A real-time detection system for offshore photovoltaics

Abstract

The present invention relates to the field of photovoltaic detection, and in particular to a real-time offshore photovoltaic detection system. The present invention obtains aerial images to calibrate moving targets in each image frame, determines the relative position relationship between each moving target and the angle between corresponding moving vectors, so as to verify whether there is a tendency of migration and aggregation; obtains image data of a photovoltaic panel area to predict a moving path of a moving target based on an image frame, determines the maximum distribution distance of the moving target relative to the moving path, so as to determine an inspection area for the photovoltaic panel; performs mobile acquisition on the inspection area, obtains detailed image data of the photovoltaic panel area, extracts panel surface texture features, determines a corrosion area of ​​the photovoltaic panel, and calls the electric energy conversion data of the photovoltaic panel in the corrosion area to determine whether to issue an early warning signal. The present invention is suitable for the inspection of photovoltaic panels laid over a large area, improves the inspection efficiency and reduces the impact of bird droppings or scratches on the photovoltaic panel while ensuring reliability.

Description

Technical Field

[0001] The present invention relates to the field of photovoltaic detection, and in particular to an offshore photovoltaic real-time detection system. Background Art

[0002] With the growing demand for renewable energy, offshore photovoltaics, as an efficient and clean form of energy, has received widespread attention and application. There are currently two main types of offshore photovoltaics: pile-based fixed photovoltaic power stations and floating photovoltaic power stations, which can be applied to various application scenarios. However, since the offshore photovoltaic deployment area may be offshore, and the deployment area is large and the regional span is wide, the detection of offshore photovoltaics is crucial, and relevant detection methods have emerged.

[0003] For example, Chinese patent publication number: CN117963093A, discloses an offshore floating photovoltaic power generation detection platform and monitoring device, including a mesh floating structure composed of multiple HDPE floats and a photovoltaic device arranged on the floating structure, the photovoltaic device is arranged on an adjusting bracket, and a HDPE float is arranged at the bottom of the adjusting bracket. The inclination angle of the photovoltaic device is changed by adjusting the bracket to face the direction of light; the HDPE floats are connected by cables to form a mesh structure, and the wave load acts on the HDPE floats and is borne by the mesh structure composed of cables; a flexible support is arranged between the HDPE float and the adjusting bracket, and the flexible support provides longitudinal elastic support and torsional elastic support, which can suppress the transmission of wave load from the HDPE float to the photovoltaic device.

[0004] However, there are still the following problems in the prior art:

[0005] In actual situations, for large-scale offshore photovoltaic installations, birds may stop at photovoltaic panels during their concentrated migration, producing feces that may corrode the panels and affect their power generation. The efficiency of traversal inspections is too low for large-area photovoltaic panels. Summary of the invention

[0006] To this end, the present invention provides a real-time detection system for offshore photovoltaics to overcome the problem in the prior art that, for large-area offshore photovoltaics, birds may stay on the photovoltaic panels during concentrated migration, and the resulting feces may corrode the photovoltaic panels and may affect the power generation of the photovoltaic panels, and the efficiency of traversal inspections for large-area photovoltaic panels is too low.

[0007] To achieve the above object, the present invention provides an offshore photovoltaic real-time detection system, which comprises:

[0008] An image acquisition module, which includes a plurality of mobile inspection units for acquiring image data of the photovoltaic panel area and an image acquisition unit arranged in each photovoltaic panel area for acquiring aerial images;

[0009] A feature analyzer connected to the image acquisition module is used to calibrate the moving targets of each image frame based on the aerial image, determine the relative position relationship between the moving targets and the angle between the corresponding moving vectors, so as to verify whether there is a migration aggregation tendency;

[0010] a trigger analyzer connected to the feature analyzer, for predicting the moving path of the moving target based on the image frame in response to the verification result of the feature analyzer, and determining the maximum distribution distance of the moving target relative to the moving path, so as to determine the inspection area for the photovoltaic panel;

[0011] A controller, which is connected to the image acquisition module and the trigger analyzer respectively, and is used to control the mobile inspection unit to move based on the inspection area to obtain detailed image data of the photovoltaic panel area;

[0012] an analysis and warning device connected to the controller, for extracting panel surface texture features based on the detail image data, determining the corrosion area of ​​the photovoltaic panel, calling the power conversion data of the photovoltaic panel in the corrosion area, and determining whether to issue a warning signal for the photovoltaic panel according to the power conversion data;

[0013] The board surface texture features include corrosion feature contours and scratches.

[0014] Furthermore, the feature analyzer is used to determine the relative position relationship between the moving targets and the angle between the corresponding moving vectors, including:

[0015] Coordinate data for identifying moving objects in each video frame;

[0016] To determine the average distance between each moving target and the nearest moving target, as well as the number of moving targets;

[0017] To construct the motion vector of each moving target and determine the average angle between each moving target and the motion vectors corresponding to the remaining moving targets;

[0018] The endpoints of the moving vector are the moving target coordinate points, and the vector direction is the moving direction of the moving target.

[0019] Furthermore, the feature analyzer is used to verify whether there is a migration and aggregation tendency, including:

[0020] If each mobile target meets the migration aggregation conditions, then the existence of migration aggregation tendency is verified;

[0021] Among them, the migration aggregation conditions include that the average distance between each moving target and the nearest moving target is less than or equal to the average distance threshold, the number of moving targets is greater than or equal to the number threshold, and the average angle between each moving target and the corresponding moving vectors of the remaining moving targets is less than the average angle threshold.

[0022] Further, the trigger analyzer is used to respond to the verification result of the feature analyzer including:

[0023] If the moving targets have a tendency to migrate and aggregate, the moving paths of the moving targets are predicted based on the image frames, and the maximum distribution distance of the moving targets relative to the moving paths is determined to determine the inspection area for the photovoltaic panels.

[0024] Furthermore, the trigger analyzer is used to predict the moving path of the moving target based on the image frame, including:

[0025] To solve the vector sum of the motion vectors corresponding to each moving target and obtain a reference vector;

[0026] It is used to solve the projection of the reference vector on the horizontal plane where the photovoltaic panel is located, and determine the projection as the moving path.

[0027] Furthermore, the trigger analyzer is used to determine the maximum distribution distance of the moving target relative to the moving path, including:

[0028] Used to determine a reference direction perpendicular to the moving path on the horizontal plane where the photovoltaic panel is located;

[0029] It is used to construct the farthest distance of each moving target in the reference direction to obtain the maximum distribution distance.

[0030] Furthermore, the trigger analyzer is used to determine the inspection area for the photovoltaic panel, including:

[0031] Constructing a rectangular area covering the moving path along the moving path, and using the rectangular area as the inspection area;

[0032] The center line of the rectangular area is the moving path, and the width of the rectangular area is the maximum distribution distance.

[0033] Furthermore, the analysis and early warning device is used to determine the corrosion area of ​​the photovoltaic panel, including:

[0034] Determine the area of ​​the corrosion feature profile and the number of scratches according to the plate surface texture characteristics;

[0035] The sum of the ratio of the area of ​​the corrosion feature profile to the area threshold and the ratio of the number of scratches to the scratch number threshold is used as a characterization value of the corrosion tendency degree;

[0036] If there is any photovoltaic panel region whose corrosion tendency degree characterization value is greater than or equal to the corrosion degree characterization threshold, the photovoltaic panel region is determined as the corrosion region.

[0037] Furthermore, the analysis and warning device is used to determine whether to issue a warning signal for the photovoltaic panel based on the electric energy conversion data, including:

[0038] Determining the power conversion efficiency of the photovoltaic panel according to the power conversion data;

[0039] Used to construct a time domain curve of the electric energy conversion efficiency within a predetermined time;

[0040] Used to calculate the variance of each peak value in the time domain curve segment corresponding to the electric energy conversion efficiency in each sub-time domain segment;

[0041] If the variance is greater than or equal to the variance threshold, it is determined to issue a warning signal for the photovoltaic panel.

[0042] Furthermore, the warning signal includes the position coordinates of the corresponding photovoltaic panel in the corrosion area.

[0043] Compared with the prior art, the present invention acquires aerial images, calibrates moving targets, determines the relative position relationship between moving targets and the angle between corresponding moving vectors, verifies whether there is a tendency of migration and aggregation, and subsequently predicts the moving path of the moving target based on the image frame, determines the maximum distribution distance of the moving target relative to the moving path, determines the inspection area for the photovoltaic panel, adaptively controls the mobile inspection unit to inspect the inspection area, collects detailed image data, determines the corrosion area, and determines whether to issue a warning signal for the photovoltaic panel based on the electric energy conversion data. Therefore, the present invention is suitable for the inspection of photovoltaic panels laid over a large area, improves the inspection efficiency and reduces the impact of bird droppings or scratches on the photovoltaic panels while ensuring reliability.

[0044] In particular, the present invention sets an image acquisition module to collect aerial images, and determines whether there is a migration and aggregation tendency based on the aerial images through a feature analyzer. In actual situations, birds have concentrated migration. Under concentrated migration, they may stay on the photovoltaic panels that are passed through a large area. The movement and trampling of birds will form scratches on the photovoltaic panels, and they may also produce feces on the photovoltaic panels they pass through. Based on this, the present invention considers collecting the relative position relationship of the moving target and the angle between the moving vectors based on the aerial images, and considers whether there is concentrated migration of birds in the aerial images, so as to provide data support for the subsequent controller to control the mobile inspection unit. For photovoltaic panels laid over a large area, it is possible to select a suitable inspection area based on actual conditions, improve the inspection efficiency while ensuring reliability, and reduce the impact of bird feces or scratches on the photovoltaic panels.

[0045] In particular, the present invention sets a trigger analyzer, which responds to the feature analyzer and can predict the moving path of the moving target based on the image frame to construct an inspection area for the photovoltaic panels. In actual situations, when inspecting photovoltaic panels laid over a large area, drone inspection is adopted. Since it is necessary to obtain images with higher precision, it is necessary to control the flight speed and altitude of the drone to obtain images with higher precision, which is convenient for analyzing the specific features on the photovoltaic panels. In this case, the area that can be collected in a single flight is relatively small, and the efficiency is low when inspecting large-area photovoltaic panels. Based on this, the present invention considers controlling the mobile inspection unit to inspect when birds have a tendency to migrate and gather, which is easy to affect the photovoltaic panels, and adaptively determining the inspection area where potential abnormalities are likely to exist, so that the mobile inspection unit can adaptively inspect and detect abnormalities in time, thereby improving the inspection efficiency when inspecting photovoltaic panels laid over a large area and reducing the impact of bird droppings or scratches on photovoltaic panels.

[0046] In particular, the present invention obtains detailed image data of the photovoltaic panel area, and then analyzes the photovoltaic panel based on the detailed image data to determine the corrosion area of ​​the photovoltaic panel, and specifically analyzes the power conversion efficiency of the corrosion area. In actual situations, when feces are attached, due to the attachment thickness or the properties of the feces themselves or the distribution of feces on the photovoltaic panel, the light transmittance of the feces is unstable, which will affect the photovoltaic panel's absorption of light energy, and thus cause slight fluctuations in the power conversion efficiency of the photovoltaic panel. The present invention takes this factor into consideration and performs targeted analysis of the photovoltaic panel, thereby promptly issuing warnings to abnormal photovoltaic panels to facilitate timely processing and reduce the impact of bird feces or scratches on the photovoltaic panel. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a functional module diagram of an offshore photovoltaic real-time detection system according to an embodiment of the invention;

[0048] Figure 2 A logical decision diagram for verifying whether there is a tendency of migration and aggregation for an embodiment of the invention;

[0049] Figure 3 A logical decision diagram for determining corrosion areas of a photovoltaic panel for an embodiment of the invention;

[0050] Figure 4 This is a logic decision diagram for whether to issue a warning signal for a photovoltaic panel according to an embodiment of the invention. DETAILED DESCRIPTION

[0051] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0052] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0053] It should be noted that, in the description of the present invention, the terms such as "on" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be understood as a limitation on the present invention.

[0054] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the term "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0055] See also Figure 1 As shown, Figure 1 : is a functional module diagram of an offshore photovoltaic real-time detection system according to an embodiment of the present invention. The offshore photovoltaic real-time detection system according to an embodiment of the present invention comprises:

[0056] An image acquisition module, which includes a plurality of mobile inspection units for acquiring image data of the photovoltaic panel area and an image acquisition unit arranged in each photovoltaic panel area for acquiring aerial images;

[0057] A feature analyzer connected to the image acquisition module is used to calibrate the moving targets of each image frame based on the aerial image, determine the relative position relationship between the moving targets and the angle between the corresponding moving vectors, so as to verify whether there is a migration aggregation tendency;

[0058] a trigger analyzer connected to the feature analyzer, for predicting the moving path of the moving target based on the image frame in response to the verification result of the feature analyzer, and determining the maximum distribution distance of the moving target relative to the moving path, so as to determine the inspection area for the photovoltaic panel;

[0059] A controller, which is connected to the image acquisition module and the trigger analyzer respectively, and is used to control the mobile inspection unit to move based on the inspection area to obtain detailed image data of the photovoltaic panel area;

[0060] an analysis and warning device connected to the controller, for extracting panel surface texture features based on the detail image data, determining the corrosion area of ​​the photovoltaic panel, calling the power conversion data of the photovoltaic panel in the corrosion area, and determining whether to issue a warning signal for the photovoltaic panel according to the power conversion data;

[0061] The board surface texture features include corrosion feature contours and scratches.

[0062] Specifically, there is no limitation on the specific structure of the mobile inspection unit, and it only needs to have the function of mobile collection of image data of the photovoltaic panel area. For example, a high-definition camera equipped with a drone can be used to perform mobile collection of the photovoltaic panel area.

[0063] Specifically, there is no limitation on the specific structure of the image acquisition unit, and it only needs to be able to capture aerial images corresponding to the photovoltaic panels. Any photographic device in the prior art that meets the application environment can be used. In some possible implementations, a bracket with sufficient strength and stability is built at the edge of the photovoltaic panel array, and an image acquisition unit is set on the bracket to ensure that the entire aerial area can be photographed. Of course, the image acquisition unit can use an optical image stabilization photography device to reduce the impact of shaking on the camera, which will not be repeated here.

[0064] It is understandable that a single image acquisition unit can acquire aerial images of a relatively large area, which will not be elaborated herein.

[0065] It is understandable that there is no limitation on the division method of photovoltaic panel areas, the area difference ratio of each photovoltaic panel area should be small, controlled within 0.1-0.3, and the difference ratio of two values ​​is the ratio of the absolute difference of the two values ​​to the corresponding mean of the two values.

[0066] Specifically, the moving target refers to birds. During the migration of birds, due to the long distance of the migration path, the birds may stop and rest on the photovoltaic panels. When the birds stop on the photovoltaic panels, the frequent trampling of their claws may cause wear on the surface of the photovoltaic panels. In particular, when the areas where the birds stay are too concentrated, it is easier to scratch the glass cover of the photovoltaic panels, accelerating the wear and aging of the photovoltaic panels in the corresponding areas. The fine scratches accumulated over a long period of time will reduce its light transmittance and affect the photovoltaic panels' absorption efficiency of solar energy. By analyzing the movement trajectories and distribution of birds, inspection areas with a tendency for birds to accumulate are determined, the damage to the photovoltaic panels in the inspection areas is analyzed to identify corrosion areas, and verification and early warning are carried out on the corrosion areas.

[0067] Specifically, there is no limitation on the specific structures of the feature analyzer, trigger analyzer, controller and analysis alarm, and they themselves or each unit therein can be composed of logic components or a combination of logic components, and the logic components include a field programmable processor, a computer or a microprocessor in a computer.

[0068] Specifically, the feature analyzer is used to determine the relative position relationship between each moving target and the angle between each corresponding moving vector, including:

[0069] Coordinate data for identifying moving objects in each video frame;

[0070] To determine the average distance between each moving target and the nearest moving target, as well as the number of moving targets;

[0071] To construct the motion vector of each moving target and determine the average angle between each moving target and the motion vectors corresponding to the remaining moving targets;

[0072] The endpoints of the moving vector are the moving target coordinate points, and the vector direction is the moving direction of the moving target.

[0073] Specifically, there is no limitation on the method of determining the moving target. An image segmentation algorithm can be used to identify the contour of the moving target and then determine the moving target. In implementation, the coordinate data of the moving target is the coordinate value of the center of the moving target contour. The reference coordinate system of the coordinate value can be established in the aerial image taken by the image acquisition unit. Of course, other methods can also be used, which will not be repeated here.

[0074] Specifically, see Figure 2 As shown, it is a logical decision diagram for verifying whether there is a migration aggregation tendency in an embodiment of the present invention. The feature analyzer is used to verify whether there is a migration aggregation tendency, including:

[0075] If each mobile target meets the migration aggregation conditions, then the existence of migration aggregation tendency is verified;

[0076] If each moving target does not meet the migration aggregation conditions, it is verified that there is no migration aggregation tendency.

[0077] Among them, the migration aggregation conditions include that the average distance between each moving target and the nearest moving target is less than or equal to the average distance threshold, the number of moving targets is greater than or equal to the number threshold, and the average angle between each moving target and the corresponding moving vectors of the remaining moving targets is less than the average angle threshold.

[0078] In this embodiment, the purpose of setting the average distance threshold between each moving target and the nearest moving target, the number threshold of the moving targets, and the average angle threshold between each moving target and the moving vectors corresponding to the remaining moving targets is to characterize that the moving situation of each moving target is a situation with a high degree of migration aggregation;

[0079] The average distance threshold between each mobile target and the nearest mobile target, the number threshold of mobile targets, and the average angle threshold between each mobile target and the mobile vectors corresponding to the remaining mobile targets are determined respectively by the average value of the average distance between each mobile target and the nearest mobile target, the average value of the number of mobile targets, and the average angle threshold between each mobile target and the mobile vectors corresponding to the remaining mobile targets;

[0080] Those skilled in the art can determine the average value of the average distance between each moving target and the nearest moving target, the average value of the number of moving targets, and the average value of the average angle between each moving target and the moving vectors corresponding to the remaining moving targets by screening out relevant historical aerial images of the migration and accumulation of moving targets in the same sea area. According to the purpose of setting the above three thresholds in this embodiment, the threshold value of the average distance between each moving target and the nearest moving target is determined between 0.8 times and 0.9 times the average value of the average distance between each moving target and the nearest moving target, the threshold value of the number of moving targets is determined between 1.1 times and 1.2 times the average value of the number of mobile targets, and the threshold value of the average angle between each moving target and the moving vectors corresponding to the remaining moving targets is determined between 0.85 times and 0.9 times the average value of the average angle between each moving target and the moving vectors corresponding to the remaining moving targets.

[0081] Specifically, the present invention sets an image acquisition module to collect aerial images, and determines whether there is a migration and aggregation tendency based on the aerial images through a feature analyzer. In actual situations, birds have concentrated migration. Under concentrated migration, they may stay on the photovoltaic panels that are passed through a large area. The movement and trampling of birds will form scratches on the photovoltaic panels, and they may also produce feces on the photovoltaic panels they pass through. Based on this, the present invention considers collecting the relative position relationship of the moving target and the angle between the moving vectors based on the aerial images, and considers whether there is concentrated migration of birds in the aerial images, so as to provide data support for the subsequent controller to control the mobile inspection unit. For photovoltaic panels laid over a large area, it is possible to select a suitable inspection area based on actual conditions, improve the inspection efficiency while ensuring reliability, and reduce the impact of bird feces or scratches on the photovoltaic panels.

[0082] Specifically, the trigger analyzer is used to respond to the verification result of the feature analyzer including:

[0083] If the moving targets have a tendency to migrate and aggregate, the moving paths of the moving targets are predicted based on the image frames, and the maximum distribution distance of the moving targets relative to the moving paths is determined to determine the inspection area for the photovoltaic panels.

[0084] Specifically, the trigger analyzer is used to predict the moving path of the moving target based on the image frame, including:

[0085] To solve the vector sum of the motion vectors corresponding to each moving target and obtain a reference vector;

[0086] It is used to solve the projection of the reference vector on the horizontal plane where the photovoltaic panel is located, and determine the projection as the moving path.

[0087] It can be understood that the movement vectors of each moving target are constructed in the same coordinate system, so the vector sum of the movement vectors can represent the overall movement direction, which will not be elaborated here.

[0088] Specifically, the trigger analyzer is used to determine the maximum distribution distance of the moving target relative to the moving path, including:

[0089] Used to determine a reference direction perpendicular to the moving path on the horizontal plane where the photovoltaic panel is located;

[0090] It is used to construct the farthest distance of each moving target in the reference direction to obtain the maximum distribution distance.

[0091] It can be understood that the mobile targets that have a tendency to migrate and gather usually move in a relatively neat formation when moving, wherein the width of the formation can determine the width of the path range covered by each mobile target during the movement, that is, the maximum distribution distance, and then, the maximum distribution distance is analyzed and detected in more detail, which will not be repeated here.

[0092] Specifically, the trigger analyzer is used to determine the inspection area for the photovoltaic panel, including:

[0093] Constructing a rectangular area covering the moving path along the moving path, and using the rectangular area as the inspection area;

[0094] The center line of the rectangular area is the moving path, and the width of the rectangular area is the maximum distribution distance.

[0095] Specifically, the present invention sets a trigger analyzer, which responds to the feature analyzer and can predict the moving path of the moving target based on the image frame to construct an inspection area for the photovoltaic panels. In actual situations, when inspecting photovoltaic panels laid over a large area, a drone inspection method is adopted. Since it is necessary to obtain images with higher precision, it is necessary to control the flight speed and flight altitude of the drone to obtain images with higher precision, which is convenient for analyzing the specific features on the photovoltaic panels. In this case, the area that can be collected in a single flight is relatively small, and the efficiency is low when inspecting large-area photovoltaic panels. Based on this, the present invention considers controlling the mobile inspection unit to inspect when birds have a tendency to migrate and gather, which is easy to affect the photovoltaic panels, and adaptively determining the inspection area where potential abnormalities are likely to exist, so that the mobile inspection unit can adaptively inspect and detect abnormalities in time, thereby improving the inspection efficiency when inspecting photovoltaic panels laid over a large area and reducing the impact of bird droppings or scratches on photovoltaic panels.

[0096] Specifically, see Figure 3As shown, it is a logical decision diagram for determining the corrosion area of ​​the photovoltaic panel according to an embodiment of the present invention. The analysis and early warning device is used to determine the corrosion area of ​​the photovoltaic panel, including:

[0097] Determine the area of ​​the corrosion feature profile and the number of scratches according to the plate surface texture characteristics;

[0098] The sum of the ratio of the area of ​​the corrosion feature profile to the area threshold and the ratio of the number of scratches to the scratch number threshold is used as a characterization value of the corrosion tendency degree;

[0099] If there is any photovoltaic panel region whose corrosion tendency degree characterization value is greater than or equal to the corrosion degree characterization threshold, the photovoltaic panel region is determined as the corrosion region.

[0100] Specifically, there is no limitation on the specific method of identifying corrosion feature contours and scratches. An algorithm or model that can implement the corresponding function can be pre-trained, and logical components can be imported to implement the corresponding function, which will not be elaborated here.

[0101] Specifically, the corrosion feature profile refers to the range area formed by the attachments existing on the photovoltaic panel, that is, the coverage area of ​​the droppings produced by the birds. Since the attachment of droppings on the photovoltaic panel will have a certain impact on the efficiency of the photovoltaic panel in absorbing light energy, the scratches caused by the bird's claws on the photovoltaic panel will also reduce the light transmittance of the photovoltaic panel. At the same time, the weak acidity of the bird droppings itself will corrode the photovoltaic panel. Combined with marine environmental factors, such as ultraviolet rays and waves, the aging and corrosion of the photovoltaic panel material corresponding to the scratched area will be accelerated. Therefore, in this embodiment, the corrosion tendency degree characterization value is determined based on the area of ​​the corrosion feature profile and the number of scratches to characterize the degree of corrosion tendency of the photovoltaic panel area caused by the behavior of the moving target;

[0102] During implementation, the area threshold of the corrosion feature profile and the scratch number threshold are determined based on the mean area of ​​the corrosion feature profile and the mean scratch number, respectively. By obtaining relevant historical data on the migration and accumulation of moving targets in the same sea area, the area historical data of the corrosion feature profile and the scratch number historical data are called to solve the area mean of the corrosion feature profile and the scratch number mean. Since the purpose of setting the above two thresholds is to characterize the situation where the photovoltaic panel area is affected by the behavior of the moving target and the degree of corrosion tendency is too high, the area threshold of the corrosion feature profile is determined between 0.8 and 0.9 times the area mean of the corrosion feature profile, and the scratch number threshold is determined between 0.85 and 0.95 times the scratch number mean.

[0103] The corrosion degree characterization threshold is selected in the interval [1.12,1.23].

[0104] Specifically, see Figure 4As shown, it is a logic determination diagram of whether to issue a warning signal for a photovoltaic panel in an embodiment of the present invention. The analysis warning device is used to determine whether to issue a warning signal for the photovoltaic panel based on the electric energy conversion data, including:

[0105] Determining the power conversion efficiency of the photovoltaic panel according to the power conversion data;

[0106] Used to construct a time domain curve of the electric energy conversion efficiency within a predetermined time;

[0107] Used to calculate the variance of each peak value in the time domain curve segment corresponding to the electric energy conversion efficiency in each sub-time domain segment;

[0108] If the variance is greater than or equal to the variance threshold, it is determined to issue a warning signal for the photovoltaic panel.

[0109] If the variance is less than the variance threshold, it is determined that there is no need to issue a warning signal for the photovoltaic panel.

[0110] Specifically, the variance threshold can be determined by obtaining relevant data obtained when the photovoltaic panel is operating normally, calling the variance data of each peak value in the time domain curve of the electric energy conversion efficiency in each sub-time domain segment when the photovoltaic panel is operating normally for several times, solving the variance mean, and determining the variance mean as the variance threshold.

[0111] It can be understood that the inverter, as a key device for converting the DC power output by the photovoltaic panel into AC power, records the input DC power and the output AC power. The overall power conversion situation of the photovoltaic panel is determined by obtaining the DC power input and the AC power output of the inverter, that is, the quotient of the input DC power and the output AC power is calculated to evaluate the power conversion efficiency of the corresponding photovoltaic panel. This will not be elaborated here.

[0112] In this embodiment, the time domain curve of the electric energy conversion efficiency is constructed in the following manner, including:

[0113] A rectangular coordinate system is constructed with time as the horizontal axis and power conversion efficiency as the vertical axis;

[0114] Marking the coordinate points of the electric energy conversion efficiency at each moment in the rectangular coordinate system;

[0115] The coordinate points are connected with a smooth curve to obtain a time domain curve of the electric energy conversion efficiency.

[0116] Specifically, there is no limitation on the method of constructing the time domain curve of the power conversion efficiency. For example, the time domain curve can be fitted by using matlab related fitting software, which will not be described in detail.

[0117] Scheduled time, selected within the range [30min, 60min].

[0118] Specifically, the present invention obtains detailed image data of the photovoltaic panel area, and then analyzes the photovoltaic panel based on the detailed image data to determine the corrosion area of ​​the photovoltaic panel, and specifically analyzes the power conversion efficiency of the corrosion area. In actual situations, when feces are attached, due to the attachment thickness or the properties of the feces themselves or the distribution of feces on the photovoltaic panel, the light transmittance of the feces is unstable, which will affect the photovoltaic panel's absorption of light energy, and thus cause slight fluctuations in the power conversion efficiency of the photovoltaic panel. The present invention takes this factor into consideration and performs targeted analysis of the photovoltaic panel, thereby promptly issuing warnings to abnormal photovoltaic panels to facilitate timely processing and reduce the impact of bird feces or scratches on the photovoltaic panel.

[0119] Specifically, the warning signal includes the position coordinates of the corresponding photovoltaic panel in the corrosion area.

[0120] Specifically, there is no limitation on the specific structure of the analysis and warning device, which can be composed of logical components to complete basic operations. If a warning signal needs to be sent to a remote end, a corresponding communication module needs to be configured to implement the corresponding function. Technical personnel in this field can configure it by themselves, so it will not be repeated here.

[0121] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. An offshore photovoltaic real-time detection system, characterized in that: include: An image acquisition module, which includes a plurality of mobile inspection units for acquiring image data of the photovoltaic panel area and an image acquisition unit arranged in each photovoltaic panel area for acquiring aerial images; A feature analyzer connected to the image acquisition module is used to calibrate the moving targets of each image frame based on the aerial image, determine the relative position relationship between the moving targets and the angle between the corresponding moving vectors, so as to verify whether there is a migration aggregation tendency; a trigger analyzer connected to the feature analyzer, for predicting the moving path of the moving target based on the image frame in response to the verification result of the feature analyzer, and determining the maximum distribution distance of the moving target relative to the moving path, so as to determine the inspection area for the photovoltaic panel; A controller, which is connected to the image acquisition module and the trigger analyzer respectively, and is used to control the mobile inspection unit to move based on the inspection area to obtain detailed image data of the photovoltaic panel area; an analysis and warning device connected to the controller, for extracting panel surface texture features based on the detail image data, determining the corrosion area of ​​the photovoltaic panel, calling the power conversion data of the photovoltaic panel in the corrosion area, and determining whether to issue a warning signal for the photovoltaic panel according to the power conversion data; Wherein, the board surface texture features include corrosion feature contours and scratches; The feature analyzer is used to determine the relative position relationship between each moving target and the angle between each corresponding moving vector, including: Coordinate data for identifying moving objects in each video frame; To determine the average distance between each moving target and the nearest moving target, as well as the number of moving targets; To construct the motion vector of each moving target and determine the average angle between each moving target and the motion vectors corresponding to the remaining moving targets; Wherein, the endpoint of the moving vector is the moving target coordinate point, and the vector direction is the moving direction of the moving target; The feature analyzer is used to verify whether there is a tendency of migration and aggregation, including: If each mobile target meets the migration aggregation conditions, then the existence of migration aggregation tendency is verified; The migration aggregation condition includes that the average distance between each mobile target and the nearest mobile target is less than or equal to the average distance threshold, the number of mobile targets is greater than or equal to the number threshold, and the average angle between each mobile target and the corresponding mobile vectors of the remaining mobile targets is less than the average angle threshold; The analysis and warning device is used to determine whether to issue a warning signal for the photovoltaic panel based on the electric energy conversion data, including: Determining the power conversion efficiency of the photovoltaic panel according to the power conversion data; Used to construct a time domain curve of the electric energy conversion efficiency within a predetermined time; Used to calculate the variance of each peak value in the time domain curve segment corresponding to the electric energy conversion efficiency in each sub-time domain segment; If the variance is greater than or equal to the variance threshold, it is determined to issue a warning signal for the photovoltaic panel.

2. The offshore photovoltaic real-time detection system according to claim 1, characterized in that: The trigger analyzer is used to respond to the verification result of the feature analyzer, including: If the moving targets have a tendency to migrate and aggregate, the moving paths of the moving targets are predicted based on the image frames, and the maximum distribution distance of the moving targets relative to the moving paths is determined to determine the inspection area for the photovoltaic panels.

3. The offshore photovoltaic real-time detection system according to claim 1 is characterized in that: The trigger analyzer is used to predict the moving path of the moving target based on the image frame, including: To solve the vector sum of the motion vectors corresponding to each moving target and obtain a reference vector; It is used to solve the projection of the reference vector on the horizontal plane where the photovoltaic panel is located, and determine the projection as the moving path.

4. The offshore photovoltaic real-time detection system according to claim 1, characterized in that: The trigger analyzer is used to determine the maximum distribution distance of the moving target relative to the moving path, including: Used to determine a reference direction perpendicular to the moving path on the horizontal plane where the photovoltaic panel is located; It is used to construct the farthest distance of each moving target in the reference direction to obtain the maximum distribution distance.

5. The offshore photovoltaic real-time detection system according to claim 1, characterized in that: The trigger analyzer is used to determine the inspection area for the photovoltaic panel, including: Constructing a rectangular area covering the moving path along the moving path, and using the rectangular area as the inspection area; The center line of the rectangular area is the moving path, and the width of the rectangular area is the maximum distribution distance.

6. The offshore photovoltaic real-time detection system according to claim 1, characterized in that: The analysis and early warning device is used to determine the corrosion area of ​​the photovoltaic panel, including: Determine the area of ​​the corrosion feature profile and the number of scratches according to the plate surface texture characteristics; The sum of the ratio of the area of ​​the corrosion feature profile to the area threshold and the ratio of the number of scratches to the scratch number threshold is used as a characterization value of the corrosion tendency degree; If there is any photovoltaic panel region whose corrosion tendency degree characterization value is greater than or equal to the corrosion degree characterization threshold, the photovoltaic panel region is determined as the corrosion region.

7. The offshore photovoltaic real-time detection system according to claim 1, characterized in that: The early warning signal includes the position coordinates of the corresponding photovoltaic panel in the corrosion area.

Images