Photovoltaic power station fault diagnosis method and system based on industrial vision

Abstract

The invention discloses a photovoltaic power station fault diagnosis method and system based on industrial vision, and belongs to the technical field of photovoltaic power station fault detection. The system comprises a power generation data acquisition module, an image data acquisition module, an intelligent analysis module, a monitoring module, a fault diagnosis module and an alarm module. The power generation data acquisition module is used for acquiring power generation data of different photovoltaic modules; the image data acquisition module is used for acquiring image data in the power generation process of a plurality of photovoltaic modules and determining attenuation factors influencing the generating capacity of different module units in the photovoltaic modules; the intelligent analysis module is used for establishing a generating capacity analysis model and analyzing the influence of different attenuation factors on the generating capacity of the component unit; the monitoring module is used for monitoring the total generating capacity of different photovoltaic modules and judging whether the photovoltaic module with the fault risk exists or not; the fault diagnosis module is used for judging the fault type of the photovoltaic module; and the alarm module is used for carrying out safety alarm.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power station fault detection, and in particular to a photovoltaic power station fault diagnosis method and system based on industrial vision. Background Art

[0002] With the continuous increase in energy demand, photovoltaic power stations, as an important renewable energy power generation facility, whether they can operate efficiently and stably directly affects the supply efficiency and stability of the power grid. Therefore, fault diagnosis of photovoltaic power stations is crucial. In recent years, with the rapid development of industrial vision technology, fault diagnosis methods based on image recognition and analysis have been widely used in photovoltaic power station fault diagnosis. In existing technologies, high-definition cameras and infrared thermal imagers carried by drones can quickly collect image information of photovoltaic components. Using advanced image processing algorithms and deep learning models, faults can be effectively and automatically identified and located, providing strong support for the intelligent operation and maintenance of photovoltaic power stations.

[0003] Fault diagnosis methods based on image recognition and analysis have the advantages of high efficiency and automation. However, the photovoltaic power station environment is complex and changeable. Some complex fault types are difficult to accurately identify using a single image recognition technology, resulting in low accuracy in judging photovoltaic module faults. In addition, the processing of large amounts of image data not only increases the maintenance cost and difficulty of the system, but also easily leads to a waste of regulatory resources. Summary of the Invention

[0004] The purpose of the present invention is to provide a photovoltaic power station fault diagnosis method and system based on industrial vision to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solution: a substation fault early warning method based on artificial intelligence, the method comprising the following steps: Step S10: generating electricity through a plurality of photovoltaic modules in a photovoltaic power station, and collecting power generation data of different photovoltaic modules; each photovoltaic module includes a plurality of module units; and the power generation data includes power generation of different module units in the photovoltaic module; Step S20: Perform regular inspections of the photovoltaic power station using a drone to collect image data of a plurality of photovoltaic modules during power generation, analyze the collected image data, and determine attenuation factors that affect power generation of different module units in the photovoltaic module; Step S30: storing the power generation data collected in step S10 and the image data collected in step S20 as historical data in a database, establishing a power generation analysis model based on the historical data in the database, and analyzing the impact of different attenuation factors on the power generation of the module units; Step S40: Based on the power generation data of different photovoltaic modules, the total power generation of the different photovoltaic modules is monitored to determine whether there is a photovoltaic module at risk of failure; if there is a photovoltaic module at risk of failure, step S50 is executed; if there is no photovoltaic module at risk of failure, step S60 is executed; Step S50: Determine the location information of the photovoltaic module at risk of failure, dispatch a drone to capture real-time image data, collect real-time power generation data of the photovoltaic module at risk of failure, determine the type of photovoltaic module failure based on the captured real-time image data, the collected real-time power generation data, and the impact of different attenuation factors on the power generation of the module unit, issue a safety alarm, and send the alarm signal and the determined photovoltaic module failure type to the management personnel; Step S60: The photovoltaic power station is safe, and steps S10 to S60 are repeated.

[0006] A photovoltaic power station fault diagnosis system based on industrial vision, which includes a power generation data acquisition module, an image data acquisition module, an intelligent analysis module, a monitoring module, a fault diagnosis module and an alarm module; The power generation data acquisition module is used to collect power generation data of different photovoltaic modules; each photovoltaic module includes a plurality of module units; the power generation data includes the power generation of different module units in the photovoltaic module; and the collected power generation data is sent to the intelligent analysis module, the monitoring module and the fault diagnosis module; The image data acquisition module is used to collect image data of a plurality of photovoltaic modules during power generation, analyze the collected image data, determine the attenuation factors that affect the power generation of different component units in the photovoltaic module, and send the collected image data to the intelligent analysis module and the fault diagnosis module; The intelligent analysis module is used to establish a database, store the power generation data sent by the power generation data acquisition module and the image data sent by the image data acquisition module as historical data in the database; establish a power generation analysis model to analyze the impact of different attenuation factors on the power generation of the component units; and send the impact of different attenuation factors on the power generation of the component units to the fault diagnosis module; The monitoring module is used to monitor the total power generation of different photovoltaic modules based on the power generation data sent by the power generation data acquisition module, and determine whether there are photovoltaic modules with a risk of failure; if there are photovoltaic modules with a risk of failure, the location information of the photovoltaic modules with a risk of failure is sent to the fault diagnosis module; if there are no photovoltaic modules with a risk of failure, the monitoring is continued; The fault diagnosis module is used to determine the location information of the photovoltaic component with a risk of failure, send a signal to the image data acquisition module, dispatch a drone to capture real-time image data, determine the real-time power generation data of the photovoltaic component with a risk of failure sent by the power generation data acquisition module, determine the type of photovoltaic component failure based on the real-time image data, real-time power generation data and the impact of different attenuation factors on the power generation of the component unit, and send the determined photovoltaic component failure type to the alarm module; The alarm module is used to generate a safety alarm and send the alarm signal and the determined photovoltaic component fault type to the management personnel.

[0007] Compared with the existing technology, the beneficial effects achieved by the present invention are: through the established power generation analysis model, the total power generation of photovoltaic modules is predicted, thereby improving the accuracy of photovoltaic power station fault judgment; preliminary monitoring of photovoltaic module failures is carried out to identify photovoltaic modules with failure risks, and image data of photovoltaic modules with failure risks are collected by drones for re-analysis, avoiding the singleness brought about by photovoltaic power station fault diagnosis based on image recognition and analysis, and effectively improving the accuracy of photovoltaic module fault identification through diversified judgment methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 This is a schematic diagram of the steps of a photovoltaic power station fault diagnosis method based on industrial vision of the present invention.

[0009] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0010] See also Figure 1 , the present invention provides a technical solution: See also Figure 1 In the first embodiment, a photovoltaic power station fault diagnosis method based on industrial vision is provided, which includes the following steps: Step S10: Generate electricity through a plurality of photovoltaic modules in a photovoltaic power station, and collect power generation data of different photovoltaic modules respectively; the photovoltaic modules respectively include a plurality of module units; the power generation data includes the power generation of different module units in the photovoltaic modules.

[0011] Step S20: Perform regular inspections of the photovoltaic power station using a drone to collect image data of several photovoltaic modules during power generation, analyze the collected image data, and determine the attenuation factors that affect the power generation of different module units in the photovoltaic module.

[0012] It should be noted that the size specifications of each photovoltaic module in the photovoltaic power station are the same, and the size and number of module units in each photovoltaic module are the same; the attenuation factor represents the different defect types that affect the power generation of the module unit and the coverage area under the corresponding defect type; the photovoltaic module failure types are divided into internal failures and external failures; the internal failure represents the interference factors that cannot be directly determined from the image data taken by the drone and will affect the power generation of the module unit; the external failure represents the interference factors that can be directly determined from the image data taken by the drone and will affect the power generation of the module unit; the internal failure includes several defects, where each defect corresponds to an interference factor that will affect the power generation of the module unit.

[0013] In this embodiment, the defects are divided into a first interference factor and a second interference factor; the first interference factor includes but is not limited to cracks, scratches and wear in the component unit, indicating the loss of the component unit itself; the second interference factor includes but is not limited to shadows, dust and foreign objects, indicating the loss of non-component units themselves; when analyzing the image data, the different defects in the image are identified by an image processing algorithm, and the coverage area corresponding to each defect is determined, so as to obtain the attenuation factor of the power generation of different component units in the photovoltaic module.

[0014] Step S30: Store the power generation data collected in step S10 and the image data collected in step S20 as historical data in a database. Based on the historical data in the database, establish a power generation analysis model to analyze the impact of different attenuation factors on the power generation of the component units.

[0015] Specifically, the method steps are: Step S31: retrieve historical power generation data and historical image data from the database for analysis, and obtain the power generation of different component units in each photovoltaic module over time based on the historical power generation data. The total set of changes ; Based on the historical image data, the attenuation factor affecting the power generation of different component units in each photovoltaic module is obtained over time The total set of changes ; in, ; Represents the power generation of different component units in each photovoltaic module over time The set of changes; ; Indicates the The power generation of different component units in a photovoltaic module changes over time The set of changes; Respectively represent The power generation of different component units in a photovoltaic module changes over time changes; Indicates the number of photovoltaic modules in a photovoltaic power station; Indicates the number of module units in a photovoltaic module; ; Represents the attenuation factor of the power generation of different component units in each photovoltaic module over time The set of changes; ; Indicates the The first photovoltaic module The attenuation factor of the power generation of each component unit changes with time The set of changes; Represents different attenuation factors over time changes; ; ; Step S32: Establish a power generation analysis model and set the total The power generation of each component unit in is taken as the dependent variable, and the total set The attenuation factor of the power generation of each component unit is used as the dependent variable to analyze the impact of different attenuation factors on the power generation of the component unit. According to the calculation formula: ; in, Indicates the power generation of the component unit; Represent different attenuation factors respectively; and Indicates the influence coefficient of the attenuation factor on the power generation of the component unit; They respectively represent the degree of influence of different attenuation factors on the power generation of component units.

[0016] It should be noted that, based on historical power generation data and historical image data, the total The power generation of different component units in each photovoltaic module changes over time The change value of The training parameters of Different decay factors over time The change value of The training parameters of They correspond to the first interference factor in the defect, Corresponding to the second interference factor in the defect respectively; the above training parameters are substituted into the calculation formula for training, and the influence coefficient of the attenuation factor on the power generation of the component unit is determined according to the least squares method and , and determine the impact of different attenuation factors on the power generation of component units ,Through the established power generation analysis model, it is convenient to ,predict the total power generation of photovoltaic modules in the future, thereby improving ,the accuracy of photovoltaic power station fault judgment.

[0017] Step S40: monitor the total power generation of different photovoltaic modules based on their power generation data to determine whether there are any photovoltaic modules with a risk of failure; if there are any photovoltaic modules with a risk of failure, execute step S50; if there are no photovoltaic modules with a risk of failure, execute step S60.

[0018] Specifically, the steps for determining whether a photovoltaic module has a failure risk are as follows: Step S41: Based on the power generation data of different photovoltaic modules, the power generation of different module units in the photovoltaic module is accumulated to obtain the total power generation of different photovoltaic modules over time. Change in value ; Step S42: Calculate the discrete value between the total power generation of each photovoltaic module and the total power generation of other photovoltaic modules according to the calculation formula: ; in, Indicates the The discrete value of the total power generation of a photovoltaic module and the total power generation of other photovoltaic modules over time changes; Indicates the The total power generation of a photovoltaic module over time The change value of Indicates the The total power generation of a photovoltaic module over time The change value of ,and ; Step S43: Based on the power generation data when the photovoltaic module fails, obtain the discrete value between the total power generation when the photovoltaic module fails and the total power generation of other photovoltaic modules, and determine the safety threshold of the photovoltaic module failure risk. ;Will and For comparison, when there is When the PV panels are at risk of failure, There is no risk of PV module failure.

[0019] It should be noted that the photovoltaic module failure refers to interference factors that affect the power generation of the module unit and cannot be directly determined from the image data taken by the drone; by analyzing the total power generation of each photovoltaic module, the photovoltaic module failure can be effectively monitored preliminarily, and there is no need to monitor the power generation of different module units in the photovoltaic module one by one, which reduces the maintenance cost and difficulty of the system and saves regulatory resources.

[0020] Step S50: Determine the location information of the photovoltaic module at risk of failure, dispatch a drone to capture real-time image data, collect real-time power generation data of the photovoltaic module at risk of failure, determine the type of photovoltaic module failure based on the captured real-time image data, the collected real-time power generation data, and the impact of different attenuation factors on the power generation of the module unit, issue a safety alarm, and send the alarm signal and the determined photovoltaic module failure type to the management personnel; Specifically, the method steps are: Step S51: Analyze the captured real-time image data to determine the attenuation factor of the power generation of different component units in the photovoltaic module with a risk of failure; analyze the collected real-time power generation data to determine the power generation of different component units in the photovoltaic module with a risk of failure over time. Changes ; Step S52: Substitute the attenuation factor of the power generation of different component units in the photovoltaic module with failure risk into the calculation formula of step S32 to predict the power generation of different component units in the photovoltaic module with failure risk. ; Step S53: Determine The timestamps of the power generation of different component units predicted in the Respectively For comparison, when there is When the fault type of the PV module is internal fault; when there is When , the fault type of the PV module is external fault; Indicates the number of PV modules predicted to have a risk of failure. The power generation of each module unit; Indicates the number of PV modules with a risk of failure. The power generation of each module unit changes with time changes; represents the error term; Step S54: Issue a safety alarm, and send the alarm signal and the determined photovoltaic module fault type to the management personnel.

[0021] It should be noted that preliminary monitoring of photovoltaic module failures, identification of photovoltaic modules with failure risks, collection of image data of photovoltaic modules with failure risks by drones, and further analysis through the established power generation analysis model avoid the singleness brought about by photovoltaic power station fault diagnosis based on image recognition and analysis. Through diversified judgment methods, the accuracy of photovoltaic module fault identification can be effectively improved.

[0022] Step S60: The photovoltaic power station is safe, and steps S10 to S60 are repeated.

[0023] In this second embodiment: a photovoltaic power station fault diagnosis system based on industrial vision is provided, which includes a power generation data acquisition module, an image data acquisition module, an intelligent analysis module, a monitoring module, a fault diagnosis module and an alarm module; The power generation data acquisition module is used to collect power generation data of different photovoltaic modules; each photovoltaic module includes a plurality of module units; the power generation data includes the power generation of different module units in the photovoltaic module; and the collected power generation data is sent to the intelligent analysis module, the monitoring module and the fault diagnosis module; The image data acquisition module is used to collect image data of a plurality of photovoltaic modules during power generation, analyze the collected image data, determine the attenuation factors that affect the power generation of different component units in the photovoltaic module, and send the collected image data to the intelligent analysis module and the fault diagnosis module; The intelligent analysis module is used to establish a database, store the power generation data sent by the power generation data acquisition module and the image data sent by the image data acquisition module as historical data in the database; establish a power generation analysis model to analyze the impact of different attenuation factors on the power generation of the component units; and send the impact of different attenuation factors on the power generation of the component units to the fault diagnosis module; The monitoring module is used to monitor the total power generation of different photovoltaic modules based on the power generation data sent by the power generation data acquisition module, and determine whether there are photovoltaic modules with a risk of failure; if there are photovoltaic modules with a risk of failure, the location information of the photovoltaic modules with a risk of failure is sent to the fault diagnosis module; if there are no photovoltaic modules with a risk of failure, the monitoring is continued; The fault diagnosis module is used to determine the location information of the photovoltaic component with a risk of failure, send a signal to the image data acquisition module, dispatch a drone to capture real-time image data, determine the real-time power generation data of the photovoltaic component with a risk of failure sent by the power generation data acquisition module, determine the type of photovoltaic component failure based on the real-time image data, real-time power generation data and the impact of different attenuation factors on the power generation of the component unit, and send the determined photovoltaic component failure type to the alarm module; The alarm module is used to generate a safety alarm and send the alarm signal and the determined photovoltaic component fault type to the management personnel.

[0024] Furthermore, the intelligent analysis module includes a historical data analysis unit and a model management unit; The historical data analysis unit is used to retrieve historical power generation data and historical image data from the database for analysis, and obtain, based on the historical power generation data, a total set of changes over time in power generation of different component units in each photovoltaic module; and, based on the historical image data, obtain a total set of changes over time in attenuation factors that affect power generation of different component units in each photovoltaic module; The model management unit is used to establish a power generation analysis model, taking the power generation of each component unit as a dependent variable and the attenuation factor of the power generation of each component unit as a dependent variable, and analyzing the impact of different attenuation factors on the power generation of the component unit.

[0025] Furthermore, the monitoring module includes a total power generation analysis unit, a discrete value calculation unit and a fault risk judgment unit; The total power generation analysis unit is used to accumulate the power generation of different component units in the photovoltaic assembly according to the power generation data of different photovoltaic assemblies to obtain the change value of the total power generation of different photovoltaic assemblies over time; The discrete value calculation unit is used to respectively calculate the discrete value between the total power generation of each photovoltaic module and the total power generation of other photovoltaic modules; The fault risk judgment unit is used to judge whether there are photovoltaic components with fault risks.

[0026] Furthermore, the fault diagnosis module includes a real-time data analysis unit, a power generation prediction unit and a fault type judgment unit; The real-time data analysis unit is used to analyze the captured real-time image data to determine the attenuation factor of the power generation of different component units in the photovoltaic module with a risk of failure; and to analyze the collected real-time power generation data to determine the change in the power generation of different component units in the photovoltaic module with a risk of failure over time; The power generation prediction unit is used to predict the power generation of different component units in the photovoltaic component with failure risk; The fault type determination unit is used to determine the fault type of the photovoltaic module.

[0027] Furthermore, the alarm module provides an interactive display platform, through which managers can view the fault types of photovoltaic modules and view the stored historical power generation data and historical image data through the database.

[0028] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A photovoltaic power station fault diagnosis method based on industrial vision, characterized by: The method comprises the following steps: Step S10: generating electricity through a plurality of photovoltaic modules in a photovoltaic power station, and collecting power generation data of different photovoltaic modules; each photovoltaic module includes a plurality of module units; and the power generation data includes power generation of different module units in the photovoltaic module; Step S20: Perform regular inspections of the photovoltaic power station using a drone to collect image data of a plurality of photovoltaic modules during power generation, analyze the collected image data, and determine attenuation factors that affect power generation of different module units in the photovoltaic module; Step S30: storing the power generation data collected in step S10 and the image data collected in step S20 as historical data in a database, establishing a power generation analysis model based on the historical data in the database, and analyzing the impact of different attenuation factors on the power generation of the module units; Step S40: Based on the power generation data of different photovoltaic modules, the total power generation of the different photovoltaic modules is monitored to determine whether there is a photovoltaic module at risk of failure; if there is a photovoltaic module at risk of failure, step S50 is executed; if there is no photovoltaic module at risk of failure, step S60 is executed; Step S50: Determine the location information of the photovoltaic module at risk of failure, dispatch a drone to capture real-time image data, collect real-time power generation data of the photovoltaic module at risk of failure, determine the type of photovoltaic module failure based on the captured real-time image data, the collected real-time power generation data, and the impact of different attenuation factors on the power generation of the module unit, issue a safety alarm, and send the alarm signal and the determined photovoltaic module failure type to the management personnel; Step S60: The photovoltaic power station is safe, and steps S10 to S60 are repeated.

2. The photovoltaic power station fault diagnosis method based on industrial vision according to claim 1 is characterized by: The size and specifications of each photovoltaic module in a photovoltaic power station are the same, and the size and number of module units in each photovoltaic module are the same; the attenuation factor represents different defect types that affect the power generation of the module unit and the coverage area under the corresponding defect type; the photovoltaic module failure types are divided into internal failures and external failures; the internal failure represents an interference factor that cannot be directly determined from the image data taken by the drone and will affect the power generation of the module unit; the external failure represents an interference factor that can be directly determined from the image data taken by the drone and will affect the power generation of the module unit; the internal failure includes several defects, wherein each defect corresponds to an interference factor that will affect the power generation of the module unit.

3. The photovoltaic power station fault diagnosis method based on industrial vision according to claim 2 is characterized by: The method steps of step S30 are: Step S31: retrieve historical power generation data and historical image data from the database for analysis, and obtain the power generation of different component units in each photovoltaic module over time based on the historical power generation data. The total set of changes ; Based on the historical image data, the attenuation factor affecting the power generation of different component units in each photovoltaic module is obtained over time The total set of changes ; Step S32: Establish a power generation analysis model and set the total The power generation of each component unit in is taken as the dependent variable, and the total set The attenuation factor of the power generation of each component unit is used as the dependent variable to analyze the impact of different attenuation factors on the power generation of the component unit. According to the calculation formula: ； in, Indicates the power generation of the component unit; Represent different attenuation factors respectively; and Indicates the influence coefficient of the attenuation factor on the power generation of the component unit; They respectively represent the degree of influence of different attenuation factors on the power generation of component units.

4. The photovoltaic power station fault diagnosis method based on industrial vision according to claim 3 is characterized by: The steps to determine whether there is a risk of PV module failure are as follows: Step S41: Based on the power generation data of different photovoltaic modules, the power generation of different module units in the photovoltaic module is accumulated to obtain the total power generation of different photovoltaic modules over time. Change in value ; Step S42: Calculate the discrete value between the total power generation of each photovoltaic module and the total power generation of other photovoltaic modules according to the calculation formula: ； in, Indicates the The discrete value of the total power generation of a photovoltaic module and the total power generation of other photovoltaic modules over time changes; Indicates the The total power generation of a photovoltaic module over time The change value of Indicates the The total power generation of a photovoltaic module over time The change value of ,and ; Step S43: Based on the power generation data when the photovoltaic module fails, obtain the discrete value between the total power generation when the photovoltaic module fails and the total power generation of other photovoltaic modules, and determine the safety threshold of the photovoltaic module failure risk. ;Will and For comparison, when there is When the PV panels are at risk of failure, There is no risk of PV module failure.

5. The photovoltaic power station fault diagnosis method based on industrial vision according to claim 4 is characterized in that: The method steps of step S50 are: Step S51: Analyze the captured real-time image data to determine the attenuation factor of the power generation of different component units in the photovoltaic module with a risk of failure; analyze the collected real-time power generation data to determine the power generation of different component units in the photovoltaic module with a risk of failure over time. Changes ; Step S52: Substitute the attenuation factor of the power generation of different component units in the photovoltaic module with failure risk into the calculation formula of step S32 to predict the power generation of different component units in the photovoltaic module with failure risk. ; Step S53: Determine The timestamps of the power generation of different component units predicted in the Respectively For comparison, when there is When the fault type of the PV module is internal fault; when there is When , the fault type of the PV module is external fault; Indicates the number of PV modules predicted to have a risk of failure. The power generation of each module unit; Indicates the number of PV modules with a risk of failure. The power generation of each module unit changes with time changes; represents the error term; Step S54: Issue a safety alarm, and send the alarm signal and the determined photovoltaic module fault type to the management personnel.

6. A photovoltaic power station fault diagnosis system based on industrial vision, characterized by: The system includes power generation data acquisition module, image data acquisition module, intelligent analysis module, monitoring module, fault diagnosis module and alarm module; The power generation data acquisition module is used to collect power generation data of different photovoltaic modules; each photovoltaic module includes a plurality of module units; the power generation data includes the power generation of different module units in the photovoltaic module; and the collected power generation data is sent to the intelligent analysis module, the monitoring module and the fault diagnosis module; The image data acquisition module is used to collect image data of a plurality of photovoltaic modules during power generation, analyze the collected image data, determine the attenuation factors that affect the power generation of different component units in the photovoltaic module, and send the collected image data to the intelligent analysis module and the fault diagnosis module; The intelligent analysis module is used to establish a database, store the power generation data sent by the power generation data acquisition module and the image data sent by the image data acquisition module as historical data in the database; establish a power generation analysis model to analyze the impact of different attenuation factors on the power generation of the component units; and send the impact of different attenuation factors on the power generation of the component units to the fault diagnosis module; The monitoring module is used to monitor the total power generation of different photovoltaic modules based on the power generation data sent by the power generation data acquisition module, and determine whether there are photovoltaic modules with a risk of failure; if there are photovoltaic modules with a risk of failure, the location information of the photovoltaic modules with a risk of failure is sent to the fault diagnosis module; if there are no photovoltaic modules with a risk of failure, the monitoring is continued; The fault diagnosis module is used to determine the location information of the photovoltaic component with a risk of failure, send a signal to the image data acquisition module, dispatch a drone to capture real-time image data, determine the real-time power generation data of the photovoltaic component with a risk of failure sent by the power generation data acquisition module, determine the type of photovoltaic component failure based on the real-time image data, real-time power generation data and the impact of different attenuation factors on the power generation of the component unit, and send the determined photovoltaic component failure type to the alarm module; The alarm module is used to generate a safety alarm and send the alarm signal and the determined photovoltaic component fault type to the management personnel.

7. The photovoltaic power station fault diagnosis system based on industrial vision according to claim 6 is characterized by: The intelligent analysis module includes a historical data analysis unit and a model management unit; The historical data analysis unit is used to retrieve historical power generation data and historical image data from the database for analysis, and obtain, based on the historical power generation data, a total set of changes over time in power generation of different component units in each photovoltaic module; and, based on the historical image data, obtain a total set of changes over time in attenuation factors that affect power generation of different component units in each photovoltaic module; The model management unit is used to establish a power generation analysis model, taking the power generation of each component unit as a dependent variable and the attenuation factor of the power generation of each component unit as a dependent variable, and analyzing the impact of different attenuation factors on the power generation of the component unit.

8. The photovoltaic power station fault diagnosis system based on industrial vision according to claim 7 is characterized by: The monitoring module includes a total power generation analysis unit, a discrete value calculation unit and a fault risk judgment unit; The total power generation analysis unit is used to accumulate the power generation of different component units in the photovoltaic assembly according to the power generation data of different photovoltaic assemblies to obtain the change value of the total power generation of different photovoltaic assemblies over time; The discrete value calculation unit is used to respectively calculate the discrete value between the total power generation of each photovoltaic module and the total power generation of other photovoltaic modules; The fault risk judgment unit is used to judge whether there are photovoltaic components with fault risks.

9. The photovoltaic power station fault diagnosis system based on industrial vision according to claim 8, characterized in that: The fault diagnosis module includes a real-time data analysis unit, a power generation prediction unit and a fault type judgment unit; The real-time data analysis unit is used to analyze the captured real-time image data to determine the attenuation factor of the power generation of different component units in the photovoltaic module with a risk of failure; and to analyze the collected real-time power generation data to determine the change in the power generation of different component units in the photovoltaic module with a risk of failure over time; The power generation prediction unit is used to predict the power generation of different component units in the photovoltaic component with failure risk; The fault type determination unit is used to determine the fault type of the photovoltaic module.

10. The photovoltaic power station fault diagnosis system based on industrial vision according to claim 9, characterized in that: The alarm module provides an interactive display platform, through which managers can view the fault types of photovoltaic components and view the stored historical power generation data and historical image data through the database.

Images