Artificial intelligence model training method and system for identifying electric power scene

Abstract

The invention provides an artificial intelligence model training method and system for identifying an electric power scene, and belongs to the technical field of computers. The method comprises the following steps: based on multiple image acquisition schemes, performing image acquisition on a pre-constructed power scene model to obtain multiple power scene images; wherein each power scene model is constructed based on different scene establishment rules; and inputting each power scene image into the artificial intelligence initial model for model training to obtain an artificial intelligence model for identifying the power scene. According to the method and the system, an artificial intelligence image recognition technology is fully utilized, the electric power scene image model and the electric power scene physical model are combined, the purpose of performing artificial intelligence model recognition and training in the target laboratory can be achieved, the recognition efficiency and accuracy of the typical working condition of the power station are improved, and the working efficiency of the power station is improved. The safety is improved, and staff reduction and efficiency improvement are realized.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to an artificial intelligence model training method for identifying power scenarios, an artificial intelligence model training system for identifying power scenarios, a machine-readable storage medium and an electronic device. Background Art

[0002] The rapid development of image processing technology has promoted the emergence and development of image recognition technology, and has gradually become an important component in the field of artificial intelligence. It is widely used in facial recognition, fingerprint recognition, medical diagnosis and other fields, playing an important role.

[0003] Image recognition is an important field of artificial intelligence. It refers to the technology of using computers to process, analyze and understand images to identify targets and objects of various different patterns. In industrial use, industrial cameras are usually used to take pictures, and then software is used to perform further recognition processing based on the grayscale difference of the pictures.

[0004] Since the emergence of image recognition technology is based on artificial intelligence, the process of computer image recognition is roughly the same as the process of human brain recognition of images. The process of existing image recognition technology includes the following four steps:

[0005] 1. Acquiring information: refers to converting information such as sound and light into electrical signals through sensors, that is, acquiring the basic information of the identified object and converting the acquired basic information of the identified object into information that can be recognized by the computer;

[0006] 2. Information preprocessing: refers to the processing of images by operations such as denoising, transformation and smoothing, based on which the important features of the image are improved;

[0007] 3. Extraction and selection of features: refers to the extraction and selection of image features in pattern recognition. In general, it is to recognize that images have various characteristics. If a certain method is used for separation, the characteristics of the image must be identified. Obtaining features is also called feature extraction.

[0008] 4. Design classifiers and classification decisions: Designing classifiers is based on training and recognition rules. Based on these recognition rules, the main types of features can be obtained, thereby continuously improving the recognition rate of image recognition. After that, by identifying special features, the image can be evaluated and confirmed.

[0009] It can be seen that the process of existing image recognition technology is complicated. How to improve the recognition efficiency and accuracy of typical working conditions of power plants to improve safety and achieve reduction in staff and increase in efficiency is an issue that needs to be solved urgently. Summary of the invention

[0010] The purpose of the embodiments of the present invention is to provide an artificial intelligence model training method and system for identifying power scenarios, so as to at least solve the above-mentioned problem of how to improve the recognition efficiency and accuracy of typical operating conditions of power plants, so as to improve safety and achieve reduction in staff and increase in efficiency.

[0011] In order to achieve the above-mentioned object, the first aspect of the present invention provides an artificial intelligence model training method for identifying power scenarios, comprising:

[0012] Based on multiple image acquisition schemes, image acquisition is performed on the pre-built power scene model to obtain multiple power scene images; wherein each power scene model is constructed based on rules established for different scenes;

[0013] The images of various power scenes are input into the initial artificial intelligence model for model training to obtain an artificial intelligence model for identifying power scenes.

[0014] Optionally, the above-mentioned power scene model includes a power scene image model;

[0015] The above scenario establishment rules include:

[0016] Control the screen in the target laboratory to display the pre-collected power scene pictures to obtain a power scene image model;

[0017] Among them, the power scene image model is used to display power scenes under different background conditions in the form of pictures. The different background conditions include one or more of light intensity data in different weather conditions, light intensity data in different time periods, and different smoke concentration data.

[0018] Optionally, the above-mentioned power scenario model includes a physical model of the power scenario;

[0019] The above scenario establishment rules include:

[0020] Control the light intensity and smoke concentration in the physical scene arranged in the target laboratory to obtain a physical model of the power scene;

[0021] Among them, the physical scene in the target laboratory is arranged based on the actual power scene data. The physical model of the power scene is used to simulate the power scenes under different background conditions by controlling the light intensity and smoke concentration in the physical scene. The different background conditions include one or more of the light intensity data in different weather conditions, the light intensity data in different time periods and the different smoke concentration data.

[0022] Optionally, the acquiring of multiple power scene images includes:

[0023] Get pictures of actual power scenes;

[0024] The background conditions in the actual power scene picture are adjusted so that the actual power scene picture is mixed with different background conditions to obtain multiple power scene images.

[0025] Optionally, the above-mentioned multiple image acquisition schemes include a scheme of acquiring images through a static camera and a scheme of acquiring images through a dynamic camera.

[0026] Optionally, the above-mentioned solution for collecting images through a dynamic camera includes:

[0027] Control the preset inspection device to collect images according to the preset inspection rate and preset inspection route;

[0028] Among them, the preset inspection device, preset inspection rate and preset inspection route are all determined by the actual inspection information of the actual power scene, and the preset inspection device is equipped with a high-definition camera and a thermal imaging camera.

[0029] Optionally, the acquiring of multiple power scene images includes:

[0030] Through cloud-edge collaboration, a video of the power production process of an actual power scene captured by a remote camera is obtained; wherein the video of the power production process of the actual power scene includes multiple power scene images.

[0031] A second aspect of the present invention provides an artificial intelligence model training system for identifying power scenarios, comprising:

[0032] An image acquisition module is used to acquire images of a pre-built power scene model based on a variety of image acquisition schemes to obtain a plurality of power scene images; wherein each power scene model is constructed based on rules for establishing different scenes;

[0033] The model training module is used to input various power scene images into the artificial intelligence initial model for model training to obtain an artificial intelligence model for identifying power scenes.

[0034] In a third aspect of the present invention, a machine-readable storage medium is provided, on which instructions are stored, and when the instructions are executed by a processor, the processor is configured to execute the above-mentioned artificial intelligence model training method for identifying power scenarios.

[0035] In a fourth aspect of the present invention, an electronic device is provided, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the artificial intelligence model training method for identifying power scenarios when executing the computer program.

[0036] Through the above technical scheme, a method and system for training an artificial intelligence model for identifying power scenes are provided. For the power production process, on the basis of a pre-constructed power scene model (including a power scene image model and a power scene physical model), multiple image acquisition schemes are used to perform image acquisition to obtain multiple power scene images. Different power scene models are constructed based on rules established in different scenarios. Each power scene image is input into the artificial intelligence initial model for model training to obtain an artificial intelligence model for identifying power scenes. The method and system make full use of the image recognition technology of artificial intelligence, combining the power scene image model with the power scene physical model, and can achieve the purpose of identifying and training the artificial intelligence model in the target laboratory, thereby improving the recognition efficiency and accuracy of typical working conditions of the power station, so as to improve safety and achieve efficiency reduction.

[0037] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0039] Figure 1 It is a flow chart of an artificial intelligence model training method for identifying power scenarios provided by one embodiment of the present invention;

[0040] Figure 2 is a block diagram of an artificial intelligence model training system for identifying power scenarios provided by one embodiment of the present invention;

[0041] Figure 3 It is a schematic diagram of the structure of an electronic device provided by a preferred embodiment of the present invention.

[0042] Description of Reference Numerals

[0043] 10 - electronic device, 100 - processor, 101 - memory, 102 - computer program. DETAILED DESCRIPTION

[0044] The specific implementation of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the present invention, and is not used to limit the present invention.

[0045] Figure 1 FIG. 1 is a flowchart of an artificial intelligence model training method for identifying power scenarios provided by an embodiment of the present invention. Figure 1As shown, an embodiment of the present invention provides an artificial intelligence model training method for identifying power scenarios, comprising:

[0046] S110: Based on multiple image acquisition schemes, perform image acquisition on a pre-constructed power scene model to obtain multiple power scene images; wherein each power scene model is constructed based on rules for establishing different scenes;

[0047] In some implementations of the present embodiment, the above-mentioned power scene model includes a power scene image model; the above-mentioned scene establishment rules include: controlling the screen in the target laboratory to display the pre-collected power scene pictures to obtain the power scene image model; wherein the power scene image model is used to display power scenes in different background conditions in the form of pictures, and the different background conditions include one or more of light intensity data in different weather conditions, light intensity data in different time periods, and different smoke concentration data.

[0048] Specifically, the artificial intelligence model training method for identifying power scenes can implement model training in a target laboratory (it should be noted that the target laboratory represents the experimental area where power scene images are centrally collected). Using a screen arranged in the target laboratory (for example, a high-definition large screen that can display in high definition), typical scenes of the power station are displayed in the form of pictures (i.e., pre-collected power scene pictures, such as power scene pictures including "on-site meter reading recognition", power scene pictures including "abnormal meter recognition", power scene pictures including "improper wearing of safety helmet recognition", power scene pictures including "fire / smoke recognition", power scene pictures including "coal spilling recognition", power scene pictures including "belt deviation recognition", power scene pictures including "personnel fall recognition", power scene pictures including "indicator light recognition", power scene pictures including "cabinet door switch status recognition", power scene pictures including "knife switch status recognition", and power scene pictures including "liquid level recognition"), so as to obtain a power scene image model displayed in the form of pictures, so that a high-definition camera can be used to collect images of the power scene image model during the image acquisition stage. The power scene pictures included in the power scene image model can also simulate the conditions of different weather, light in different time periods, or when there are interfering backgrounds such as smoke and dust on site, so as to enhance the authenticity of the subsequently trained AI recognition model (i.e., the artificial intelligence model used to identify power scenes).

[0049] In the above implementation process, the power scene picture displayed on the screen can be derived from the input of offline power scene pictures, or can be obtained through a field model uploaded from the Internet.

[0050] In some implementations of the present embodiment, the above-mentioned power scene model includes a physical model of the power scene; the above-mentioned scene establishment rules include: controlling the light intensity and smoke concentration in the physical scene arranged in the target laboratory to obtain the physical model of the power scene; wherein the physical scene in the target laboratory is arranged based on actual power scene data, and the physical model of the power scene is used to simulate power scenes in different background conditions by controlling the light intensity and smoke concentration in the physical scene, and the different background conditions include one or more of light intensity data in different weather conditions, light intensity data in different time periods, and different smoke concentration data.

[0051] Specifically, according to the actual power scene data, a real scene is arranged in the target laboratory (for example, a real scene of "electric welding operation", a real scene of "excavator operation", etc.) to obtain a physical scene (a physical scene refers to a scene containing real objects set up in the target laboratory). By controlling the light intensity and smoke concentration in the physical scene, real working conditions such as different light and / or smoke background can be simulated, so that the obtained physical model of the power scene can simulate the real power scenes under different background conditions. Therefore, in the image acquisition stage, a high-definition camera can be used to capture images of the real power scenes in the physical model of the power scene, so as to further ensure the authenticity and accuracy of the subsequently trained AI recognition model (that is, the artificial intelligence model used to identify power scenes).

[0052] In some implementations of this embodiment, the above-mentioned acquisition of multiple power scene images includes: acquiring an actual power scene picture; adjusting the background conditions in the actual power scene picture to mix the actual power scene picture with different background conditions to obtain multiple power scene images.

[0053] Specifically, the method can also receive on-site pictures taken by a mobile camera (i.e., pictures of actual power scenes), mix the actual power scene pictures with light and climate scenes of different backgrounds, and send them to the AI ​​server for state recognition and model training to further improve the authenticity and accuracy of the trained AI recognition model (i.e., the artificial intelligence model used to identify power scenes).

[0054] In some implementations of this embodiment, the above-mentioned multiple image acquisition schemes include a scheme of acquiring images through a static camera and a scheme of acquiring images through a dynamic camera.

[0055] Specifically, the specific content of the solution for image acquisition through a static camera may be to collect image information and thermal imaging information by arranging a fixed intelligent camera (i.e., to collect images of a pre-built power scene model through a fixed intelligent camera). The specific content of the solution for image acquisition through a dynamic camera may be to collect and identify object image information through a high-definition camera and a thermal imaging camera installed on a track inspection robot and / or a patrol robot dog (i.e., to collect images of a pre-built power scene model through a high-definition camera and a thermal imaging camera installed on a track inspection robot and / or a patrol robot dog).

[0056] During the above implementation process, the physical scene and the video scene are displayed on a high-definition large screen, captured by static cameras and dynamic cameras, and the AI ​​model is recognized and trained to finally obtain a model of typical power application scenarios (i.e., an artificial intelligence model for identifying power scenarios).

[0057] In some implementations of this embodiment, the content of the above-mentioned scheme for image acquisition through a dynamic camera includes: controlling the preset inspection device to perform image acquisition according to a preset inspection rate and a preset inspection route; wherein the preset inspection device, the preset inspection rate and the preset inspection route are all determined by actual inspection information of the actual power scene, and the preset inspection device is provided with a high-definition camera and a thermal imaging camera.

[0058] Exemplarily, the above-mentioned preset inspection device may be a track inspection robot and / or an inspection robot dog.

[0059] Specifically, the track inspection robot performs inspection and image recognition at a preset inspection rate to simulate the track inspection machine system at the power production site. The inspection robot dog performs automatic inspection and image recognition at a preset inspection rate and preset inspection route to simulate the inspection machine system at the power production site.

[0060] In some implementations of this embodiment, the above-mentioned acquisition of multiple power scene images includes: acquiring, through cloud-edge collaboration, a video of the power production process of an actual power scene captured by a remote camera; wherein the video of the power production process of the actual power scene includes multiple power scene images.

[0061] Specifically, the video of the power production process of the actual power scene captured by the remote camera is transmitted to the computer of the target laboratory through the server and the Internet dedicated line to realize the cloud-edge collaboration of the power scene image data used to train the AI ​​model, and expand the service scope of the target laboratory to the power production site. Through the cloud-edge collaboration, the purpose of remote service of obtaining power scene images at the power site can be achieved, which also achieves the purpose of obtaining remote power scene images.

[0062] S120: Input each power scene image into an artificial intelligence initial model for model training to obtain an artificial intelligence model for identifying power scenes.

[0063] Specifically, each power scene image collected on the basis of the power scene image model and the power scene physical model is input into the artificial intelligence initial model for model training to generate an AI recognition model (i.e., an artificial intelligence model for identifying power scenes). This method makes full use of video image recognition technology and robot inspection technology, combines the large-screen image recognition process with the physical scene recognition process, and simulates various working conditions such as various lighting conditions and backgrounds with interference conditions such as smoke. It combines image recognition with the model training server to achieve the purpose of image recognition of real power scenes. And combined with cloud-edge collaborative technology, the input end of the target laboratory is extended to the power production site, which can realize remote status recognition and service. It is of great significance to improve the recognition efficiency of AI models, reduce costs, achieve staff reduction and efficiency improvement, and improve economic efficiency.

[0064] In the above implementation process, the method uses a variety of image acquisition schemes to acquire images based on the pre-constructed power scene model (including the power scene image model and the power scene physical model) for the power production process to obtain multiple power scene images. Different power scene models are constructed based on rules for different scenarios. Each power scene image is input into the artificial intelligence initial model for model training to obtain an artificial intelligence model for identifying power scenes. This method makes full use of the image recognition technology of artificial intelligence, combines the power scene image model with the power scene physical model, and can achieve the purpose of identifying and training the artificial intelligence model in the target laboratory, thereby improving the recognition efficiency and accuracy of typical working conditions of the power station, so as to improve safety and achieve efficiency reduction.

[0065] Figure 2 FIG. 1 is a block diagram of an artificial intelligence model training system for identifying power scenarios provided by an embodiment of the present invention. Figure 2 As shown, an embodiment of the present invention provides an artificial intelligence model training system for identifying power scenarios, including:

[0066] An image acquisition module is used to acquire images of a pre-built power scene model based on a variety of image acquisition schemes to obtain a plurality of power scene images; wherein each power scene model is constructed based on rules for establishing different scenes;

[0067] The model training module is used to input various power scene images into the artificial intelligence initial model for model training to obtain an artificial intelligence model for identifying power scenes.

[0068] Specifically, the system uses a variety of image acquisition schemes to acquire images based on pre-built power scene models (including power scene image models and power scene physical models) for the power production process to obtain multiple power scene images. Different power scene models are constructed based on rules established in different scenarios. Each power scene image is input into the initial artificial intelligence model for model training to obtain an artificial intelligence model for identifying power scenes. The system makes full use of artificial intelligence image recognition technology, combining power scene image models with power scene physical models, and can achieve the purpose of identifying and training artificial intelligence models in the target laboratory, thereby improving the recognition efficiency and accuracy of typical working conditions of power plants, so as to improve safety and achieve efficiency reduction.

[0069] An embodiment of the present invention also provides a machine-readable storage medium, on which instructions are stored. When the instructions are executed by the processor 100, the processor 100 is configured to execute the above-mentioned artificial intelligence model training method for identifying power scenarios.

[0070] Machine-readable storage media include permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology to store information. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0071] An embodiment of the present invention also provides an electronic device 10, which includes a memory 101, a processor 100, and a computer program 102 stored in the memory 101 and executable on the processor 100. When the processor 100 executes the computer program 102, the above-mentioned artificial intelligence model training method for identifying power scenarios is implemented.

[0072] like Figure 3 FIG. 1 is a schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 3As shown, the electronic device 10 of this embodiment includes: a processor 100, a memory 101, and a computer program 102 stored in the memory 101 and executable on the processor 100. When the processor 100 executes the computer program 102, the steps in the above method embodiment are implemented. Alternatively, when the processor 100 executes the computer program 102, the functions of each module / unit in the above device embodiment are implemented.

[0073] Exemplarily, the computer program 102 may be divided into one or more modules / units, one or more modules / units are stored in the memory 101, and are executed by the processor 100 to complete the present invention. One or more modules / units may be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer program 102 in the electronic device 10. For example, the computer program 102 may be divided into an image acquisition module and a model training module.

[0074] The electronic device 10 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The electronic device 10 may include, but is not limited to, a processor 100 and a memory 101. Those skilled in the art will appreciate that Figure 3 It is only an example of the electronic device 10 and does not constitute a limitation of the electronic device 10. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0075] The processor 100 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0076] The memory 101 may be an internal storage unit of the electronic device 10, such as a hard disk or memory of the electronic device 10. The memory 101 may also be an external storage device of the electronic device 10, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 10. Further, the memory 101 may also include both an internal storage unit of the electronic device 10 and an external storage device. The memory 101 is used to store computer programs and other programs and data required by the electronic device 10. The memory 101 may also be used to temporarily store data that has been output or is to be output.

[0077] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0078] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program 102 products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program 102 product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0079] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program 102 products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by the computer program 102 instructions. These computer program 102 instructions can be provided to a processor 100 of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor 100 of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0080] These computer program 102 instructions may also be stored in a computer readable memory 101 that can direct a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory 101 produce an article of manufacture including an instruction device that implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0081] These computer program 102 instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0082] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0083] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. An artificial intelligence model training method for identifying power scenarios, characterized in that: include: Based on multiple image acquisition schemes, image acquisition is performed on the pre-built power scene model to obtain multiple power scene images; wherein each power scene model is constructed based on rules established for different scenes; The images of various power scenes are input into the initial artificial intelligence model for model training to obtain an artificial intelligence model for identifying power scenes.

2. The artificial intelligence model training method for identifying power scenarios according to claim 1 is characterized in that: The electric power scene model includes an electric power scene image model; The scene establishment rules include: Control the screen in the target laboratory to display the pre-collected power scene pictures to obtain a power scene image model; The power scene image model is used to display power scenes under different background conditions in the form of pictures, and the different background conditions include one or more of light intensity data in different weather conditions, light intensity data in different time periods, and different smoke concentration data.

3. The artificial intelligence model training method for identifying power scenarios according to claim 1 is characterized in that: The power scenario model includes a physical model of the power scenario; The scene establishment rules include: Control the light intensity and smoke concentration in the physical scene arranged in the target laboratory to obtain a physical model of the power scene; Among them, the physical scene in the target laboratory is arranged based on the actual power scene data, and the physical model of the power scene is used to simulate the power scenes under different background conditions by controlling the light intensity and smoke concentration in the physical scene. The different background conditions include one or more of light intensity data in different weather conditions, light intensity data in different time periods, and different smoke concentration data.

4. The artificial intelligence model training method for identifying power scenarios according to claim 3 is characterized in that: The acquiring of multiple power scene images comprises: Get pictures of actual power scenes; The background conditions in the actual power scene picture are adjusted so that the actual power scene picture is mixed with different background conditions to obtain multiple power scene images.

5. The artificial intelligence model training method for identifying power scenarios according to claim 1 is characterized in that: The multiple image acquisition schemes include a scheme of acquiring images through a static camera and a scheme of acquiring images through a dynamic camera.

6. The artificial intelligence model training method for identifying power scenarios according to claim 5 is characterized in that: The contents of the scheme for image acquisition by a dynamic camera include: Control the preset inspection device to collect images according to the preset inspection rate and preset inspection route; Among them, the preset inspection device, the preset inspection rate and the preset inspection route are all determined by actual inspection information of actual power scenarios, and the preset inspection device is provided with a high-definition camera and a thermal imaging camera.

7. The artificial intelligence model training method for identifying power scenarios according to claim 1, characterized in that: The acquiring of multiple power scene images comprises: Through cloud-edge collaboration, a video of the power production process of an actual power scene captured by a remote camera is obtained; wherein the video of the power production process of the actual power scene includes multiple power scene images.

8. An artificial intelligence model training system for identifying power scenarios, characterized in that: include: An image acquisition module is used to acquire images of a pre-built power scene model based on a variety of image acquisition schemes to obtain a plurality of power scene images; wherein each power scene model is constructed based on rules for establishing different scenes; The model training module is used to input various power scene images into the artificial intelligence initial model for model training to obtain an artificial intelligence model for identifying power scenes.

9. A machine-readable storage medium having instructions stored thereon, characterized in that: When the instruction is executed by a processor, the processor is configured to execute the artificial intelligence model training method for identifying power scenarios as described in any one of claims 1 to 7.

10. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, it implements the artificial intelligence model training method for identifying power scenarios as described in any one of claims 1 to 7.

Images