An artificial intelligence-based photovoltaic cell panel inspection system and method

Abstract

The application discloses a kind of photovoltaic cell panel inspection systems and methods based on artificial intelligence, belong to photovoltaic cleaning technical field.The surface image information of photovoltaic cell panel, air dust concentration, environmental wind speed and environmental wind direction are collected to carry out early warning to the dust accumulation degree of photovoltaic cell panel surface;Considering that dust accumulation is not uniform speed with time increase, seasonal characteristics, local interference factors or regional natural environment influence can make dust accumulation speed change significantly, although the seasonal characteristics of different regions are different, but the precipitation brought by seasonal characteristics is regular, and the influence of the change trend of local factors such as air dust concentration, environmental wind speed and environmental wind direction on photovoltaic cell panel dust accumulation is quantitatively analyzed to determine the influence rate;So it can monitor and early warning the dust accumulation degree of photovoltaic cell panel surface in real time, make power generation efficiency can be maximized, suitable for photovoltaic power station in different environment, region.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic clean technology, specifically to an artificial intelligence-based photovoltaic panel inspection system and method. Background Technology

[0002] Photovoltaic panels are a crucial core component of photovoltaic power plants, converting solar energy into electricity. Since photovoltaic panels collect solar energy outdoors, they accumulate dust over time, affecting power generation efficiency. Therefore, regular cleaning of photovoltaic panels is necessary. Current technologies typically employ manual or mechanical cleaning, but the drawback is that dust accumulation does not increase uniformly over time. Seasonal characteristics, local disturbances, or regional environmental factors can significantly alter the rate of dust accumulation. Therefore, a scheduled dust removal model cannot automatically adapt to these changes, lacking artificial intelligence, resulting in untimely dust removal and preventing the maximization of power generation efficiency. Summary of the Invention

[0003] The purpose of this invention is to provide an artificial intelligence-based photovoltaic panel inspection system and method to solve the problems mentioned in the background art.

[0004] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0005] An artificial intelligence-based photovoltaic panel inspection method, comprising the following steps:

[0006] Step S100: Obtain photovoltaic panel surface image information from historical data, generate photovoltaic panel surface grayscale value based on the image information, and calculate the growth rate of photovoltaic panel surface grayscale value;

[0007] Step S200: Obtain the air dust concentration and ambient wind value of the area where the photovoltaic panel is located from historical data. The ambient wind value includes ambient wind speed and ambient wind direction. Calculate the impact rate of the increase in gray value on the surface of the photovoltaic panel.

[0008] Step S300: Calculate the dust accumulation rate on the photovoltaic panel surface based on the growth rate of the gray value on the photovoltaic panel surface and the influence rate of the gray value growth on the photovoltaic panel surface;

[0009] Step S400: Obtain the current image information of the photovoltaic panel surface and obtain the current grayscale value of the photovoltaic panel surface; based on the current grayscale value of the photovoltaic panel surface and the dust accumulation rate of the photovoltaic panel surface, establish a dust accumulation prediction model for the photovoltaic panel surface, perform artificial intelligence prediction on the amount of dust accumulation on the photovoltaic panel surface, and output early warning information;

[0010] Step S500: Obtain meteorological information of the area where the photovoltaic panel is located from historical data, wherein the meteorological information is the precipitation in the area where the photovoltaic panel is located; establish a rainwater washing model based on the precipitation, calculate the surface gray value of the photovoltaic panel after rainwater washing, perform artificial intelligence calibration on the early warning information, and output calibration information;

[0011] In the above method, the calculated growth rate of the gray value on the surface of the photovoltaic panel does not include precipitation, because photovoltaic power stations are usually not built in areas with frequent rainfall. Photovoltaic power stations need to be built in areas with abundant sunshine and long hours of sunlight. However, the impact of air dust and ambient wind occurs frequently, even daily. Wind sometimes blows away dust from the surface of the photovoltaic panel, and at the same time, it brings dust from the air into the surface of the photovoltaic panel. Air dust and ambient wind are the main factors affecting the accumulation of dust on the surface of photovoltaic panels. Therefore, it is necessary to take into account the concentration of air dust and ambient wind value, and to treat precipitation as a special case. At the same time, precipitation can wash away some of the dust on the surface of the photovoltaic panel.

[0012] Furthermore, the specific implementation process for calculating the growth rate of grayscale values ​​on the surface of the photovoltaic panel includes:

[0013] The grayscale value of the photovoltaic panel surface is denoted as . ,in Indicates the first During the second inspection, image information of the photovoltaic panel surface is collected to generate grayscale values ​​of the photovoltaic panel surface. The specific calculation formula for the growth rate of the grayscale values ​​of the photovoltaic panel surface is as follows:

[0014]

[0015] in, This indicates the growth rate of grayscale value on the surface of the photovoltaic panel. This represents the total number of times the image information of the photovoltaic panel surface was collected to generate the grayscale value of the photovoltaic panel surface. Indicates the first During the second inspection, image information of the photovoltaic panel surface is collected to generate grayscale values ​​of the photovoltaic panel surface. Indicates the first The second inspection to the first The time interval between collecting image information of the photovoltaic panel surface during the second inspection and generating grayscale values ​​of the photovoltaic panel surface. ;

[0016] In the above method, the amount of dust accumulation does not increase at a uniform rate over time, so it is necessary to subtract the gray values ​​of each inspection in turn to refine the gray value changes and make the data analysis more accurate; the inspection can be carried out by drones.

[0017] Furthermore, the specific implementation process for calculating the impact rate of the increase in grayscale value on the surface of the photovoltaic panel includes:

[0018] Step S201: Obtain the ambient wind value of the area where the photovoltaic panel is located from historical data, taking the direction parallel to the photovoltaic panel as the fixed direction, where the photovoltaic panel direction is the direction in which the photovoltaic panel is tilted downwards; decompose the ambient wind direction of the area where the photovoltaic panel is located into two ambient wind vectors, one parallel to the fixed direction and one perpendicular to the fixed direction, and record the ambient wind vector parallel to the fixed direction as the parallel ambient wind direction, and record the angle between the decomposed parallel ambient wind directions as... ,and ;

[0019] In the above method, the ambient wind direction can be easily obtained through a wind vane. The ambient wind is decomposed to study the impact of the ambient wind on the photovoltaic panel. The parallel ambient wind direction is the key influencing factor that blows away or brings dust into the surface of the photovoltaic panel, while the ambient wind vector perpendicular to the fixed direction is a normal force and its role in the force is negligible. Therefore, the parallel ambient wind needs to be studied as the main focus of the two ambient wind vectors.

[0020] Step S202: Record the air dust concentration as The ambient wind speed is recorded as Record the ambient wind direction as ,in Indicates the first Next to During the inspection This monitoring was conducted; using the air dust concentration, ambient wind speed, and surface gray value of the photovoltaic panels in the area where they are located as independent variables, and the influence rate of the increase in surface gray value of the photovoltaic panels as the dependent variable, a training model was established. The specific calculation formula is as follows:

[0021]

[0022] Wherein, K represents the impact rate of the increase in gray value on the surface of the photovoltaic panel. This represents the total number of times the image information of the photovoltaic panel surface was collected to generate the grayscale value of the photovoltaic panel surface. and They represent the first Second and third The decomposed angle of the parallel environmental wind direction in the second monitoring, where n represents the angle of the first monitoring. Next to Total number of monitoring sessions during the inspection period , , and Represents the training coefficient values. , and They represent the first The air dust concentration, ambient wind speed, and ambient wind direction were monitored in this instance. ;

[0023] In the above method, the denominator of the training model calculation formula is actually based on the idea of ​​a multivariate linear model algorithm. The three factors affecting the dust accumulation rate of photovoltaic panels—air dust concentration, ambient wind speed, and ambient wind direction—are used as independent variables for training. This allows us to obtain the relationship between the fluctuation of dust accumulation on the surface of the photovoltaic panel and the changes in the three environmental factors, and to determine the impact of the changing trends of the three factors on the dust accumulation of the photovoltaic panel. The calculated growth rate of the gray value of the photovoltaic panel surface actually includes the influence of air dust concentration, ambient wind speed, and ambient wind direction. Therefore, it is necessary to train and convert the influence rate of the gray value growth of the photovoltaic panel surface.

[0024] Furthermore, the specific implementation process of establishing a prediction model for dust accumulation on the surface of photovoltaic panels, using artificial intelligence to predict the amount of dust accumulation on the surface of photovoltaic panels, and outputting early warning information includes:

[0025] Step S401: Acquire the current image information of the photovoltaic panel surface, and generate the grayscale value of the photovoltaic panel surface from the current image information; based on the current grayscale value of the photovoltaic panel surface and the dust accumulation rate of the photovoltaic panel surface, establish a dust accumulation prediction model for the photovoltaic panel surface in units of time. The specific calculation formula is as follows:

[0026]

[0027] in, The function representing the grayscale value of the photovoltaic panel surface with respect to time. This indicates the current grayscale value of the photovoltaic panel surface. Indicates the current time. Indicates the rate of dust accumulation on the surface of photovoltaic panels Functions of time

[0028] Step S402: Set the warning value for the grayscale of the photovoltaic panel surface, denoted as... ;if , The system indicates the warning time when dust needs to be cleaned from the surface of the photovoltaic panels. It then uses artificial intelligence to predict the amount of dust accumulation on the photovoltaic panel surface, outputting warning information to remind staff to clean it. Within a specified timeframe, the surface of the photovoltaic panels is cleaned using artificial intelligence; if If so, continue with step S401.

[0029] Furthermore, based on the rainfall, a rainwater washing model is established, and the early warning information is calibrated using artificial intelligence. The specific implementation process for outputting calibration information includes:

[0030] Step S501: Obtain the precipitation amount for each rainfall event in the area where the photovoltaic panels are located from historical data, and record it as follows: To obtain the duration of each precipitation event, denoted as... Using the rainfall amount and duration of each precipitation event as independent variables, and the surface gray value of the photovoltaic panel after each precipitation event as the dependent variable, a rainwater cleaning model is established to calculate the surface gray value of the photovoltaic panel after rainwater cleaning. The specific calculation formula is as follows:

[0031]

[0032] in, This represents the surface grayscale value of the photovoltaic panel after each rainfall event. and This indicates the adjustment coefficient value;

[0033] Step S502: In If precipitation occurs in the area where the photovoltaic panels are located within the specified time frame, then step S503 is executed; If no precipitation occurs in the area where the photovoltaic panels are located within the time frame, then proceed to step S400.

[0034] Step S503: Obtain the rainfall and duration of rainfall in the area where the photovoltaic panel is located. Based on the rainwater washing model, calculate the surface gray value of the photovoltaic panel after rainwater washing. Replace with B, execute step S400, re-acquire the warning time when dust needs to be cleaned from the photovoltaic panel surface, and record the new warning time as the calibration time. The system performs AI-based calibration on the early warning information, outputs calibration data, and prompts staff to... The system performs AI-powered cleaning of the photovoltaic panel surface within a short timeframe.

[0035] An artificial intelligence-based photovoltaic panel inspection system, comprising: a data acquisition module, a data processing module, a model building module, and an early warning module;

[0036] The data acquisition module is used to acquire image information of the photovoltaic panel surface, air dust concentration in the area where the photovoltaic panel is located, and ambient wind value, including ambient wind speed and ambient wind direction; it is used to generate grayscale values ​​of the photovoltaic panel surface based on the image information; and it is used to decompose the ambient wind direction in the area where the photovoltaic panel is located into two ambient wind vectors, one parallel to the fixed direction and the other perpendicular to the fixed direction, with the direction parallel to the fixed direction being the direction in which the photovoltaic panel is tilted downwards. The ambient wind vector parallel to the fixed direction is denoted as the parallel ambient wind direction, and the angle between the decomposed parallel ambient wind directions is denoted as... ,and Used to collect rainfall data in the area where photovoltaic panels are located, denoted as . The duration of precipitation is recorded as follows: ;

[0037] The data processing module is used to calculate the growth rate of grayscale value on the surface of the photovoltaic panel; to calculate the impact rate of the growth of grayscale value on the surface of the photovoltaic panel; and to calculate the dust accumulation rate on the surface of the photovoltaic panel.

[0038] The model building module is used to build a prediction model for dust accumulation on the surface of photovoltaic panels and to build a rainwater cleaning model.

[0039] The early warning module is used to predict the amount of dust accumulation on the surface of the photovoltaic panel using artificial intelligence and output early warning information; it is also used to calibrate the early warning information using artificial intelligence and output calibration information.

[0040] Furthermore, the data processing module also includes a growth rate calculation unit, an impact rate calculation unit, and a dust accumulation rate calculation unit;

[0041] The growth rate calculation unit is used to calculate the growth rate of the gray value on the surface of the photovoltaic panel. ,in, This indicates the growth rate of grayscale value on the surface of the photovoltaic panel. This represents the total number of times the image information of the photovoltaic panel surface was collected to generate the grayscale value of the photovoltaic panel surface. Indicates the first During the second inspection, image information of the photovoltaic panel surface is collected to generate grayscale values ​​of the photovoltaic panel surface. Indicates the first During the second inspection, image information of the photovoltaic panel surface is collected to generate grayscale values ​​of the photovoltaic panel surface. Indicates the first The second inspection to the first The time interval between collecting image information of the photovoltaic panel surface during the second inspection and generating grayscale values ​​of the photovoltaic panel surface. ;

[0042] The influence rate calculation unit is used to calculate the influence rate of the increase in gray value on the surface of the photovoltaic panel. Where K represents the rate of increase in the grayscale value of the photovoltaic panel surface. This represents the total number of times the image information of the photovoltaic panel surface was collected to generate the grayscale value of the photovoltaic panel surface. and They represent the first Second and third The decomposed angle of the parallel environmental wind direction in the second monitoring, where n represents the angle of the first monitoring. Next to Total number of monitoring sessions during the inspection period , , and Indicates the training coefficient value. , and They represent the first The air dust concentration, ambient wind speed, and ambient wind direction were monitored in this instance. , and They represent the first The air dust concentration, ambient wind speed, and ambient wind direction were monitored in this instance. ;

[0043] Furthermore, the model building module also includes a dust accumulation prediction model unit and a rainwater washing model unit;

[0044] The dust accumulation prediction model unit is used to establish a dust accumulation prediction model on the surface of photovoltaic panels. ,in, The function representing the grayscale value of the photovoltaic panel surface with respect to time. This indicates the current grayscale value of the photovoltaic panel surface. Indicates the current time. The function representing the rate of dust accumulation on the surface of a photovoltaic panel with respect to time.

[0045] The rainwater washing model unit is used to establish a rainwater washing model. ,in, This represents the surface grayscale value of the photovoltaic panel after each rainfall event. and This indicates the adjustment coefficient value.

[0046] Furthermore, the early warning module also includes an early warning unit and a calibration unit;

[0047] The early warning unit is used to set an early warning value for the grayness of the photovoltaic panel surface; if the amount of dust accumulated on the photovoltaic panel surface exceeds the early warning value, an artificial intelligence early warning is issued, and early warning information is output to prompt staff to perform artificial intelligence cleaning on the photovoltaic panel surface within the early warning time.

[0048] The calibration unit is used to perform artificial intelligence calibration on the early warning information, output calibration information, and prompt staff to perform artificial intelligence cleaning on the surface of the photovoltaic panel within the calibration time.

[0049] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: The photovoltaic panel inspection system and method based on artificial intelligence provided by this invention collects image information of the photovoltaic panel surface, air dust concentration, ambient wind speed, and ambient wind direction to provide early warning of the degree of dust accumulation on the photovoltaic panel surface. Through big data artificial intelligence algorithms, data analysis becomes more accurate. Considering that dust accumulation does not increase uniformly over time, seasonal characteristics, local interference factors, or regional natural environmental influences can significantly change the dust accumulation rate. While seasonal characteristics differ across regions, the resulting precipitation patterns are regular. Furthermore, the impact of local factors such as air dust concentration, ambient wind speed, and ambient wind direction on dust accumulation on photovoltaic panels is quantitatively analyzed to determine the impact rate. This enables real-time monitoring and early warning of the degree of dust accumulation on the photovoltaic panel surface, ensuring timely dust removal and maximizing power generation efficiency. This system is applicable to photovoltaic power plants in different environments and regions. Attached Figure Description

[0050] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0051] Figure 1 This is a schematic diagram of the structure of a photovoltaic panel inspection system based on artificial intelligence according to the present invention;

[0052] Figure 2 This is a flowchart illustrating an artificial intelligence-based photovoltaic panel inspection method according to the present invention. Detailed Implementation

[0053] 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, and 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.

[0054] Please see Figures 1-2The present invention provides the following technical solution:

[0055] Please see Figure 1 In this first embodiment: an artificial intelligence-based photovoltaic panel inspection system is provided. The system includes: a data acquisition module, a data processing module, a model building module, and an early warning module.

[0056] The data acquisition module is used to collect image information of the photovoltaic panel surface, air dust concentration in the area where the photovoltaic panel is located, and ambient wind values, including ambient wind speed and direction. It is used to generate grayscale values ​​for the photovoltaic panel surface based on the image information. It is used to decompose the ambient wind direction in the area where the photovoltaic panel is located into two ambient wind vectors, one parallel to the fixed direction and the other perpendicular to the fixed direction, using a direction parallel to the photovoltaic panel (where the photovoltaic panel direction is the direction in which it tilts downwards). The ambient wind vector parallel to the fixed direction is denoted as the parallel ambient wind direction, and the angle between the decomposed parallel ambient wind directions is denoted as... ,and Used to collect rainfall data in the area where photovoltaic panels are located, denoted as . The duration of precipitation is recorded as follows: ;

[0057] The data processing module is used to calculate the growth rate of grayscale values ​​on the surface of photovoltaic panels; to calculate the impact rate of the growth of grayscale values ​​on the surface of photovoltaic panels; and to calculate the dust accumulation rate on the surface of photovoltaic panels.

[0058] The model building module is used to build a prediction model for dust accumulation on the surface of photovoltaic panels; it is also used to build a rainwater cleaning model.

[0059] The early warning module is used to predict the amount of dust accumulation on the surface of photovoltaic panels using artificial intelligence and output early warning information; it is also used to calibrate the early warning information using artificial intelligence and output calibration information.

[0060] The data processing module also includes a growth rate calculation unit, an impact rate calculation unit, and a dust accumulation rate calculation unit.

[0061] The growth rate calculation unit is used to calculate the growth rate of grayscale values ​​on the surface of photovoltaic panels. ,in, This indicates the growth rate of grayscale value on the surface of the photovoltaic panel. This represents the total number of times the image information of the photovoltaic panel surface was collected to generate the grayscale value of the photovoltaic panel surface. Indicates the first During the second inspection, image information of the photovoltaic panel surface is collected to generate grayscale values ​​of the photovoltaic panel surface. Indicates the first During the second inspection, image information of the photovoltaic panel surface is collected to generate grayscale values ​​of the photovoltaic panel surface. Indicates the first The second inspection to the first The time interval between collecting image information of the photovoltaic panel surface during the second inspection and generating grayscale values ​​of the photovoltaic panel surface. ;

[0062] The impact rate calculation unit is used to calculate the impact rate of the increase in gray value on the surface of the photovoltaic panel. Where K represents the rate of increase in the grayscale value of the photovoltaic panel surface. This represents the total number of times the image information of the photovoltaic panel surface was collected to generate the grayscale value of the photovoltaic panel surface. and They represent the first Second and third The decomposed angle of the parallel environmental wind direction in the second monitoring, where n represents the angle of the first monitoring. Next to Total number of monitoring sessions during the inspection period , , and Indicates the training coefficient value. , and They represent the first The air dust concentration, ambient wind speed, and ambient wind direction were monitored in this instance. , and They represent the first The air dust concentration, ambient wind speed, and ambient wind direction were monitored in this instance. ;

[0063] The model building module also includes a dust accumulation prediction model unit and a rainwater washing model unit.

[0064] The dust accumulation prediction model unit is used to establish a dust accumulation prediction model on the surface of photovoltaic panels. ,in, The function representing the grayscale value of the photovoltaic panel surface with respect to time. This indicates the current grayscale value of the photovoltaic panel surface. Indicates the current time. The function representing the rate of dust accumulation on the surface of a photovoltaic panel with respect to time.

[0065] Rainwater washing model unit, used to build rainwater washing model ,in, This represents the surface grayscale value of the photovoltaic panel after each rainfall event. and This indicates the adjustment coefficient value.

[0066] The early warning module also includes an early warning unit and a calibration unit;

[0067] The early warning unit is used to set the early warning value for the grayness of the photovoltaic panel surface; if the amount of dust accumulated on the photovoltaic panel surface exceeds the early warning value, an artificial intelligence early warning will be issued, and the early warning information will be output to prompt the staff to clean the photovoltaic panel surface with artificial intelligence within the early warning time.

[0068] The calibration unit is used to perform artificial intelligence calibration on the early warning information, output calibration information, and prompt staff to perform artificial intelligence cleaning on the surface of the photovoltaic panel within the calibration time.

[0069] Please see Figure 2 In this second embodiment, an artificial intelligence-based photovoltaic panel inspection method is provided, which includes the following steps:

[0070] Obtain image information of the photovoltaic panel surface from historical data, and generate grayscale values ​​for the photovoltaic panel surface from the image information, denoted as . ,in Indicates the first During each inspection, images of the photovoltaic (PV) panel surface are collected to generate grayscale values. The growth rate of these grayscale values ​​is then calculated. For example, if images of the PV panel surface are collected four times, generating grayscale values ​​of 70, 80, 90, and 100 respectively, with intervals of 10, 20, and 20 seconds between each collection, then the growth rate of the grayscale values ​​is calculated. ;

[0071] Obtain historical data on air dust concentration and ambient wind speed in the area where photovoltaic (PV) panels are located, using the direction parallel to the PV panels as a fixed direction (where the PV panels are tilted downwards). Decompose the ambient wind direction in the area of ​​the PV panels into two ambient wind vectors: one parallel to the fixed direction and one perpendicular to the fixed direction. The ambient wind vector parallel to the fixed direction is denoted as the parallel ambient wind direction, and the angle between the decomposed parallel ambient wind directions is denoted as... ,and The concentration of airborne dust is recorded as... The ambient wind speed is recorded as Record the ambient wind direction as ,in Indicates the first Next to During the inspection This monitoring was conducted; using the air dust concentration, ambient wind speed, and surface gray value of the photovoltaic panels in the area where they are located as independent variables, and the influence rate of the increase in surface gray value of the photovoltaic panels as the dependent variable, a training model was established. The specific calculation formula is as follows:

[0072]

[0073] Wherein, K represents the impact rate of the increase in gray value on the surface of the photovoltaic panel. This represents the total number of times the image information of the photovoltaic panel surface was collected to generate the grayscale value of the photovoltaic panel surface. and They represent the first Second and third The decomposed angle of the parallel environmental wind direction in the second monitoring, where n represents the angle of the first monitoring. Next to Total number of monitoring sessions during the inspection period , , and Indicates the training coefficient value. , and They represent the first The air dust concentration, ambient wind speed, and ambient wind direction were monitored in this instance. ;

[0074] Collect current image information of the photovoltaic panel surface and generate a grayscale value of 80 for the photovoltaic panel surface. Based on the current grayscale value and the dust accumulation rate on the photovoltaic panel surface, establish a dust accumulation prediction model for the photovoltaic panel surface in units of time. ;

[0075] The warning value for the grayscale level on the photovoltaic panel surface is set to 120; according to the photovoltaic panel surface dust accumulation prediction model, the prediction value is... hour, This indicates the warning time when the dust on the surface of the photovoltaic panel needs to be cleaned. If the amount of dust on the surface of the photovoltaic panel exceeds 120, staff will be prompted to perform artificial intelligence cleaning in 15 days.

[0076] Obtain the rainfall amount for each precipitation event in the area where the photovoltaic panels are located from historical data, and record it as follows: The duration of each precipitation event is obtained and denoted as... A rainwater cleaning model was established, using the rainfall amount and duration of each precipitation event as independent variables and the surface gray value of the photovoltaic panels after each precipitation event as the dependent variable. ,in, This represents the surface grayscale value of the photovoltaic panel after each rainfall event. and This represents the adjustment coefficient value; for example, in Within a given timeframe, if rainfall occurs in the area where the photovoltaic panels are located, the amount and duration of rainfall in that area are obtained. Based on the rainwater washing model, the surface grayscale value of the photovoltaic panels after rainwater washing is calculated to be 60. Replace with B and substitute. It should be noted that when there is a sudden rainfall, the surface gray value of the photovoltaic panel needs to be re-collected. According to the photovoltaic panel surface dust accumulation prediction model, the warning time for re-collecting the dust on the photovoltaic panel surface is 20 seconds. This prompts the staff to carry out artificial intelligence cleaning of the photovoltaic panel surface within 20 seconds.

[0077] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0078] 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 foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A photovoltaic panel inspection method based on artificial intelligence, characterized in that, The method includes the following steps: Step S100: Obtain photovoltaic panel surface image information from historical data, generate photovoltaic panel surface grayscale value based on the image information, and calculate the growth rate of photovoltaic panel surface grayscale value; Step S200: Obtain the air dust concentration and ambient wind value of the area where the photovoltaic panel is located from historical data. The ambient wind value includes ambient wind speed and ambient wind direction. Calculate the impact rate of the increase in gray value on the surface of the photovoltaic panel. Step S300: Calculate the dust accumulation rate on the photovoltaic panel surface based on the growth rate of the gray value on the photovoltaic panel surface and the influence rate of the gray value growth on the photovoltaic panel surface; Step S400: Obtain the current image information of the photovoltaic panel surface and obtain the current grayscale value of the photovoltaic panel surface; based on the current grayscale value of the photovoltaic panel surface and the dust accumulation rate of the photovoltaic panel surface, establish a dust accumulation prediction model for the photovoltaic panel surface, perform artificial intelligence prediction on the amount of dust accumulation on the photovoltaic panel surface, and output early warning information; Step S500: Obtain meteorological information of the area where the photovoltaic panel is located from historical data, wherein the meteorological information is the precipitation in the area where the photovoltaic panel is located; establish a rainwater washing model based on the precipitation, calculate the surface gray value of the photovoltaic panel after rainwater washing, perform artificial intelligence calibration on the early warning information, and output calibration information; The specific implementation process for calculating the impact rate of the increase in gray value on the surface of the photovoltaic panel includes: Step S201: Obtain the ambient wind value of the area where the photovoltaic panel is located from historical data, taking the direction parallel to the photovoltaic panel as the fixed direction, where the photovoltaic panel direction is the direction in which the photovoltaic panel is tilted downwards; decompose the ambient wind direction of the area where the photovoltaic panel is located into two ambient wind vectors, one parallel to the fixed direction and one perpendicular to the fixed direction, and record the ambient wind vector parallel to the fixed direction as the parallel ambient wind direction, and record the angle between the decomposed parallel ambient wind directions as... ,and ; Step S202: Record the air dust concentration as The ambient wind speed is recorded as Record the ambient wind direction as ,in Indicates the first Next to During the inspection This monitoring was conducted; using the air dust concentration, ambient wind speed, and surface gray value of the photovoltaic panels in the area where they are located as independent variables, and the influence rate of the increase in surface gray value of the photovoltaic panels as the dependent variable, a training model was established. The specific calculation formula is as follows: ； Wherein, K represents the impact rate of the increase in gray value on the surface of the photovoltaic panel. This represents the total number of times the image information of the photovoltaic panel surface was collected to generate the grayscale value of the photovoltaic panel surface. and They represent the first Second and third The decomposed angle of the parallel environmental wind direction in the second monitoring, where n represents the angle of the first monitoring. Next to Total number of monitoring sessions during the inspection period , , and Indicates the training coefficient value. , and They represent the first The air dust concentration, ambient wind speed, and ambient wind direction were monitored in this instance. , and They represent the first The air dust concentration, ambient wind speed, and ambient wind direction were monitored in this instance. .

2. The photovoltaic panel inspection method based on artificial intelligence according to claim 1, characterized in that, The specific implementation process for calculating the growth rate of grayscale value on the surface of a photovoltaic panel includes: The grayscale value of the photovoltaic panel surface is denoted as . ,in Indicates the first During the second inspection, image information of the photovoltaic panel surface is collected to generate grayscale values ​​of the photovoltaic panel surface. The specific calculation formula for the growth rate of the grayscale values ​​of the photovoltaic panel surface is as follows: ； in, This indicates the growth rate of grayscale value on the surface of the photovoltaic panel. This represents the total number of times the image information of the photovoltaic panel surface was collected to generate the grayscale value of the photovoltaic panel surface. Indicates the first During the second inspection, image information of the photovoltaic panel surface is collected to generate grayscale values ​​of the photovoltaic panel surface. Indicates the first The second inspection to the first The time interval between collecting image information of the photovoltaic panel surface during the second inspection and generating grayscale values ​​of the photovoltaic panel surface. .

3. The photovoltaic panel inspection method based on artificial intelligence according to claim 2, characterized in that, The specific implementation process of establishing a prediction model for dust accumulation on the surface of photovoltaic panels, using artificial intelligence to predict the amount of dust accumulation on the surface of photovoltaic panels, and outputting early warning information includes: Step S401: Acquire the current image information of the photovoltaic panel surface, and generate the grayscale value of the photovoltaic panel surface from the current image information; based on the current grayscale value of the photovoltaic panel surface and the dust accumulation rate of the photovoltaic panel surface, establish a dust accumulation prediction model for the photovoltaic panel surface in units of time. The specific calculation formula is as follows: ； in, The function representing the grayscale value of the photovoltaic panel surface with respect to time. This indicates the current grayscale value of the photovoltaic panel surface. Indicates the current time. Indicates the rate of dust accumulation on the surface of photovoltaic panels A function of time; Step S402: Set the warning value for the grayscale of the photovoltaic panel surface, denoted as... ;if , The system indicates the warning time when dust needs to be cleaned from the surface of the photovoltaic panels. It then uses artificial intelligence to predict the amount of dust accumulation on the photovoltaic panel surface, outputting warning information to remind staff to clean it. Within a specified timeframe, the surface of the photovoltaic panels is cleaned using artificial intelligence; if If so, continue with step S401.

4. The photovoltaic panel inspection method based on artificial intelligence according to claim 3, characterized in that, The specific implementation process of establishing a rainwater washing model based on the rainfall, performing artificial intelligence calibration on the early warning information, and outputting calibration information includes: Step S501: Obtain the precipitation amount for each rainfall event in the area where the photovoltaic panels are located from historical data, and record it as follows: To obtain the duration of each precipitation event, denoted as... Using the rainfall amount and duration of each precipitation event as independent variables, and the surface gray value of the photovoltaic panel after each precipitation event as the dependent variable, a rainwater cleaning model is established to calculate the surface gray value of the photovoltaic panel after rainwater cleaning. The specific calculation formula is as follows: ； in, This represents the surface grayscale value of the photovoltaic panel after each rainfall event. and This indicates the adjustment coefficient value; Step S502: In If precipitation occurs in the area where the photovoltaic panels are located within the specified time frame, then step S503 is executed; If no precipitation occurs in the area where the photovoltaic panels are located within the time frame, then proceed to step S400. Step S503: Obtain the rainfall and duration of rainfall in the area where the photovoltaic panel is located. Based on the rainwater washing model, calculate the surface gray value of the photovoltaic panel after rainwater washing. Replace with B, execute step S400, re-acquire the warning time when dust needs to be cleaned from the photovoltaic panel surface, and record the new warning time as the calibration time. The system performs AI-based calibration on the early warning information, outputs calibration data, and prompts staff to... The system performs AI-powered cleaning of the photovoltaic panel surface within a short timeframe.

5. An AI-based photovoltaic panel inspection system that implements an AI-based photovoltaic panel inspection method as described in any one of claims 1-4, characterized in that, The system includes: a data acquisition module, a data processing module, a model building module, and an early warning module; The data acquisition module is used to acquire image information of the photovoltaic panel surface, air dust concentration in the area where the photovoltaic panel is located, and ambient wind value, including ambient wind speed and ambient wind direction; it is used to generate grayscale values ​​of the photovoltaic panel surface based on the image information; and it is used to decompose the ambient wind direction in the area where the photovoltaic panel is located into two ambient wind vectors, one parallel to the fixed direction and the other perpendicular to the fixed direction, with the direction parallel to the fixed direction being the direction in which the photovoltaic panel is tilted downwards. The ambient wind vector parallel to the fixed direction is denoted as the parallel ambient wind direction, and the angle between the decomposed parallel ambient wind directions is denoted as... ,and Used to collect rainfall data in the area where photovoltaic panels are located, denoted as . The duration of precipitation is recorded as follows: ; The data processing module is used to calculate the growth rate of grayscale value on the surface of the photovoltaic panel; to calculate the impact rate of the growth of grayscale value on the surface of the photovoltaic panel; and to calculate the dust accumulation rate on the surface of the photovoltaic panel. The model building module is used to build a prediction model for dust accumulation on the surface of photovoltaic panels and to build a rainwater cleaning model. The early warning module is used to predict the amount of dust accumulation on the surface of the photovoltaic panel using artificial intelligence and output early warning information; it is also used to calibrate the early warning information using artificial intelligence and output calibration information.

6. The photovoltaic panel inspection system based on artificial intelligence according to claim 5, characterized in that: The data processing module also includes a growth rate calculation unit, an impact rate calculation unit, and a dust accumulation rate calculation unit; The growth rate calculation unit is used to calculate the growth rate of the gray value on the surface of the photovoltaic panel. ,in, This indicates the growth rate of grayscale value on the surface of the photovoltaic panel. This represents the total number of times the image information of the photovoltaic panel surface was collected to generate the grayscale value of the photovoltaic panel surface. Indicates the first During the second inspection, image information of the photovoltaic panel surface is collected to generate grayscale values ​​of the photovoltaic panel surface. Indicates the first During the second inspection, image information of the photovoltaic panel surface is collected to generate grayscale values ​​of the photovoltaic panel surface. Indicates the first The second inspection to the first The time interval between collecting image information of the photovoltaic panel surface during the second inspection and generating grayscale values ​​of the photovoltaic panel surface. ; The influence rate calculation unit is used to calculate the influence rate of the increase in gray value on the surface of the photovoltaic panel. Where K represents the rate of increase in the grayscale value of the photovoltaic panel surface. This represents the total number of times the image information of the photovoltaic panel surface was collected to generate the grayscale value of the photovoltaic panel surface. and They represent the first Second and third The decomposed angle of the parallel environmental wind direction in the second monitoring, where n represents the angle of the first monitoring. Next to Total number of monitoring sessions during the inspection period , , and Indicates the training coefficient value. , and They represent the first The air dust concentration, ambient wind speed, and ambient wind direction were monitored in this instance. , and They represent the first The air dust concentration, ambient wind speed, and ambient wind direction were monitored in this instance. .

7. The photovoltaic panel inspection system based on artificial intelligence according to claim 6, characterized in that: The model building module also includes a dust accumulation prediction model unit and a rainwater washing model unit; The dust accumulation prediction model unit is used to establish a dust accumulation prediction model on the surface of photovoltaic panels. ,in, The function representing the grayscale value of the photovoltaic panel surface with respect to time. This indicates the current grayscale value of the photovoltaic panel surface. Indicates the current time. The function representing the rate of dust accumulation on the surface of a photovoltaic panel with respect to time; The rainwater washing model unit is used to establish a rainwater washing model. ,in, This represents the surface grayscale value of the photovoltaic panel after each rainfall event. and This indicates the adjustment coefficient value.

8. The photovoltaic panel inspection system based on artificial intelligence according to claim 7, characterized in that: The early warning module also includes an early warning unit and a calibration unit; The early warning unit is used to set an early warning value for the grayness of the photovoltaic panel surface; if the amount of dust accumulated on the photovoltaic panel surface exceeds the early warning value, an artificial intelligence early warning is issued, and early warning information is output to prompt staff to perform artificial intelligence cleaning on the photovoltaic panel surface within the early warning time. The calibration unit is used to perform artificial intelligence calibration on the early warning information, output calibration information, and prompt staff to perform artificial intelligence cleaning on the surface of the photovoltaic panel within the calibration time.

Images