing speed of power big data and reducing transmission delay. The schematic diagram of the whole process is shown in Figure 5.

[0094] Figure 6 Schematic diagram of an intelligent detection device for power big data according to some embodiments of the present invention. The intelligent detection device for power big data 600 is used to implement Figure 2 The detection method described in S210-S230, as Figure 6As shown, it includes an acquisition module 610, an identification module 620 and a reporting module 630.

[0095] in:

[0096] An acquisition module 610 is used to acquire a video stream of a power consumption scene through a camera and acquire user power consumption information through a power data acquisition device;

[0097] The recognition module 620 is used to perform face recognition on the video stream. If the person in the video stream is identified as an electric power worker, it further identifies whether the clothing of the electric power worker complies with the dress code, and reports the recognition result to the cloud server;

[0098] The reporting module 630 is used to determine whether the user's electricity usage is abnormal based on the user's electricity usage information. If there is an abnormality, the alarm information and the user's electricity usage information are reported to the cloud server. If there is no abnormality, the user's electricity usage information is reported to the cloud server.

[0099] One embodiment of the present invention provides a computer device. Figure 7 As shown, the computer device 700 includes a memory 720 and a processor 710, wherein the memory 720 is used to store executable instructions; and the processor 710 is used to execute the executable instructions to implement Figure 2 The detection method described in S210-S230.

[0100] An embodiment of the present invention provides a computer-readable storage medium. The computer-readable storage medium stores instructions, and when the instructions are executed, Figure 2 The detection method described in S210-S230.

[0101] To summarize, the electric power big data intelligent detection system, method, device and equipment provided by the embodiments of the present invention, by setting an edge server on the terminal side of the power substation, the edge server reads the local video data and electricity consumption data in real time, and processes the big data file on the terminal side while ensuring the integrity, security and stability of the data, thereby improving the overall processing speed of the electric power big data, reducing the transmission delay, and realizing the intelligent real-time monitoring of the substation video surveillance data and electricity consumption data. At the same time, because there is no need to transmit normal video surveillance data to the cloud, it also avoids the need to set up a wireless / optical fiber communication system for transmitting big data, thereby simplifying the complexity of the system.

[0102] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and modules can refer to the corresponding descriptions in the aforementioned device embodiments and will not be repeated here.

[0103] Although the subject matter described herein is provided in the general context of being executed in conjunction with the execution of operating systems and application programs on computer systems, it will be appreciated by those skilled in the art that other implementations may also be performed in conjunction with other types of program modules. In general, program modules include routines, programs, components, data structures, and other types of structures that perform specific tasks or implement specific abstract data types. It will be appreciated by those skilled in the art that the subject matter described herein may be practiced using other computer system configurations, including handheld devices, multiprocessor systems, microprocessor-based or programmable consumer electronics, minicomputers, mainframe computers, etc., and may also be used in a distributed computing environment in which tasks are performed by remote processing devices connected via a communication network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices.

[0104] Those of ordinary skill in the art will appreciate that the units and method 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 the present invention.

[0105] It should be understood that the above specific embodiments of the present invention are only used to illustrate or explain the principles of the present invention, and do not constitute a limitation of the present invention. Therefore, any modifications, equivalent substitutions, improvements, etc. made without departing from the spirit and scope of the present invention should be included in the protection scope of the present invention. In addition, the appended claims of the present invention are intended to cover all changes and modifications that fall within the scope and boundaries of the appended claims, or the equivalent forms of such scope and boundaries.

Claims

1. An intelligent detection system for power big data, characterized by: It includes a cloud server and an edge server, wherein the edge server is connected to the cloud server network, wherein: The edge server is connected to a camera network, the camera is used to capture a video stream of a power usage scene, the edge server is used to perform face recognition on the video stream using a deep learning model, is used to identify the power worker in the video stream, is used to further identify whether the power worker's attire complies with the dress code, and is used to report the recognition result to the cloud server; The edge server is also connected to a power data collection device, which is used to collect user power usage information; the edge server is used to determine whether the user's power usage is abnormal based on the user's power usage information through a machine learning model, and to report alarm information and user power usage information to the cloud server when the power usage is abnormal, and to report user power usage information to the cloud server when there is no abnormality in the power usage.

2. The power big data intelligent detection system according to claim 1 is characterized by: It also includes a concentrator, a collector and a terminal device; the concentrator is connected to the cloud server network, a plurality of collectors are connected to the concentrator via a high-speed power line carrier communication network, the edge server is connected one-to-one with the collector via RS-485, and the terminal device includes the camera and the power data acquisition device.

3. A method for intelligent detection of power big data, implemented based on the power big data intelligent detection system according to claim 2, characterized in that: include: The video stream of the power consumption scene is obtained through the camera, and the user's power consumption information is obtained through the power data collection equipment; Performing face recognition on the video stream, if the person in the video stream is identified as an electric power worker, further identifying whether the clothing of the electric power worker complies with the dress code, and reporting the recognition result to the cloud server; Based on the user's electricity usage information, determine whether the user's electricity usage is abnormal. If there is an abnormality, report the alarm information and the user's electricity usage information to the cloud server. If there is no abnormality, report the user's electricity usage information to the cloud server.

4. The power big data intelligent detection method according to claim 3 is characterized in that: The method for performing face recognition on the video stream includes: Acquire a frame image from a video stream, and preprocess the frame image, wherein the preprocessing includes one or more of denoising, cropping, scaling, and grayscale conversion; Performing face detection on the frame image using a deep learning model to determine whether a face exists in the frame image; If there is a face, a convolutional neural network is used to extract facial features, and the facial features are matched with facial features in a preset face library, and corresponding operations are performed on the matching results according to preset rules.

5. The power big data intelligent detection method according to claim 3 is characterized in that: Methods for identifying whether the power workers' attire complies with the dress code include: Capture the image of the power worker in his or her uniform; The power worker's attire image is input into the power worker's attire standard recognition model to identify whether the power worker's attire is standard.

6. The power big data intelligent detection method according to claim 5 is characterized in that: The training method of the power worker dress code recognition model includes: Obtaining standard image materials of electric power workers in uniform and marking them as standard, and non-standard image materials of electric power workers in uniform and marking them as non-standard, and pre-processing the image materials; wherein the pre-processing includes one or more of denoising, cropping, and scaling; The preprocessed image material is used as a sample and the annotation is used as a label to train a convolutional neural network model to obtain a recognition model for the dress code of power workers.

7. The intelligent detection method for electric power big data according to claim 3 is characterized in that: The method for judging whether the user's power consumption is abnormal based on the user's power consumption information includes: Obtain user electricity consumption information collected by power data collection equipment; Inputting the user's electricity usage information into an anti-electricity theft model to determine whether the user's electricity usage is abnormal; If there is any abnormality, an alarm message corresponding to the abnormality is triggered.

8. The power big data intelligent detection method according to claim 7 is characterized in that: The training method of the anti-electricity theft model includes: Collecting electricity consumption data of a preset number of users and cleaning and preprocessing the data; wherein the electricity consumption data includes the user's electricity consumption, electricity consumption time, voltage, and current, and the cleaning and preprocessing include removing outliers, processing missing data, and performing time series analysis; Selecting features related to the user's electricity usage behavior from the electricity usage data to form a sample set, and marking the data in the sample set as a normal electricity usage mode or an abnormal electricity usage mode according to a specific task, wherein the features include electricity usage mode, peak and valley distribution of electricity consumption, daily / weekly / monthly electricity usage mode, and energy-saving behavior; The data in the sample set is used as samples, and the annotations corresponding to the data are used as labels to train the machine learning model to obtain an anti-electricity theft model.

9. An intelligent detection device for power big data, characterized in that: include: An acquisition module is used to acquire a video stream of a power consumption scene through a camera, and to acquire user power consumption information through a power data acquisition device; a recognition module, for performing face recognition on the video stream, for identifying the power workers in the video stream, for further identifying whether the clothing of the power workers complies with the dress code, and for reporting the recognition result to the cloud server; The reporting module is used to determine whether the user's electricity usage is abnormal based on the user's electricity usage information, and to report the alarm information and the user's electricity usage information to the cloud server when the electricity usage is abnormal, and to report the user's electricity usage information to the cloud server when there is no abnormality in the electricity usage.

10. A computer device, characterized in that: Including memory and processor: The memory is used to store executable instructions; The processor is used to execute the executable instructions to implement the detection method according to any one of claims 3-8.

Images