A cloud-edge collaborative power distribution model application service method and device

Abstract

The application provides a cloud-edge collaborative power distribution model application service method and device, including: taking a single physical quantity collected by an end-side device as an input of a pre-trained end-side device machine learning model to obtain a judgment result of an end-side device control object output by the pre-trained end-side device machine learning model; collecting the single physical quantity collected by each end-side device in a region and the judgment result of the end-side device control object, and taking the single physical quantity and the judgment result as an input of a pre-trained edge-side device machine learning model to obtain a region judgment result output by the pre-trained edge-side device machine learning model; collecting the region judgment results, and taking the region judgment results and power distribution network characteristic information as an input of a pre-trained cloud-end machine learning model to obtain a power distribution system judgment result output by the pre-trained cloud-end machine learning model; and the scheme can economically and efficiently obtain and utilize the pre-trained model to solve actual problems.

Description

Technical Field

[0001] This invention relates to the field of digital and intelligent power system technology, specifically to a cloud-edge collaborative power distribution model application service method and device. Background Technology

[0002] With the widespread integration of distributed power sources and electric vehicles, power distribution networks are undergoing unprecedented transformation, evolving from a single, closed structure into a complex and diverse network system, becoming a crucial platform for driving energy transformation and resource allocation. Faced with this challenge, cutting-edge digital and intelligent technologies, such as artificial intelligence and the Internet of Things (IoT), are rapidly emerging, becoming the core driving force for technological innovation in power distribution networks. For example, smart IoT systems have enabled comprehensive perception and intelligent management of the power grid. Smart vehicle-to-everything (V2X) platforms provide convenient and intelligent charging and battery swapping services for the electric vehicle industry, promoting efficient energy interaction. Based on the "π model theory," the concept of a digital power grid has constructed an architecture that deeply integrates physical, information, and business aspects, further enhancing the intelligence level and operational efficiency of power distribution networks. These achievements have collectively driven a comprehensive upgrade in the power distribution field, laying a solid foundation for future energy development.

[0003] However, with the emergence of a large number of energy producers and consumers, the increasing demand for distributed source-load localization and local supply-demand balance, and the challenges of flexible interaction between the power grid and users across domains and applications, the traditional closed and fixed data interaction and model application models are becoming unsustainable. Summary of the Invention

[0004] To overcome the above-mentioned shortcomings, this invention proposes a cloud-edge collaborative power distribution model application service method and device.

[0005] Firstly, a cloud-edge collaborative power distribution model application service method is provided, the method comprising:

[0006] The single physical quantity collected by the edge device is used as the input of the pre-trained edge device machine learning model to obtain the judgment result of the edge device inspection object output by the pre-trained edge device machine learning model.

[0007] The system collects the single physical quantities collected by each end-side device within the region and the judgment results of the objects monitored by each end-side device, and uses them as input to a pre-trained side-side device machine learning model to obtain the region judgment result output by the pre-trained side-side device machine learning model.

[0008] The judgment results from each region are collected, and the judgment results from each region and the distribution network characteristic information are used as input to a pre-trained cloud machine learning model to obtain the distribution system judgment result output by the pre-trained cloud machine learning model.

[0009] The judgment results include at least one of the following: system optimization decision results, health evaluation results, user profile tags, defect identification results, energy efficiency prediction results, and control command results supporting power supply operation.

[0010] Preferably, the end-side device includes: a smart switch, a smart meter, and an end-side sensor.

[0011] Preferably, the single physical quantity includes: video signal, sound signal, power consumption data, temperature signal, vibration signal, humidity signal, or pressure signal.

[0012] Preferably, the training process of the pre-trained edge device machine learning model includes:

[0013] Training data is constructed using historical data of a single physical quantity collected by edge devices and their corresponding judgment results;

[0014] The machine learning model of the edge device is trained using the training data to obtain the pre-trained machine learning model of the edge device.

[0015] Preferably, the training process of the pre-trained side-device machine learning model includes:

[0016] Training data is constructed by using historical data of single physical quantities collected by each end-side device in the region, historical data of judgment results of the objects monitored by each end-side device, and their corresponding regional judgment results;

[0017] The edge device machine learning model is trained using the training data to obtain the pre-trained edge device machine learning model.

[0018] Preferably, the power distribution network characteristic information includes at least one of the following: power distribution network ledger data and meteorological data.

[0019] Preferably, the edge device includes: an edge computing terminal and a smart gateway.

[0020] Preferably, the training process of the pre-trained cloud machine learning model includes:

[0021] Training data is constructed using historical data of judgment results in each region, historical data of distribution network characteristic information, and their corresponding judgment results in the distribution system.

[0022] The cloud-based machine learning model is trained using the training data to obtain the pre-trained cloud-based machine learning model.

[0023] Secondly, a cloud-edge collaborative power distribution model application service device is provided, the cloud-edge collaborative power distribution model application service device comprising:

[0024] The first analysis module is used to take a single physical quantity collected by the end-side device as input to a pre-trained end-side device machine learning model and obtain the judgment result of the end-side device inspection object output by the pre-trained end-side device machine learning model.

[0025] The second analysis module is used to collect the single physical quantities collected by each end-side device in the area and the judgment results of the objects monitored by each end-side device, and use them as input to the pre-trained side-side device machine learning model to obtain the area judgment result output by the pre-trained side-side device machine learning model.

[0026] The third analysis module is used to collect the judgment results of each region and use the judgment results of each region and the distribution network characteristic information as input to a pre-trained cloud machine learning model to obtain the distribution system judgment result output by the pre-trained cloud machine learning model.

[0027] The judgment results include at least one of the following: system optimization decision results, health evaluation results, user profile tags, defect identification results, energy efficiency prediction results, and control command results supporting power supply operation.

[0028] Preferably, the end-side device includes: a smart switch, a smart meter, and an end-side sensor.

[0029] Preferably, the single physical quantity includes: video signal, sound signal, power consumption data, temperature signal, vibration signal, humidity signal, or pressure signal.

[0030] Preferably, the training process of the pre-trained edge device machine learning model includes:

[0031] Training data is constructed using historical data of a single physical quantity collected by edge devices and their corresponding judgment results;

[0032] The machine learning model of the edge device is trained using the training data to obtain the pre-trained machine learning model of the edge device.

[0033] Preferably, the training process of the pre-trained side-device machine learning model includes:

[0034] Training data is constructed by using historical data of single physical quantities collected by each end-side device in the region, historical data of judgment results of the objects monitored by each end-side device, and their corresponding regional judgment results;

[0035] The edge device machine learning model is trained using the training data to obtain the pre-trained edge device machine learning model.

[0036] Preferably, the power distribution network characteristic information includes at least one of the following: power distribution network ledger data and meteorological data.

[0037] Preferably, the edge device includes: an edge computing terminal and a smart gateway.

[0038] Preferably, the training process of the pre-trained cloud machine learning model includes:

[0039] Training data is constructed using historical data of judgment results in each region, historical data of distribution network characteristic information, and their corresponding judgment results in the distribution system.

[0040] The cloud-based machine learning model is trained using the training data to obtain the pre-trained cloud-based machine learning model.

[0041] Thirdly, a computer device is provided, comprising: one or more processors;

[0042] The processor is used to execute one or more programs;

[0043] When the one or more programs are executed by the one or more processors, the cloud-edge collaborative power distribution model application service method is implemented.

[0044] Fourthly, a computer-readable storage medium is provided, on which a computer program is stored, wherein when the computer program is executed, the cloud-edge collaborative power distribution model application service method is implemented.

[0045] The above-described technical solutions of the present invention have at least one or more of the following beneficial effects:

[0046] This invention provides a cloud-edge collaborative power distribution model application service method and apparatus, comprising: using a single physical quantity collected by an end-side device as input to a pre-trained end-side device machine learning model to obtain the judgment result of the end-side device monitoring object output by the pre-trained end-side device machine learning model; collecting the single physical quantity collected by each end-side device in the region and the judgment results of each end-side device monitoring object, and using them as input to a pre-trained edge-side device machine learning model to obtain the regional judgment result output by the pre-trained edge-side device machine learning model; collecting the regional judgment results, and using the regional judgment results and power distribution network characteristic information as input to a pre-trained cloud machine learning model to obtain the power distribution system judgment result output by the pre-trained cloud machine learning model; wherein, the judgment result includes at least one of the following: system optimization decision result, health evaluation result, user profile label, defect identification result, energy efficiency prediction result, and control command result supporting power supply operation. This invention, by deploying the pre-trained model at different levels of cloud-edge-end, can economically and efficiently acquire and utilize the pre-trained model to solve practical problems. Attached Figure Description

[0047] Figure 1 This is a schematic diagram of the main steps of the cloud-edge collaborative power distribution model application service method according to an embodiment of the present invention;

[0048] Figure 2 This is a schematic diagram of the cloud-edge collaborative power distribution model application service method according to an embodiment of the present invention. Detailed Implementation

[0049] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0050] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] As disclosed in the background section, with the widespread integration of distributed power sources and electric vehicles, the power distribution network is undergoing unprecedented transformation, evolving from a single, closed structure into a complex and diverse network system, becoming a crucial platform for driving energy transformation and resource allocation. Faced with this challenge, cutting-edge digital and intelligent technologies such as artificial intelligence and the Internet of Things (IoT) are rapidly emerging, becoming the core force driving technological innovation in the power distribution network. For example, smart IoT systems have enabled comprehensive perception and intelligent management of the power grid. Smart vehicle-to-everything (V2X) platforms provide convenient and intelligent charging and battery swapping services for the electric vehicle industry, promoting efficient energy interaction. Based on the "π model theory," the concept of a digital power grid has constructed an architecture that deeply integrates physical, information, and business aspects, further enhancing the intelligence level and operational efficiency of the power distribution network. These achievements have collectively driven a comprehensive upgrade in the power distribution field, laying a solid foundation for future energy development.

[0052] However, with the emergence of a large number of energy producers and consumers, the increasing demand for distributed source-load localization and local supply-demand balance, and the accompanying issues of flexible interaction between the power grid and users across domains and applications, the traditional closed and fixed data interaction and model application models are becoming unsustainable.

[0053] To address the aforementioned issues, this invention provides a cloud-edge collaborative power distribution model application service method and apparatus, comprising: using a single physical quantity collected by an end-side device as input to a pre-trained end-side device machine learning model to obtain the judgment result of the end-side device monitoring object output by the pre-trained end-side device machine learning model; collecting the single physical quantities collected by each end-side device within a region and the judgment results of each end-side device monitoring object, and using them as input to a pre-trained edge-side device machine learning model to obtain the regional judgment result output by the pre-trained edge-side device machine learning model; collecting the regional judgment results, and using the regional judgment results and power distribution network characteristic information as input to a pre-trained cloud-based machine learning model to obtain the power distribution system judgment result output by the pre-trained cloud-based machine learning model; wherein the judgment result includes at least one of the following: system optimization decision result, health evaluation result, user profile label, defect identification result, energy efficiency prediction result, and control command result supporting power supply operation. This invention, by deploying the pre-trained model at different levels of cloud-edge-end, can economically and efficiently acquire and utilize the pre-trained model to solve practical problems.

[0054] The above plan will be explained in detail below.

[0055] Example 1

[0056] See appendix Figure 1 , Figure 1 This is a schematic flowchart illustrating the main steps of a cloud-edge collaborative power distribution model application service method according to an embodiment of the present invention. Figure 1 As shown, the cloud-edge collaborative power distribution model application service method in this embodiment of the invention mainly includes the following steps:

[0057] Step S101: Use the single physical quantity collected by the edge device as the input of the pre-trained edge device machine learning model to obtain the judgment result of the edge device inspection object output by the pre-trained edge device machine learning model;

[0058] Step S102: Collect the single physical quantities collected by each end-side device in the area and the judgment results of the objects monitored by each end-side device, and use them as input to the pre-trained side-side device machine learning model to obtain the area judgment result output by the pre-trained side-side device machine learning model.

[0059] Step S103: Collect the judgment results of each region, and use the judgment results of each region and the distribution network characteristic information as input to the pre-trained cloud machine learning model to obtain the distribution system judgment result output by the pre-trained cloud machine learning model;

[0060] The judgment results include at least one of the following: system optimization decision results, health evaluation results, user profile tags, defect identification results, energy efficiency prediction results, and control command results supporting power supply operation.

[0061] In this embodiment, the end-side device includes: a smart switch, a smart meter, and an end-side sensor.

[0062] In this embodiment, the single physical quantity includes: video signal, sound signal, power consumption data, temperature signal, vibration signal, humidity signal, or pressure signal.

[0063] In this embodiment, the training process of the pre-trained edge device machine learning model includes:

[0064] Training data is constructed using historical data of a single physical quantity collected by edge devices and their corresponding judgment results;

[0065] The machine learning model of the edge device is trained using the training data to obtain the pre-trained machine learning model of the edge device.

[0066] In this embodiment, the training process of the pre-trained side-device machine learning model includes:

[0067] Training data is constructed by using historical data of single physical quantities collected by each end-side device in the region, historical data of judgment results of the objects monitored by each end-side device, and their corresponding regional judgment results;

[0068] The edge device machine learning model is trained using the training data to obtain the pre-trained edge device machine learning model.

[0069] In this embodiment, the power distribution network characteristic information includes at least one of the following: power distribution network ledger data and meteorological data.

[0070] In this embodiment, the edge device includes: an edge computing terminal and a smart gateway.

[0071] In this embodiment, the training process of the pre-trained cloud machine learning model includes:

[0072] Training data is constructed using historical data of judgment results in each region, historical data of distribution network characteristic information, and their corresponding judgment results in the distribution system.

[0073] The cloud-based machine learning model is trained using the training data to obtain the pre-trained cloud-based machine learning model.

[0074] In one specific implementation, such as Figure 2As shown, on the cloud side, this invention combines intelligent and customized model services for the power distribution industry. For specific needs, training or optimization can be performed on specific data and tasks to develop large-scale industry models and applications with professional knowledge and capabilities, providing users with more professional and efficient AI services. In the power industry, relying on the massive data and powerful computing resources of the cloud platform, large-scale power industry models and specific power industry scenario-specific models are trained to support the needs of power supply operation mode orchestration, transformer area overload prediction, and power equipment defect identification. Simultaneously, the cloud platform manages the intelligent models, continuously training and optimizing them through collected data to achieve model updates and iterations. Furthermore, the cloud can perform comprehensive reasoning based on multiple models to generate optimal decisions and push them to users.

[0075] On the edge side, the integration of the model and edge devices enhances the intelligence level of these devices. Furthermore, the model's proximity to real-time data at the edge facilitates real-time inference and decision-making. The edge layer receives data uploaded from its managed terminal devices in real time and uploads this data to the cloud. The intelligent model deployed at the edge optimizes and iterates based on the continuously growing and updated data uploaded from these terminals. The edge-deployed intelligent model performs inference and generates decisions based on the real-time data uploaded from the terminals, then pushes these decisions to the cloud and the terminals.

[0076] Taking substations as an example, traditional monitoring methods mainly rely on manual inspections and fixed camera surveillance, which suffer from problems such as long inspection cycles, slow response times, and numerous blind spots. The cloud-edge-device collaborative audio-video AI monitoring model for substations integrates the collaborative capabilities of cloud computing, edge computing, and terminal devices. It deploys model services on the cloud, edge, and device sides, including audio data noise suppression models, vibration signal feature extraction models, vibration signal imaging models, and spatiotemporal correlation feature fusion for partial discharge type identification, enabling intelligent, efficient, and safe monitoring of substations.

[0077] The cloud-edge-device collaborative AI monitoring model application deployment solution monitors the operation status of power distribution stations based on sound field visualization imaging and video AI recognition technology; it enables plug-and-play data, message forwarding and proxying, and edge computing and cloud collaboration with edge video AI terminals; it deploys the power distribution station AI monitoring model in the cloud, performs comprehensive mining and analysis on various types of online monitoring data, and achieves dynamic model optimization; and it achieves information sharing and collaborative management with PMS, power supply service system and other operation and maintenance management systems through interface specifications and interaction mechanisms.

[0078] The "end" side equipment consists of sensors and a convergence unit. While collecting information, the sensors perform preliminary local information processing. Threshold exceeding limits, change events, etc., are directly collected, judged, and recorded by the sensor equipment and uploaded to the convergence unit in a standard format.

[0079] "Edge" devices mainly refer to aggregation unit devices, including edge video AI computing terminals and smart gateways. The edge video AI computing terminals support simultaneous access, processing, and analysis of no fewer than 16 video feeds and 8 acoustic array sensor data streams. Aggregation unit devices collect information within the station to achieve data aggregation, and simultaneously construct self-describing files of station information using standard models to achieve plug-and-play functionality at the substation level. The aggregation unit utilizes station information to perform diagnostic mining algorithms for defect identification, improving or correcting the sensitivity or accuracy of sensor warning information. After recording the correlation information across the entire station before and after an event, the completed accident sample data is uploaded to the cloud system to support deep mining and statistical algorithms.

[0080] The "cloud" side equipment includes service cloud and management cloud. The service cloud, working in conjunction with edge devices, forms a complete algorithm system and operation and maintenance management, and is a key component of the substation audio-video AI monitoring system. It is generally deployed on a provincial cloud platform, acquiring data from the IoT platform via micro-applications and linking with work order systems. When conditions permit, it can also be integrated with PMS and supply chain management systems to expand the raw data for diagnostic algorithms and improve diagnostic effectiveness. For well-developed enterprise middleware cloud platforms, they can obtain various data and computing resources from resource and business middleware platforms to fully leverage the efficiency of these platforms. The service cloud primarily implements functions such as algorithm upgrades and maintenance, defect sample collection and summarization, and evaluation of improved operation and maintenance efficiency, supporting the development and improvement of equipment diagnostic technology and promoting the practical application of defect identification and health assessment.

[0081] Example 2

[0082] Based on the same inventive concept, the present invention also provides a cloud-edge collaborative power distribution model application service device, the cloud-edge collaborative power distribution model application service device comprising:

[0083] The first analysis module is used to take a single physical quantity collected by the end-side device as input to a pre-trained end-side device machine learning model and obtain the judgment result of the end-side device inspection object output by the pre-trained end-side device machine learning model.

[0084] The second analysis module is used to collect the single physical quantities collected by each end-side device in the area and the judgment results of the objects monitored by each end-side device, and use them as input to the pre-trained side-side device machine learning model to obtain the area judgment result output by the pre-trained side-side device machine learning model.

[0085] The third analysis module is used to collect the judgment results of each region and use the judgment results of each region and the distribution network characteristic information as input to a pre-trained cloud machine learning model to obtain the distribution system judgment result output by the pre-trained cloud machine learning model.

[0086] The judgment results include at least one of the following: system optimization decision results, health evaluation results, user profile tags, defect identification results, energy efficiency prediction results, and control command results supporting power supply operation.

[0087] Preferably, the end-side device includes: a smart switch, a smart meter, and an end-side sensor.

[0088] Preferably, the single physical quantity includes: video signal, sound signal, power consumption data, temperature signal, vibration signal, humidity signal, or pressure signal.

[0089] Preferably, the training process of the pre-trained edge device machine learning model includes:

[0090] Training data is constructed using historical data of a single physical quantity collected by edge devices and their corresponding judgment results;

[0091] The machine learning model of the edge device is trained using the training data to obtain the pre-trained machine learning model of the edge device.

[0092] Preferably, the training process of the pre-trained side-device machine learning model includes:

[0093] Training data is constructed by using historical data of single physical quantities collected by each end-side device in the region, historical data of judgment results of the objects monitored by each end-side device, and their corresponding regional judgment results;

[0094] The edge device machine learning model is trained using the training data to obtain the pre-trained edge device machine learning model.

[0095] Preferably, the power distribution network characteristic information includes at least one of the following: power distribution network ledger data and meteorological data.

[0096] Preferably, the edge device includes: an edge computing terminal and a smart gateway.

[0097] Preferably, the training process of the pre-trained cloud machine learning model includes:

[0098] Training data is constructed using historical data of judgment results in each region, historical data of distribution network characteristic information, and their corresponding judgment results in the distribution system.

[0099] The cloud-based machine learning model is trained using the training data to obtain the pre-trained cloud-based machine learning model.

[0100] Example 3

[0101] Based on the same inventive concept, this invention also provides a computer device, which includes a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in the computer storage medium to implement corresponding method flows or corresponding functions, thereby realizing the steps of a cloud-edge collaborative power distribution model application service method in the above embodiments.

[0102] Example 4

[0103] Based on the same inventive concept, this invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the computer device and extended storage media supported by the computer device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, this storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the steps of the cloud-edge collaborative power distribution model application service method in the above embodiments.

[0104] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied 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.

[0105] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0106] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0107] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0108] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A cloud-edge collaborative power distribution model application service method, characterized in that, The method includes: The single physical quantity collected by the edge device is used as the input of the pre-trained edge device machine learning model to obtain the judgment result of the edge device inspection object output by the pre-trained edge device machine learning model. The system collects the single physical quantities collected by each end-side device within the region and the judgment results of the objects monitored by each end-side device, and uses them as input to a pre-trained side-side device machine learning model to obtain the region judgment result output by the pre-trained side-side device machine learning model. The judgment results from each region are collected, and the judgment results from each region and the distribution network characteristic information are used as input to a pre-trained cloud machine learning model to obtain the distribution system judgment result output by the pre-trained cloud machine learning model. The judgment results include at least one of the following: system optimization decision results, health evaluation results, user profile tags, defect identification results, energy efficiency prediction results, and control command results supporting power supply operation.

2. The method as described in claim 1, characterized in that, The end-side devices include: smart switches, smart meters, and end-side sensors.

3. The method as described in claim 1, characterized in that, The single physical quantity includes: video signal, sound signal, power consumption data, temperature signal, vibration signal, humidity signal, or pressure signal.

4. The method as described in claim 1, characterized in that, The training process of the pre-trained edge device machine learning model includes: Training data is constructed using historical data of a single physical quantity collected by edge devices and their corresponding judgment results; The machine learning model of the edge device is trained using the training data to obtain the pre-trained machine learning model of the edge device.

5. The method as described in claim 1, characterized in that, The training process of the pre-trained side-device machine learning model includes: Training data is constructed by using historical data of single physical quantities collected by each end-side device in the region, historical data of judgment results of the objects monitored by each end-side device, and their corresponding regional judgment results; The edge device machine learning model is trained using the training data to obtain the pre-trained edge device machine learning model.

6. The method as described in claim 1, characterized in that, The power distribution network characteristic information includes at least one of the following: power distribution network ledger data and meteorological data.

7. The method as described in claim 1, characterized in that, The edge devices include: edge computing terminals and smart gateways.

8. The method as described in claim 1, characterized in that, The training process of the pre-trained cloud machine learning model includes: Training data is constructed using historical data of judgment results in each region, historical data of distribution network characteristic information, and their corresponding judgment results in the distribution system. The cloud-based machine learning model is trained using the training data to obtain the pre-trained cloud-based machine learning model.

9. A cloud-edge collaborative power distribution model application service device, characterized in that, The device includes: The first analysis module is used to take a single physical quantity collected by the end-side device as input to a pre-trained end-side device machine learning model and obtain the judgment result of the end-side device inspection object output by the pre-trained end-side device machine learning model. The second analysis module is used to collect the single physical quantities collected by each end-side device in the area and the judgment results of the objects monitored by each end-side device, and use them as input to the pre-trained side-side device machine learning model to obtain the area judgment results output by the pre-trained side-side device machine learning model. The third analysis module is used to collect the judgment results of each region and use the judgment results of each region and the distribution network characteristic information as input to a pre-trained cloud machine learning model to obtain the distribution system judgment result output by the pre-trained cloud machine learning model. The judgment results include at least one of the following: system optimization decision results, health evaluation results, user profile tags, defect identification results, energy efficiency prediction results, and control command results supporting power supply operation.

10. The apparatus as claimed in claim 9, characterized in that, The end-side devices include: smart switches, smart meters, and end-side sensors.

11. The apparatus as claimed in claim 9, characterized in that, The single physical quantity includes: video signal, sound signal, power consumption data, temperature signal, vibration signal, humidity signal, or pressure signal.

12. The apparatus as claimed in claim 9, characterized in that, The training process of the pre-trained edge device machine learning model includes: Training data is constructed using historical data of a single physical quantity collected by edge devices and their corresponding judgment results; The machine learning model of the edge device is trained using the training data to obtain the pre-trained machine learning model of the edge device.

13. The apparatus as claimed in claim 9, characterized in that, The training process of the pre-trained side-device machine learning model includes: Training data is constructed by using historical data of single physical quantities collected by each end-side device in the region, historical data of judgment results of the objects monitored by each end-side device, and their corresponding regional judgment results; The edge device machine learning model is trained using the training data to obtain the pre-trained edge device machine learning model.

14. The apparatus as claimed in claim 9, characterized in that, The power distribution network characteristic information includes at least one of the following: power distribution network ledger data and meteorological data.

15. The apparatus as claimed in claim 9, characterized in that, The edge devices include: edge computing terminals and smart gateways.

16. The apparatus as claimed in claim 9, characterized in that, The training process of the pre-trained cloud machine learning model includes: Training data is constructed using historical data of judgment results in each region, historical data of distribution network characteristic information, and their corresponding judgment results in the distribution system. The cloud-based machine learning model is trained using the training data to obtain the pre-trained cloud-based machine learning model.

17. A computer device, characterized in that, include: One or more processors; The processor is used to execute one or more programs; When the one or more programs are executed by the one or more processors, the cloud-edge collaborative power distribution model application service method as described in any one of claims 1 to 8 is implemented.

18. A computer-readable storage medium, characterized in that, It contains a computer program, which, when executed, implements the cloud-edge collaborative power distribution model application service method as described in any one of claims 1 to 8.

Images