Photovoltaic inspection management and control method and system based on digital twinning

Abstract

The invention relates to photovoltaic inspection management and control, in particular to a photovoltaic inspection management and control method and system based on digital twinning, and the method comprises the steps: collecting an orthoimage collected by an unmanned aerial vehicle for a photovoltaic power station, and marking each equipment unit and a corresponding equipment name number in the orthoimage; extracting the topographic three-dimensional geographic information of the photovoltaic power station, the equipment three-dimensional geographic information of each equipment unit and the equipment name number from the orthoimage; creating a power station simulation model of the photovoltaic power station, an equipment simulation model of each equipment unit and a hierarchical management relationship among the equipment simulation models, and obtaining a digital twin simulation model of the photovoltaic power station; performing fault prediction on the photovoltaic power station by using a machine learning algorithm, and arranging / adjusting a photovoltaic inspection task in real time according to a fault prediction result; according to the technical scheme provided by the invention, the defects that a digital twinborn simulation model of a photovoltaic power station is difficult to accurately construct and intelligent arrangement and adjustment of a photovoltaic inspection task cannot be realized in the prior art can be effectively overcome.

Description

Technical Field

[0001] This invention relates to photovoltaic inspection and control, specifically to a photovoltaic inspection and control method and system based on digital twins. Background Technology

[0002] With the growth of global energy demand and the increasing emphasis on clean energy, photovoltaic power plants, as important renewable energy power generation facilities, are constantly expanding in number and scale. Effective inspection and management are crucial to ensuring the stable and efficient operation of photovoltaic power plants.

[0003] Traditional photovoltaic inspection methods mainly rely on regular manual inspections. This method not only consumes a lot of manpower and resources, but also has low inspection efficiency, making it difficult to discover potential equipment unit failures and to deal with emergencies in a timely manner, which seriously affects the power generation efficiency and economic benefits of photovoltaic power plants.

[0004] In recent years, drone technology has been gradually applied to the field of photovoltaic inspection. By collecting images of photovoltaic power plants through drones, the coverage and efficiency of inspections have been improved to a certain extent. However, relying solely on image data is insufficient to comprehensively and intuitively reflect the actual operating status and geographical location information of photovoltaic power plant equipment units. The ability to accurately locate and predict faults is limited, making it difficult to meet the technical requirements of refined inspection and management of large-scale photovoltaic power plants.

[0005] Meanwhile, digital twins, as an emerging technology, enable real-time monitoring, simulation, and analysis of physical equipment by constructing virtual mapping models of physical entities. However, research on applying digital twin technology to the field of photovoltaic inspection and management is still in its early stages. A complete and effective method and system for photovoltaic inspection and management based on digital twins has not yet been formed to achieve accurate prediction and visualization of photovoltaic power station faults, as well as intelligent scheduling and adjustment of photovoltaic inspection tasks. Summary of the Invention

[0006] (a) Technical problems to be solved

[0007] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a photovoltaic inspection and control method and system based on digital twin, which can effectively overcome the shortcomings of the existing technology, such as the difficulty in accurately constructing a digital twin simulation model of a photovoltaic power station and the inability to realize intelligent arrangement and adjustment of photovoltaic inspection tasks.

[0008] (II) Technical Solution

[0009] To achieve the above objectives, the present invention provides the following technical solution:

[0010] A photovoltaic inspection and control method based on digital twins includes the following steps:

[0011] S1. Collect orthophotos of the photovoltaic power station taken by the drone and mark each equipment unit and its corresponding equipment name and number in the orthophotos;

[0012] S2. Extract the three-dimensional geographic information of the terrain of the photovoltaic power station, the three-dimensional geographic information of the equipment of each equipment unit, and the equipment name and number from the orthophoto;

[0013] S3. Create a power plant simulation model, equipment simulation models of each equipment unit, and hierarchical management relationships between the equipment simulation models to obtain a digital twin simulation model of the photovoltaic power plant.

[0014] S4. Utilize machine learning algorithms to predict faults in photovoltaic power plants, display the fault prediction results in a digital twin simulation model of the photovoltaic power plant for visualization, and arrange / adjust photovoltaic inspection tasks in real time based on the fault prediction results.

[0015] Preferably, in S1, marking each device unit in the orthophoto image includes:

[0016] In the orthophoto, mark the corner points of each photovoltaic string and the center points of other equipment units;

[0017] Other equipment units include transformers, inverters, and hub circuits.

[0018] Preferably, in S2, the extraction of the three-dimensional geographic information of each equipment unit from the orthophoto image includes:

[0019] Extract the geographic information of each corner point of each photovoltaic string from the orthophoto; extract the geographic information of the center point of other equipment units from the orthophoto.

[0020] The geographic information of the string corner points includes the longitude and latitude of the corner points, and the geographic information of the equipment center includes the longitude, latitude, and altitude of the center location point.

[0021] Preferably, the simulation model of the photovoltaic power station and the equipment simulation models of each equipment unit created in S3 include:

[0022] A simulation model of the photovoltaic power station is created based on the three-dimensional geographic information of the terrain.

[0023] Based on the geographical information of the corner points of each photovoltaic string, calculate the geographical information of the string center of each photovoltaic string;

[0024] Based on the geographic information of the corner points and center of each photovoltaic string, as well as the geographic information of the center of other equipment units, an equipment simulation model for each equipment unit is created in the power plant simulation model of the photovoltaic power plant.

[0025] Preferably, in S3, a hierarchical management relationship is created between the simulation models of each device to obtain a digital twin simulation model of the photovoltaic power station, including:

[0026] Construct POI data for the equipment simulation models of each equipment unit;

[0027] Based on the equipment name and number of each equipment unit, determine the hierarchical management relationship between the equipment simulation models of each equipment unit;

[0028] Based on the hierarchical management relationship, the POI data of the equipment simulation models of each equipment unit are displayed in the equipment simulation model of the equipment unit, so as to obtain the digital twin simulation model of the photovoltaic power station.

[0029] Preferably, after creating the equipment simulation model of each equipment unit in the power plant simulation model of the photovoltaic power plant, the process includes:

[0030] Multiple Level of Detail (LOD) models of different levels are configured for the equipment simulation models of each equipment unit;

[0031] When displaying a digital twin simulation scene using a digital twin simulation model of a photovoltaic power station, the corresponding level of LOD model is selected for display according to the different display distances of each equipment simulation model in the display interface, and each equipment simulation model in the display interface is rendered according to the corresponding rendering parameters.

[0032] The closer the equipment simulation model is displayed in the display interface, the higher the corresponding LOD model level.

[0033] Preferably, the step of selecting and displaying LOD models of corresponding levels according to the different display distances of each device simulation model in the display interface, and rendering each device simulation model in the display interface according to the corresponding rendering parameters, includes:

[0034] The rendering priority of each device simulation model is determined based on the LOD level displayed for each device simulation model in the display interface.

[0035] The simulation models of each device in the display interface are rendered according to their respective rendering parameters based on their rendering priority.

[0036] Among them, the higher the LOD level of the equipment simulation model, the higher the rendering priority.

[0037] Preferably, in step S4, machine learning algorithms are used to predict faults in the photovoltaic power station, and the fault prediction results are displayed in a digital twin simulation model of the photovoltaic power station for visualization. Photovoltaic inspection tasks are also scheduled / adjusted in real time based on the fault prediction results, including:

[0038] Based on the environmental data in the digital twin simulation model of the photovoltaic power station, and the data characteristics and hierarchical management relationships of the equipment simulation models of each equipment unit in the digital twin simulation model of the photovoltaic power station, a prediction algorithm is matched in the machine learning algorithm library.

[0039] The matching prediction algorithm is used to perform fault prediction on the working data of the equipment simulation model of each equipment unit in the digital twin simulation model of the photovoltaic power station, and the fault prediction results are obtained.

[0040] The fault prediction results are displayed in the digital twin simulation model of the photovoltaic power station for visualization, and photovoltaic inspection tasks are arranged / adjusted in real time based on the fault prediction results.

[0041] Preferably, the step of visually displaying the fault prediction results in the digital twin simulation model of the photovoltaic power station and scheduling / adjusting photovoltaic inspection tasks in real time based on the fault prediction results includes:

[0042] Machine learning algorithms are used regularly to predict faults in photovoltaic power plants. In the digital twin simulation model of the photovoltaic power plant, different colors are used to distinguish the equipment units that have changed from the fault prediction results of the previous cycle.

[0043] When generating fault prediction reports, the equipment units with continuously changing colors are placed at the top of the report, and photovoltaic inspection tasks are arranged / adjusted in real time based on the fault prediction reports.

[0044] A photovoltaic inspection and control system based on digital twins is used to execute the aforementioned photovoltaic inspection and control method based on digital twins, including an image collection module, an equipment unit marking module, an information extraction module, a digital twin simulation model creation module, and a fault prediction module;

[0045] Image collection module, which collects orthophotos of photovoltaic power plants taken by drones;

[0046] The equipment unit marking module marks each equipment unit and its corresponding equipment name and number in the orthophoto;

[0047] The information extraction module extracts the three-dimensional geographic information of the photovoltaic power station's terrain, the three-dimensional geographic information of each equipment unit, and the equipment name and number from the orthophoto.

[0048] The digital twin simulation model creation module creates a power plant simulation model, equipment simulation models of each equipment unit, and hierarchical management relationships between the equipment simulation models to obtain a digital twin simulation model of the photovoltaic power plant.

[0049] The fault prediction module uses machine learning algorithms to predict faults in photovoltaic power plants, displays the fault prediction results in a digital twin simulation model of the photovoltaic power plant, and schedules / adjusts photovoltaic inspection tasks in real time based on the fault prediction results.

[0050] (III) Beneficial Effects

[0051] Compared with existing technologies, the photovoltaic inspection and control method and system based on digital twins provided by this invention have the following beneficial effects:

[0052] 1) Collect orthophotos of photovoltaic power plants taken by drones, and mark each equipment unit and its corresponding equipment name and number in the orthophotos. Extract the 3D geographic information of the photovoltaic power plant terrain, the 3D geographic information of each equipment unit and the equipment name and number from the orthophotos. Create a power plant simulation model, an equipment simulation model of each equipment unit, and the hierarchical management relationship between the equipment simulation models to obtain a digital twin simulation model of the photovoltaic power plant. By extracting the 3D geographic information of the photovoltaic power plant terrain terrain, the 3D geographic information of each equipment unit and the equipment name and number from the orthophotos, the digital twin simulation model of the photovoltaic power plant can be constructed. It can comprehensively and intuitively reflect the actual operating status and geographical location information of the photovoltaic power plant equipment units, which helps to improve the accurate location and prediction of faults and fully meet the technical needs of refined inspection and control of large-scale photovoltaic power plants.

[0053] 2) Utilize machine learning algorithms to predict faults in photovoltaic power plants, and display the fault prediction results in a digital twin simulation model of the photovoltaic power plant for visualization. Based on the fault prediction results, photovoltaic inspection tasks can be arranged / adjusted in real time. This enables accurate prediction and visualization of photovoltaic power plant faults, as well as intelligent arrangement and adjustment of photovoltaic inspection tasks, helping operation and maintenance personnel to better manage photovoltaic inspections and effectively improve the power generation efficiency and economic benefits of photovoltaic power plants. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0055] Figure 1 This is a schematic diagram of the process of the present invention;

[0056] Figure 2 This is a flowchart illustrating the process of obtaining a digital twin simulation model of a photovoltaic power station in this invention.

[0057] Figure 3 This is a flowchart illustrating the process of visualizing faults and intelligently scheduling and adjusting photovoltaic inspection tasks in this invention. Detailed Implementation

[0058] 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, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0059] A photovoltaic inspection and control method based on digital twins, such as Figure 1 and Figure 2 As shown, S1 collects orthophotos of the photovoltaic power station taken by the drone and marks each equipment unit and its corresponding equipment name and number in the orthophotos.

[0060] Specifically, each equipment unit is marked in the orthophoto, including:

[0061] In the orthophoto, mark the corner points of each photovoltaic string and the center points of other equipment units;

[0062] Other equipment units include transformers, inverters, and hub circuits.

[0063] S2. Extract the 3D topographic geographic information of the photovoltaic power station, the 3D geographic information of each equipment unit, and the equipment name and number from the orthophoto.

[0064] Specifically, the three-dimensional geographic information of each equipment unit is extracted from the orthophoto, including:

[0065] Extract the geographic information of each corner point of each photovoltaic string from the orthophoto; extract the geographic information of the center point of other equipment units from the orthophoto.

[0066] The geographic information of the string corner points includes the longitude and latitude of the corner points, and the geographic information of the equipment center includes the longitude, latitude, and altitude of the center location point.

[0067] S3. Create a power plant simulation model, equipment simulation models of each equipment unit, and hierarchical management relationships between the equipment simulation models to obtain a digital twin simulation model of the photovoltaic power plant.

[0068] 1) Create a power plant simulation model and equipment simulation models for each equipment unit of the photovoltaic power plant, including:

[0069] A simulation model of the photovoltaic power station is created based on the three-dimensional geographic information of the terrain.

[0070] Based on the geographical information of the corner points of each photovoltaic string, calculate the geographical information of the string center of each photovoltaic string;

[0071] Based on the geographic information of the corner points and center of each photovoltaic string, as well as the geographic information of the center of other equipment units, an equipment simulation model for each equipment unit is created in the power plant simulation model of the photovoltaic power plant.

[0072] 2) Create a hierarchical management relationship between the simulation models of each device to obtain a digital twin simulation model of the photovoltaic power station, including:

[0073] Construct POI data for the equipment simulation models of each equipment unit;

[0074] Based on the equipment name and number of each equipment unit, determine the hierarchical management relationship between the equipment simulation models of each equipment unit;

[0075] Based on the hierarchical management relationship, the POI data of the equipment simulation models of each equipment unit are displayed in the equipment simulation model of the equipment unit, so as to obtain the digital twin simulation model of the photovoltaic power station.

[0076] Specifically, after creating the equipment simulation models for each equipment unit in the power plant simulation model, the process includes:

[0077] Multiple Level of Detail (LOD) models of different levels are configured for the equipment simulation models of each equipment unit;

[0078] When displaying a digital twin simulation scene using a digital twin simulation model of a photovoltaic power station, the corresponding level of LOD model is selected for display according to the different display distances of each equipment simulation model in the display interface, and each equipment simulation model in the display interface is rendered according to the corresponding rendering parameters.

[0079] The closer the equipment simulation model is displayed in the display interface, the higher the corresponding LOD model level.

[0080] Specifically, based on the different display distances of each device simulation model in the display interface, the corresponding LOD model is selected for display, and each device simulation model in the display interface is rendered according to the corresponding rendering parameters, including:

[0081] The rendering priority of each device simulation model is determined based on the LOD level displayed for each device simulation model in the display interface.

[0082] The simulation models of each device in the display interface are rendered according to their respective rendering parameters based on their rendering priority.

[0083] Among them, the higher the LOD level of the equipment simulation model, the higher the rendering priority.

[0084] The above technical solution collects orthophotos of photovoltaic power plants taken by drones, marks each equipment unit and its corresponding equipment name and number in the orthophotos, extracts the 3D geographic information of the photovoltaic power plant's terrain, the 3D geographic information of each equipment unit, and the equipment name and number from the orthophotos, creates a power plant simulation model, equipment simulation models of each equipment unit, and hierarchical management relationships between the equipment simulation models, and obtains a digital twin simulation model of the photovoltaic power plant. By extracting the 3D geographic information of the photovoltaic power plant's terrain, the 3D geographic information of each equipment unit, and the equipment name and number from the orthophotos to construct the digital twin simulation model of the photovoltaic power plant, it can comprehensively and intuitively reflect the actual operating status and geographical location information of the photovoltaic power plant's equipment units, which helps to improve the accurate location and prediction capabilities of faults and fully meets the technical needs of refined inspection and management of large-scale photovoltaic power plants.

[0085] like Figure 1 and Figure 3 As shown in Figure S4, machine learning algorithms are used to predict faults in photovoltaic power plants. The fault prediction results are then visualized in the digital twin simulation model of the photovoltaic power plant. Based on the fault prediction results, photovoltaic inspection tasks are scheduled / adjusted in real time. Specifically, this includes:

[0086] Based on the environmental data in the digital twin simulation model of the photovoltaic power station, and the data characteristics and hierarchical management relationships of the equipment simulation models of each equipment unit in the digital twin simulation model of the photovoltaic power station, a prediction algorithm is matched in the machine learning algorithm library.

[0087] The matching prediction algorithm is used to perform fault prediction on the working data of the equipment simulation model of each equipment unit in the digital twin simulation model of the photovoltaic power station, and the fault prediction results are obtained.

[0088] The fault prediction results are displayed in the digital twin simulation model of the photovoltaic power station for visualization, and photovoltaic inspection tasks are arranged / adjusted in real time based on the fault prediction results.

[0089] Specifically, the fault prediction results are visualized in the digital twin simulation model of the photovoltaic power station, and photovoltaic inspection tasks are scheduled / adjusted in real time based on the fault prediction results, including:

[0090] Machine learning algorithms are used regularly to predict faults in photovoltaic power plants. In the digital twin simulation model of the photovoltaic power plant, different colors are used to distinguish the equipment units that have changed from the fault prediction results of the previous cycle.

[0091] When generating fault prediction reports, the equipment units with continuously changing colors are placed at the top of the report, and photovoltaic inspection tasks are arranged / adjusted in real time based on the fault prediction reports.

[0092] The above technical solution utilizes machine learning algorithms to predict faults in photovoltaic power plants, displays the fault prediction results in a digital twin simulation model of the photovoltaic power plant, and schedules / adjusts photovoltaic inspection tasks in real time based on the fault prediction results. This enables accurate prediction and visualization of photovoltaic power plant faults, as well as intelligent scheduling and adjustment of photovoltaic inspection tasks, helping operation and maintenance personnel to better manage photovoltaic inspections and effectively improve the power generation efficiency and economic benefits of photovoltaic power plants.

[0093] In the technical solution of this application, based on the above-mentioned photovoltaic inspection and control method based on digital twin, a photovoltaic inspection and control system based on digital twin is also disclosed, including an image collection module, an equipment unit marking module, an information extraction module, a digital twin simulation model creation module, and a fault prediction module.

[0094] Image collection module, which collects orthophotos of photovoltaic power plants taken by drones;

[0095] The equipment unit marking module marks each equipment unit and its corresponding equipment name and number in the orthophoto;

[0096] The information extraction module extracts the three-dimensional geographic information of the photovoltaic power station's terrain, the three-dimensional geographic information of each equipment unit, and the equipment name and number from the orthophoto.

[0097] The digital twin simulation model creation module creates a power plant simulation model, equipment simulation models of each equipment unit, and hierarchical management relationships between the equipment simulation models to obtain a digital twin simulation model of the photovoltaic power plant.

[0098] The fault prediction module uses machine learning algorithms to predict faults in photovoltaic power plants, displays the fault prediction results in a digital twin simulation model of the photovoltaic power plant, and schedules / adjusts photovoltaic inspection tasks in real time based on the fault prediction results.

[0099] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A photovoltaic inspection and control method based on digital twins, characterized in that: Includes the following steps: S1. Collect orthophotos of the photovoltaic power station taken by the drone and mark each equipment unit and its corresponding equipment name and number in the orthophotos; S2. Extract the three-dimensional geographic information of the terrain of the photovoltaic power station, the three-dimensional geographic information of the equipment of each equipment unit, and the equipment name and number from the orthophoto; S3. Create a power plant simulation model, equipment simulation models of each equipment unit, and hierarchical management relationships between the equipment simulation models to obtain a digital twin simulation model of the photovoltaic power plant. S4. Utilize machine learning algorithms to predict faults in photovoltaic power plants, display the fault prediction results in a digital twin simulation model of the photovoltaic power plant for visualization, and arrange / adjust photovoltaic inspection tasks in real time based on the fault prediction results.

2. The photovoltaic inspection and control method based on digital twins according to claim 1, characterized in that: In S1, each device unit is marked in the orthophoto, including: In the orthophoto, mark the corner points of each photovoltaic string and the center points of other equipment units; Other equipment units include transformers, inverters, and hub circuits.

3. The photovoltaic inspection and control method based on digital twins according to claim 2, characterized in that: S2 extracts the 3D geographic information of each equipment unit from the orthophoto, including: Extract the geographic information of each corner point of each photovoltaic string from the orthophoto; extract the geographic information of the center point of other equipment units from the orthophoto. The geographic information of the string corner points includes the longitude and latitude of the corner points, and the geographic information of the equipment center includes the longitude, latitude, and altitude of the center location point.

4. The photovoltaic inspection and control method based on digital twins according to claim 3, characterized in that: In S3, a power plant simulation model and equipment simulation models for each equipment unit are created, including: A simulation model of the photovoltaic power station is created based on the three-dimensional geographic information of the terrain. Based on the geographical information of the corner points of each photovoltaic string, calculate the geographical information of the string center of each photovoltaic string; Based on the geographic information of the corner points and center of each photovoltaic string, as well as the geographic information of the center of other equipment units, an equipment simulation model for each equipment unit is created in the power plant simulation model of the photovoltaic power plant.

5. The photovoltaic inspection and control method based on digital twins according to claim 4, characterized in that: In S3, a hierarchical management relationship is created between the simulation models of various devices to obtain a digital twin simulation model of the photovoltaic power station, including: Construct POI data for the equipment simulation models of each equipment unit; Based on the equipment name and number of each equipment unit, determine the hierarchical management relationship between the equipment simulation models of each equipment unit; Based on the hierarchical management relationship, the POI data of the equipment simulation models of each equipment unit are displayed in the equipment simulation model of the equipment unit, so as to obtain the digital twin simulation model of the photovoltaic power station.

6. The photovoltaic inspection and control method based on digital twins according to claim 5, characterized in that: After creating the equipment simulation models of each equipment unit in the power plant simulation model of the photovoltaic power plant, the process includes: Multiple Level of Detail (LOD) models of different levels are configured for the equipment simulation models of each equipment unit; When displaying a digital twin simulation scene using a digital twin simulation model of a photovoltaic power station, the corresponding level of LOD model is selected for display according to the different display distances of each equipment simulation model in the display interface, and each equipment simulation model in the display interface is rendered according to the corresponding rendering parameters. The closer the equipment simulation model is displayed in the display interface, the higher the corresponding LOD model level.

7. The photovoltaic inspection and control method based on digital twins according to claim 6, characterized in that: The process involves selecting and displaying LOD models of corresponding levels based on the different display distances of each device simulation model in the display interface, and rendering each device simulation model in the display interface according to the corresponding rendering parameters, including: The rendering priority of each device simulation model is determined based on the LOD level displayed for each device simulation model in the display interface. The simulation models of each device in the display interface are rendered according to their respective rendering parameters based on their rendering priority. Among them, the higher the LOD level of the equipment simulation model, the higher the rendering priority.

8. The photovoltaic inspection and control method based on digital twin according to claim 1, characterized in that: S4 utilizes machine learning algorithms to predict faults in photovoltaic power plants. The prediction results are then visualized in a digital twin simulation model of the power plant. Furthermore, photovoltaic inspection tasks are scheduled and adjusted in real-time based on the prediction results, including: Based on the environmental data in the digital twin simulation model of the photovoltaic power station, and the data characteristics and hierarchical management relationships of the equipment simulation models of each equipment unit in the digital twin simulation model of the photovoltaic power station, a prediction algorithm is matched in the machine learning algorithm library. The matching prediction algorithm is used to perform fault prediction on the working data of the equipment simulation model of each equipment unit in the digital twin simulation model of the photovoltaic power station, and the fault prediction results are obtained. The fault prediction results are displayed in the digital twin simulation model of the photovoltaic power station for visualization, and photovoltaic inspection tasks are arranged / adjusted in real time based on the fault prediction results.

9. The photovoltaic inspection and control method based on digital twins according to claim 8, characterized in that: The process of visually displaying the fault prediction results in the digital twin simulation model of the photovoltaic power station and scheduling / adjusting photovoltaic inspection tasks in real time based on the fault prediction results includes: Machine learning algorithms are used regularly to predict faults in photovoltaic power plants. In the digital twin simulation model of the photovoltaic power plant, different colors are used to distinguish the equipment units that have changed from the fault prediction results of the previous cycle. When generating fault prediction reports, the equipment units with continuously changing colors are placed at the top of the report, and photovoltaic inspection tasks are arranged / adjusted in real time based on the fault prediction reports.

10. A photovoltaic inspection and control system based on digital twins, used to execute the photovoltaic inspection and control method based on digital twins as described in claim 1, characterized in that: It includes an image acquisition module, an equipment unit marking module, an information extraction module, a digital twin simulation model creation module, and a fault prediction module; Image collection module, which collects orthophotos of photovoltaic power plants taken by drones; The equipment unit marking module marks each equipment unit and its corresponding equipment name and number in the orthophoto; The information extraction module extracts the three-dimensional geographic information of the photovoltaic power station's terrain, the three-dimensional geographic information of each equipment unit, and the equipment name and number from the orthophoto. The digital twin simulation model creation module creates a power plant simulation model, equipment simulation models of each equipment unit, and hierarchical management relationships between the equipment simulation models to obtain a digital twin simulation model of the photovoltaic power plant. The fault prediction module uses machine learning algorithms to predict faults in photovoltaic power plants, displays the fault prediction results in a digital twin simulation model of the photovoltaic power plant, and schedules / adjusts photovoltaic inspection tasks in real time based on the fault prediction results.

Images