Defect position determination method and device, storage medium and electronic device

Abstract

The invention discloses a defect position determination method and device, a storage medium and an electronic device. The method comprises the following steps: extracting multiple pieces of real-time feature information from a dual-light image corresponding to a target plant station; identifying the difference between the sample feature information in a preset defect feature library and the plurality of pieces of real-time feature information to obtain defect information; target coordinates corresponding to the defect information in the photovoltaic array vector diagram are determined, and the target coordinates are defect positions where defects exist in the positioned target station. By adopting the technical scheme, the problem that the defect position of the photovoltaic panel cannot be accurately positioned is solved. Furthermore, multiple pieces of real-time feature information are extracted and compared with the preset defect feature library, so that the defect position in the photovoltaic array is determined, and the positioning precision of the defect position of the photovoltaic array is improved.

Description

Technical Field

[0001] The present application relates to the technical field of photovoltaic defect detection, and in particular, to a method and device for determining a defect position, a storage medium, and an electronic device. Background Art

[0002] With the rapid development of solar energy technology, the installed capacity of photovoltaic power stations continues to grow. Large-area photovoltaic module arrays are prone to surface contamination, junction box failure, cell burnout and other defects during long-term operation, which have a great impact on the power generation efficiency of the power station. At present, the photovoltaic panel defect positioning technology mainly uses a dual-light camera mounted on a drone for shooting, and the defect situation is reflected by the color difference of the infrared image and the coordinates of the defect position are recorded. However, due to the high flying altitude of the drone, the system has a deviation in defect positioning, and cannot accurately locate the defect position and correspond to the photovoltaic panel number. To solve this problem, GPS positioning (Global Positioning System) and image recognition methods are used. After the inspection is completed, the defect recognition algorithm is combined with the GPS positioning system to measure and locate the photovoltaic panel and the defect position. However, relying solely on coordinate information cannot display the layout and defect distribution of the photovoltaic array in real time, and cannot help operation and maintenance personnel quickly locate and deal with defects, which has become one of the current challenges.

[0003] With regard to the problem in related technologies that it is impossible to quickly and accurately locate the defect positions of photovoltaic panels, no effective solution has been proposed so far.

[0004] Therefore, it is necessary to improve the related technology to overcome the above-mentioned defects in the related technology. Summary of the invention

[0005] The embodiments of the present application provide a method and device for determining a defect position, a storage medium, and an electronic device, so as to at least solve the problem that the defect position of a photovoltaic panel cannot be quickly and accurately located.

[0006] According to one aspect of an embodiment of the present application, a method for determining a defect position is provided, comprising: extracting multiple real-time feature information from a dual-light image corresponding to a target plant; identifying the difference between sample feature information in a preset defect feature library and multiple real-time feature information to obtain defect information; determining the target coordinates corresponding to the defect information in a photovoltaic array vector diagram, wherein the target coordinates are the defect position of the defect in the located target plant.

[0007] In an exemplary embodiment, before extracting multiple real-time feature information from the dual-light image corresponding to the target plant, the method further includes: acquiring electronic map data corresponding to the target station; determining a photovoltaic array vector diagram of the target station based on the electronic map data; setting a monitoring operation trajectory of the unmanned detection equipment according to the photovoltaic array vector diagram, and receiving a real-time image of the unmanned detection equipment; wherein the real-time image is an image collected from the operation monitoring operation trajectory, and the real-time image includes at least: a real-time infrared area image of the target station, and a real-time visible light image of the target station; and determining a dual-light image based on the real-time image.

[0008] In an exemplary embodiment, after determining the target coordinates corresponding to the photovoltaic array vector diagram of the defect information, the above method also includes: adding attribute data to the target coordinates, wherein the attribute data at least includes: the photovoltaic material attribute data corresponding to the associated detailed defect content and the target coordinates; fusing the attribute data, the photovoltaic array vector diagram, the target coordinates and the electronic map data corresponding to the target site to generate a defect distribution diagram.

[0009] In an exemplary embodiment, a photovoltaic array vector diagram of a target station is determined based on electronic map data, including: obtaining a photovoltaic power station logic diagram corresponding to the target station, wherein the photovoltaic power station logic diagram contains the distribution positions of all photovoltaic arrays in the target station; extracting array data from the photovoltaic power station logic diagram to obtain row and column numbers of each photovoltaic array and array vertex coordinates of each photovoltaic array; when a management object has identified a vector base point of a map interface corresponding to the electronic map data, the row and column numbers and the array vertex coordinates are identified on the map interface, and the map interface is extracted using a preset extraction application to obtain the photovoltaic array vector diagram of the target station.

[0010] In an exemplary embodiment, multiple real-time feature information is extracted from a dual-light image corresponding to a target plant, including: performing image preprocessing on the dual-light image to obtain a first image; performing feature extraction on the first image using a preset deep learning algorithm to obtain multiple real-time feature information, wherein the real-time feature information includes at least one of the following: crack size of the photovoltaic array, crack shape of the photovoltaic array, and crack direction of the photovoltaic array.

[0011] In an exemplary embodiment, the difference between sample feature information in a preset defect feature library and multiple real-time feature information is identified to obtain defect information, including: determining the defect type and defect position corresponding to the multiple real-time feature information respectively; wherein the defect type includes at least one of the following: photovoltaic array hot spot, photovoltaic array multiple spots, photovoltaic array missing, photovoltaic array fragmentation, photovoltaic array abnormality, photovoltaic array surface abnormality; using the sample feature information corresponding to the sample type and sample position to screen the defect type and defect position; and determining the defect information according to the screening result.

[0012] According to another aspect of an embodiment of the present application, a device for determining a defect position is also provided, including: an extraction module for extracting multiple real-time feature information from a dual-light image corresponding to a target plant; an identification module for identifying the difference between sample feature information in a preset defect feature library and multiple real-time feature information to obtain defect information; a determination module for determining the target coordinates corresponding to the defect information in a photovoltaic array vector diagram, wherein the target coordinates are the defect position of the defect in the located target plant.

[0013] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned defect location determination method when running.

[0014] According to another aspect of an embodiment of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the defect location determination method through the computer program.

[0015] According to another aspect of the embodiments of the present application, a computer program product is provided, including a computer program, and the method for determining the defect location when the computer program is executed by a processor.

[0016] Through this application, multiple real-time feature information is extracted from the dual-light image corresponding to the target plant; the difference between the sample feature information in the preset defect feature library and the multiple real-time feature information is identified to obtain the defect information; the target coordinates corresponding to the defect information in the photovoltaic array vector diagram are determined, wherein the target coordinates are the defect positions of the defects in the located target plant. The above technical solution solves the problem of being unable to quickly and accurately locate the defect positions of photovoltaic panels. Furthermore, by extracting multiple real-time feature information and comparing them with the preset defect feature library, the positions of the defects in the photovoltaic array are determined, thereby improving the positioning accuracy of the defect positions of the photovoltaic array. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0019] Figure 1It is a hardware structure block diagram of a computer terminal according to a method for determining a defect position in an embodiment of the present application;

[0020] Figure 2 is a flow chart of a method for determining a defect location according to an embodiment of the present application;

[0021] Figure 3 is a schematic diagram of the structure of a photovoltaic defect positioning system based on the combination of defect positioning and vector data according to an embodiment of the present application;

[0022] Figure 4 is a flow chart of a photovoltaic defect location method based on the combination of defect location and vector data according to an embodiment of the present application;

[0023] Figure 5 This is an effect diagram of superimposing defect information and a vector layout diagram of a photovoltaic array according to an embodiment of the present application;

[0024] Figure 6 It is a structural block diagram of determining a defect location according to an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0026] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0027] The method embodiments provided in the embodiments of the present application can be executed in a computer terminal, a mobile terminal or a similar computing device. Taking running on a computer terminal as an example, Figure 1 1 is a hardware structure block diagram of a computer terminal according to a method for determining a defect location in an embodiment of the present application. Figure 1 As shown, the computer terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor (Central Processing Unit, CPU) or a programmable logic device (Field Programmable Gate Array, FPGA)) and a memory 104 for storing data, wherein the above-mentioned computer terminal may also include a transmission device 106 and an input and output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above-mentioned computer terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0028] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the method for determining the defect location in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories can be connected to the computer terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0029] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of a computer terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0030] In this embodiment, a method for determining a defect position is provided. Figure 2 is a flow chart of a method for determining a defect location according to an embodiment of the present application, such as Figure 2 As shown, the process includes the following steps S202-S206:

[0031] Step S202: extracting a plurality of real-time feature information from the bi-optical image corresponding to the target plant;

[0032] Step S204: identifying the difference between the sample feature information in the preset defect feature library and the plurality of real-time feature information to obtain defect information;

[0033] Step S206: determining the target coordinates corresponding to the defect information in the photovoltaic array vector diagram, wherein the target coordinates are defect positions where defects exist in the located target station.

[0034] Through the above steps, multiple real-time feature information is extracted from the dual-light image corresponding to the target plant; the difference between the sample feature information in the preset defect feature library and the multiple real-time feature information is identified to obtain defect information; the target coordinates corresponding to the defect information in the photovoltaic array vector diagram are determined, wherein the target coordinates are the defect positions of the defects in the located target plant. The above technical solution solves the problem of being unable to quickly and accurately locate the defect positions of photovoltaic panels. Furthermore, by extracting multiple real-time feature information and comparing them with the preset defect feature library, the positions of defects in the photovoltaic array are determined, thereby improving the positioning accuracy of the defect positions of the photovoltaic array.

[0035] In an exemplary embodiment, before extracting multiple real-time feature information from the dual-light image corresponding to the target plant, the method further includes: acquiring electronic map data corresponding to the target station; determining a photovoltaic array vector diagram of the target station based on the electronic map data; setting a monitoring operation trajectory of the unmanned detection equipment according to the photovoltaic array vector diagram, and receiving a real-time image of the unmanned detection equipment; wherein the real-time image is an image collected from the operation monitoring operation trajectory, and the real-time image includes at least: a real-time infrared area image of the target station, and a real-time visible light image of the target station; and determining a dual-light image based on the real-time image.

[0036] It is understandable that by accurately obtaining the geographical location, terrain and other information of the target station through electronic map data, and determining the photovoltaic array vector map through the obtained information, the layout and structure of the photovoltaic array can be clearly understood, providing a basis for setting the monitoring and operation trajectory of unmanned detection equipment, ensuring that the unmanned detection equipment conducts comprehensive and systematic monitoring of the target station, and promptly discovers problems and takes corresponding measures. By acquiring real-time images, the situation of the target station can be monitored in real time, abnormal situations can be discovered in time, and the normal operation of the photovoltaic array can be guaranteed. By acquiring dual-light images, the information of infrared and visible light images can be comprehensively utilized to improve the accuracy and reliability of monitoring and ensure the safe and stable operation of the target station.

[0037] In an exemplary embodiment, after determining the target coordinates corresponding to the photovoltaic array vector diagram of the defect information, the above method also includes: adding attribute data to the target coordinates, wherein the attribute data at least includes: the photovoltaic material attribute data corresponding to the associated detailed defect content and the target coordinates; fusing the attribute data, the photovoltaic array vector diagram, the target coordinates and the electronic map data corresponding to the target site to generate a defect distribution diagram.

[0038] Optionally, after determining the target coordinates of a defect in the photovoltaic array vector diagram, attribute data is added according to the coordinates, including the details of the defect type (such as the defect type is component damage and the severity is severe) and the photovoltaic material attribute data corresponding to the target coordinates (such as the model is type A and the production date is 2020), and then these data are fused together to generate a defect distribution map. According to the defect distribution map, the defects are classified into categories 1, 2, and 3, and then stratified according to the defect categories, and each layer is set to represent the severity level of the defect. The first layer represents the most serious defects, the second layer represents the second most serious defects, and the third layer represents relatively less serious defects. Different maintenance strategies are adopted for different levels of severity. First, for the first layer of serious defects, maintenance is given priority to ensure the normal operation of the photovoltaic system. These serious defects may have a serious impact on the performance of the system, so they must be repaired in time. For the second and third layers of less serious defects, periodic maintenance is performed. Although these defects do not immediately affect the performance of the system, they may cause system failures if accumulated over a long period of time. Therefore, regular inspection and maintenance of these defects are necessary to ensure the long-term stable operation of the system. Through such a process, the efficiency and accuracy of photovoltaic array maintenance can be improved to ensure the normal operation of the photovoltaic system. Timely repair of serious defects and regular maintenance of minor defects can effectively prevent system failures, extend the service life of the system, and ensure the stability and reliability of the photovoltaic system.

[0039] In an exemplary embodiment, a photovoltaic array vector diagram of a target station is determined based on electronic map data, including: obtaining a photovoltaic power station logic diagram corresponding to the target station, wherein the photovoltaic power station logic diagram contains the distribution positions of all photovoltaic arrays in the target station; extracting array data from the photovoltaic power station logic diagram to obtain row and column numbers of each photovoltaic array and array vertex coordinates of each photovoltaic array; when a management object has identified a vector base point of a map interface corresponding to the electronic map data, the row and column numbers and the array vertex coordinates are identified on the map interface, and the map interface is extracted using a preset extraction application to obtain the photovoltaic array vector diagram of the target station.

[0040] It can be understood that in the above embodiment, the row and column numbers of the photovoltaic array and the array vertex coordinates of each photovoltaic array are generated in order to determine the layout and position of the photovoltaic panel. First, the number of rows and columns of the photovoltaic array is determined, and the size of the array is determined according to the size of the photovoltaic panel and the installation requirements. Then the layout of the photovoltaic array is determined, including the direction and tilt angle of the array. According to the determined number of rows and columns, a unique row and column number is assigned to each photovoltaic panel of the photovoltaic array. For example, the photovoltaic panel in the first row and the first column can be numbered as (1,1), the photovoltaic panel in the first row and the second column can be numbered as (1,2), and so on. Next, the coordinates of the four vertices of the photovoltaic array are determined, that is, the specific position and size of the photovoltaic panel are determined. The coordinates of the four vertices are determined according to the starting point and the size of the photovoltaic module. For example, the coordinates of the upper left vertex (0,0) are used as the coordinates of the starting point, that is, the origin of the photovoltaic array, and the size of the photovoltaic module is 100cm x 50cm. The coordinates of the four vertices can be determined according to the size and position of the photovoltaic array. The coordinates of the upper left vertex are (0,0), the coordinates of the upper right vertex are (0+100,0)=(100,0), the coordinates of the lower left vertex are (0,0+50)=(0,50), and the coordinates of the lower right vertex are (0+100,0+50=(100,50). Finally, according to the determined row and column numbers of the photovoltaic array and the coordinates of the four vertices, the row and column numbers and array vertex coordinates are marked on the map interface of the electronic map data. The map interface is extracted using the preset photovoltaic array vector graphics extraction application. This application can identify the row and column numbers and array vertex coordinates marked on the map interface and extract them based on this information. Finally, the photovoltaic array vector diagram of the target station is obtained through the processing of this application. Through the above method, all arrays of the photovoltaic power station logic diagram can be automatically detected and numbered quickly and accurately, and the vector diagram of the photovoltaic array can be extracted, thereby improving work efficiency, reducing manual errors, and facilitating subsequent analysis and design work.

[0041] In an exemplary embodiment, multiple real-time feature information is extracted from a dual-light image corresponding to a target plant, including: performing image preprocessing on the dual-light image to obtain a first image; performing feature extraction on the first image using a preset deep learning algorithm to obtain multiple real-time feature information, wherein the real-time feature information includes at least one of the following: crack size of the photovoltaic array, crack shape of the photovoltaic array, and crack direction of the photovoltaic array.

[0042] Optionally, in the above embodiment, the bi-optical image is first preprocessed, including denoising, contrast enhancement, edge detection and other steps, to obtain a first image. Then, the first image is feature extracted using a preset deep learning algorithm. In order to ensure the efficiency of feature extraction, the feature can be automatically processed by constructing a network model corresponding to the deep learning algorithm. Specifically, a convolutional neural network (CNN) model is designed, which includes but is not limited to multiple convolution layers and pooling layers. Through the convolution operation, CNN can extract various features from the input image, such as edges, textures, and colors. Specifically, the convolution kernel is element-wise multiplied with a small area of ​​the input image, and then the results are added to obtain a new value; the convolution kernel is slid to all positions on the entire input image, and the above operation is repeated to obtain a new feature map; the pooling operation is to reduce the size and amount of calculation of the feature map while retaining important feature information. The pooling operation is usually after the convolution operation. It performs an aggregation operation (such as the maximum value or average value) on each small area (such as a 2x2 or 3x3 window) in the feature map to merge the feature values ​​of this area into a single value. After multiple convolution and pooling operations, high-level features are extracted and enter the fully connected layer, which is processed and output by the activation function. When training this CNN model, a dataset of crack images with annotations can be prepared, and the model parameters can be continuously optimized through the back-propagation algorithm so that the model can accurately extract the characteristics of the cracks. Once the model training is completed, it can be applied to the dual-light images corresponding to the target plant station to extract the characteristic information such as the size, shape and direction of the crack in real time. Through the above process, rapid detection and analysis of cracks in photovoltaic arrays can be achieved. At the same time, the use of deep learning algorithms for feature extraction can improve accuracy and efficiency, and realize the automated processing and analysis of large amounts of image data.

[0043] In an exemplary embodiment, the difference between sample feature information in a preset defect feature library and multiple real-time feature information is identified to obtain defect information, including: determining the defect type and defect position corresponding to the multiple real-time feature information respectively; wherein the defect type includes at least one of the following: photovoltaic array hot spot, photovoltaic array multiple spots, photovoltaic array missing, photovoltaic array fragmentation, photovoltaic array abnormality, photovoltaic array surface abnormality; using the sample feature information corresponding to the sample type and sample position to screen the defect type and defect position; and determining the defect information according to the screening result.

[0044] Optionally, in the photovoltaic array monitoring system, the system collects real-time photovoltaic array feature information, including temperature, voltage, current and other parameters, and compares it with the sample feature information in the preset defect feature library. The system finds that the temperature of a photovoltaic component rises abnormally and the current fluctuates greatly. According to the feature library comparison, it determines that the component may have the defect type of photovoltaic array hot spot. Then, the system further analyzes the environment and feature information around the component to determine the specific location of the hot spot defect, helping the staff to accurately locate the problem and repair or replace it. Next, the system screens the sample type and position corresponding to the sample feature information according to the similarity and compares the defect information, and confirms the high similarity as the photovoltaic array hot spot defect, rather than other types of problems. Finally, according to the screening results, it is determined that the component has the defect information of photovoltaic array hot spot, which provides the staff with a clear fault diagnosis and repair direction. The whole process accurately and quickly identifies and locates the problem by comparing the feature information, thereby improving the accuracy and efficiency of troubleshooting.

[0045] Obviously, the embodiments described above are only some embodiments of the present application, not all embodiments. In order to better understand the above method, the above process is described below in conjunction with the embodiments, but it is not intended to limit the technical solutions of the embodiments of the present application, specifically:

[0046] An optional embodiment of the present application provides a photovoltaic defect positioning method based on the combination of defect positioning and vector data, which mainly obtains the coordinate information of the center position of the image screen through rough positioning, and then uses the position, angle and other information of the photovoltaic defect position in the picture to calculate and obtain the precise coordinates of the defect position to achieve final precise positioning, and matches it with the vector electronic map of the photovoltaic field area to achieve a one-to-one correspondence between the fault position and the photovoltaic panel operation number, and helps operation and maintenance personnel to quickly locate and handle defects by displaying the layout and defect distribution of the photovoltaic array in real time.

[0047] Optional, Figure 3 It is a structural schematic diagram of a photovoltaic defect positioning system based on a combination of defect positioning and vector data according to an embodiment of the present application. The above system at least includes: generating an aerial electronic map 31, extracting a photovoltaic array vector map 32, station route and track planning 33, fully autonomous drone inspection 34, photovoltaic image defect recognition 35, and defect information and vector data fusion 36.

[0048] Optionally, generating an aerial electronic map 31 mainly serves as a basis for extracting a high-definition map and vector data of the field, and specifically includes the following steps:

[0049] Step 1: Data collection. Use a drone mounted with an aerial camera to obtain high-definition image data of the target area. During the flight, the drone records the flight trajectory, altitude, speed and other parameters in real time, as well as the captured image data.

[0050] Step 2: Data processing. Specifically includes: data import, importing the data collected by the drone into the map processing software; image stitching, using software to stitch multiple aerial images to form a complete image of the target area; terrain extraction, using digital elevation model (DEM) and other technologies to extract terrain information from the stitched images; map generation, generating an aerial electronic map based on the extracted terrain information and other related data.

[0051] Step 3: Output: The final output aerial electronic map serves as the basis for high-definition map and vector data extraction of the field.

[0052] Optionally, the photovoltaic array vector diagram 32 is extracted, which mainly provides support for the automatic planning of the cruise route when the drone takes infrared images of the power station, and returns the specific number and location information of the array when a fault (such as a hot spot) is detected. It is an important technical basis for realizing the automatic analysis and positioning function of the power station fault. Specifically, the following steps are included:

[0053] Step 1: Import the generated high-resolution aerial electronic map into the self-developed photovoltaic array vector graphics extraction software.

[0054] Step 2: Automatic detection and numbering of arrays based on the logic diagram of photovoltaic power plants. Automatically detect and identify all arrays in the logic diagram of photovoltaic power plants, generate the row and column numbers of photovoltaic arrays and the coordinates of the four vertices, and finally extract the photovoltaic array vector diagram. The above steps provide support for the automatic planning of cruise routes when drones take infrared images of power plants, and for returning the specific number and location information of arrays when faults (such as hot spots) are detected. They are an important technical basis for realizing the automatic analysis and location function of power plant faults.

[0055] Optionally, the station route and track planning 33 realizes the fully autonomous flight of the UAV, ensures full coverage of all photovoltaic modules and minimizes repeated paths. Specifically, the following steps are included:

[0056] Step 1: Use the vector data of the extracted photovoltaic array to provide location information support for UAV trajectory planning.

[0057] Step 2: Design reasonable flight routes based on the layout of the PV site. Generally, each route covers 3-4 rows of PV strings to ensure full coverage of all PV modules and minimize repeated paths.

[0058] Step 3: According to the performance and safety requirements of the drone, set the appropriate flight altitude and speed. Generally, keep the aircraft at a fixed altitude of 30-35 meters from the photovoltaic panel and the flight speed of 6m / s to ensure the shooting quality and flight safety.

[0059] Step 4: Incorporate obstacle avoidance strategies into the flight path and enable the obstacle avoidance function to ensure that the drone can automatically avoid obstacles when it encounters them.

[0060] Optionally, the drone can perform fully autonomous inspection 34, fly at a constant speed according to the planned route, altitude, speed, gimbal angle and other parameters, and collect dual-light data of the on-site photovoltaic station. The specific steps include:

[0061] Step 1: Send the planned station route file to the drone responsible for execution.

[0062] Step 2: The drone will fly at a constant speed according to the planned route, altitude, speed, gimbal angle and other parameters to collect dual-light data of the on-site photovoltaic station.

[0063] Step 3: After the acquisition is completed, it automatically returns to the starting position and completes the data upload via the network or SD card.

[0064] Optionally, photovoltaic image defect recognition 35 mainly performs preprocessing, feature extraction and classification recognition on the images collected by the drone, and the algorithm automatically identifies various defects on the photovoltaic panel, such as damage, stains, hot spots, broken wires, etc. Specifically, it includes the following steps:

[0065] Step 1: Image preprocessing, denoising, enhancing and adjusting the acquired image to improve the accuracy of subsequent recognition. This may include image cropping, scaling, filtering and other operations.

[0066] Step 2: Feature extraction, using deep learning algorithms (such as convolutional neural networks (CNN)) to automatically extract feature information from photovoltaic images. These feature information may include the shape, size, direction, etc. of the crack.

[0067] Step 3: Defect identification, compare and identify the extracted feature information with the preset defect feature library. The comparison results are used to determine whether there are defects on the surface of the photovoltaic module and the type and location of the defects, including eight major categories: module hot spots, module multiple spots, module diodes, module missing, module fragmentation, module abnormality, and module surface abnormality.

[0068] Optionally, the defect information is fused with the vector data 36 to achieve a one-to-one correspondence between the fault location and the operation number, which can display the layout and defect distribution of the photovoltaic array in real time, helping the operation and maintenance personnel to quickly locate and handle the defects. Specifically, the following steps are included:

[0069] Step 1: Data preprocessing: digitize the extracted PV array vector information to ensure its consistency with the defect information in format and coordinate system. Clean and organize the PV defect information to remove noise and redundant data.

[0070] Step 2: Spatial fusion, superimposing the defect information (such as the hot spot location in the infrared thermal imaging image) with the vector layout map of the photovoltaic array to achieve accurate positioning of the defect in the photovoltaic array.

[0071] Step 3: Attribute fusion, associate the detailed information of the defect (such as defect type, severity, etc.) with the attribute information of the photovoltaic panel (such as model, production date, etc.) to form a complete fault record.

[0072] Step 4: Fault analysis and diagnosis: Use the fused information to conduct a comprehensive analysis of the faults in the photovoltaic array to determine the cause, scope and severity of the faults. According to the analysis results, formulate targeted maintenance plans to improve operation and maintenance efficiency.

[0073] As an optional embodiment, Figure 4 This is a flow chart of a photovoltaic defect location method based on defect location and vector data combination according to an embodiment of the present application, which specifically includes the following steps:

[0074] Step 1: Overall aerial survey of the site. Use a drone equipped with an oblique camera to conduct an overall aerial survey of the site at an altitude of about 120 meters from the photovoltaic panels.

[0075] Step 2: Generate an aerial electronic map. The aerial survey images collected by the photovoltaic station are stitched into a high-definition electronic map of the station.

[0076] Step 3: Extract the photovoltaic array vector map. Use software to extract the photovoltaic array vector map from the high-definition aerial electronic map.

[0077] Step 4: Autonomous inspection by drones. The route planning of the station is carried out based on the avionics map and vector map as the basis for track planning, and autonomous inspection data collection of the photovoltaic field is realized.

[0078] Step 5: Photovoltaic image defect recognition. The planned station route file is sent to the specific execution drone. The drone will fly at a constant speed according to the planned route, altitude, speed, gimbal angle and other parameters to collect dual-light data of the on-site photovoltaic station.

[0079] Step 6: Unify the fault information and operation number. Compare and identify the feature information extracted from the autonomous inspection images with the preset defect feature library. The comparison results are used to determine whether there are defects on the surface of the photovoltaic module and the type and location of the defects, including eight major categories: module hot spots, module multiple spots, module diodes, module missing, module fragmentation, module abnormality, and module surface abnormality.

[0080] Step 7: Fusion of defect information and appropriate data: Overlay the defect information (such as the Beidou location of hot spots in the infrared thermal imaging image) with the vector layout map of the photovoltaic array to achieve accurate positioning of the defect in the photovoltaic array.

[0081] Optionally, the defect information (such as the Beidou location of hotspots in the infrared thermal imaging image) can be overlaid with the vector layout map of the PV array. First, collect the infrared thermal imaging image and the vector layout map of the PV array. Use professional software to open these two images and make sure they are in the same coordinate system. Find the defect information in the infrared thermal imaging image, usually hot spots or hot spots, which indicate that the PV module may be faulty or damaged. Use software tools to overlay the defect information with the vector layout map of the PV array. Precise positioning can be achieved by dragging and adjusting the hotspot position. Confirm the accuracy of the defect location, which can be verified by comparing the physical location with the features of the surrounding PV modules. Mark the location of the defect in the overlay image, and use text or marking tools to add additional information. Such as Figure 5 As shown, Figure 5 This is an effect diagram of superimposing defect information and a vector layout diagram of a photovoltaic array according to an embodiment of the present application.

[0082] Step 8: Unify the fault information and operation number. The fault information corresponds to the operation number of the PV module one by one, showing the layout and defect distribution of the PV array, helping operation and maintenance personnel to quickly locate and handle defects. Figure 5 It can be clearly seen that the fusion of defect information and vector data realizes the one-to-one correspondence between the fault location and the operation number. Among them, 1-16-1, 1-16-2, and 1-16-3 represent the three fault locations in the same photovoltaic panel, and 1-16, 1-15, 1-14, 1-17, and 1-5 all represent the operation number, which represents the unique identification number of each component in the photovoltaic array.

[0083] It should be noted that when generating an aerial electronic map, it is necessary to ensure that the map resolution is high enough to clearly display the layout and detailed information of the photovoltaic array, so as to extract the vector map and locate the defects in the future. When extracting the vector map of the photovoltaic array, it is necessary to ensure that the extracted vector map accurately reflects the layout and structure of the actual photovoltaic array, so as to locate the defects in the future. When planning the route and track of the drone, it is necessary to consider the layout and structure of the photovoltaic array, reasonably design the route and track, ensure that the drone can fully cover the photovoltaic array area, and improve the inspection efficiency. When conducting fully autonomous inspections by drones, it is necessary to ensure that the drone equipment operates normally, the flight is stable, and it can inspect according to the pre-planned route and track, and timely discover photovoltaic defects. When performing photovoltaic image defect recognition, high-resolution images are required to ensure that defects in the photovoltaic array, such as cracks, pollution, etc., can be accurately identified. When fusing defect information with vector data, it is necessary to ensure the matching and accuracy between the two, so as to locate and repair defects in the future. At the same time, it is also necessary to consider how to effectively display and analyze the fused data to help optimize photovoltaic defect location and maintenance work.

[0084] In summary, the optional embodiment of the present application combines Beidou positioning data with image recognition results through vector data fusion technology, and conducts a comprehensive analysis of the two through a data fusion algorithm to further improve positioning accuracy and efficiency. Among them, the vector data contains key information such as the layout of the photovoltaic array, the size and position of the photovoltaic panel, etc., which, combined with Beidou positioning data, can achieve accurate measurement and positioning of the photovoltaic panel and defect position. At the same time, vector data fusion also helps to optimize the inspection route, reduce unnecessary flight or walking distance, and improve inspection efficiency. In addition, the above technology can also achieve real-time monitoring and rapid response of photovoltaic power stations. Operation and maintenance personnel can view the operating status and inspection screen of the photovoltaic array in real time through the remote monitoring system. Once an abnormal situation is found, the system can quickly generate an alarm message and locate the defect position. Furthermore, according to the positioning information and fault description provided by the system, maintenance resources can be quickly dispatched and the fault can be handled on site to ensure the stable operation of the photovoltaic power station. Furthermore, the above technology also integrates a data analysis and management platform, which can centrally store, process and analyze the collected image, positioning, defect and other data. Through big data analysis technology, the platform can mine the laws and trends behind the data and provide decision support for the operation and maintenance management of photovoltaic power plants. Based on the data analysis results, operation and maintenance personnel can formulate targeted operation and maintenance strategies and optimization plans to improve the power generation efficiency and operation and maintenance level of photovoltaic power plants.

[0085] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of each embodiment of the present application.

[0086] In this embodiment, a device for determining a defect position is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0087] Figure 6 : is a structural block diagram of a device for determining a defect position according to an embodiment of the present application, the device comprising:

[0088] An extraction module 62, used to extract a plurality of real-time feature information from the bi-optical image corresponding to the target plant station;

[0089] The identification module 64 is used to identify the difference between the sample feature information in the preset defect feature library and the plurality of real-time feature information to obtain defect information;

[0090] The determination module 66 is used to determine the target coordinates corresponding to the defect information in the photovoltaic array vector diagram, wherein the target coordinates are the defect positions of the defects in the located target station.

[0091] Based on the above device, multiple real-time feature information is extracted from the dual-light image corresponding to the target plant; the difference between the sample feature information in the preset defect feature library and the multiple real-time feature information is identified to obtain defect information; the target coordinates corresponding to the defect information in the photovoltaic array vector diagram are determined, wherein the target coordinates are the defect positions of the defects in the located target plant. The above technical solution solves the problem of being unable to quickly and accurately locate the defect positions of photovoltaic panels. Furthermore, by extracting multiple real-time feature information and comparing them with the preset defect feature library, the positions of defects in the photovoltaic array are determined, thereby improving the positioning accuracy of the defect positions of the photovoltaic array.

[0092] In an exemplary embodiment, the above-mentioned device also includes: an acquisition module, which is used to obtain electronic map data corresponding to the target station before extracting multiple real-time feature information from the dual-light image corresponding to the target plant; determine the photovoltaic array vector diagram of the target station based on the electronic map data; set the monitoring operation trajectory of the unmanned detection equipment according to the photovoltaic array vector diagram, and receive the real-time image of the unmanned detection equipment; wherein the real-time image is an image collected from the operation monitoring operation trajectory, and the real-time image at least includes: a real-time infrared area image of the target station, and a real-time visible light image of the target station; determine the dual-light image based on the real-time image.

[0093] In an exemplary embodiment, the above-mentioned device also includes: an adding module, which is used to determine that after the defect information is in the target coordinates corresponding to the photovoltaic array vector diagram, attribute data is added to the target coordinates, wherein the attribute data at least includes: the photovoltaic material attribute data corresponding to the associated detailed defect content and the target coordinates; the attribute data, the photovoltaic array vector diagram, the target coordinates and the electronic map data corresponding to the target station are integrated to generate a defect distribution diagram.

[0094] In an exemplary embodiment, the above-mentioned acquisition module is also used to obtain a logical diagram of the photovoltaic power station corresponding to the target station, wherein the distribution positions of all photovoltaic arrays in the target station are present in the logical diagram of the photovoltaic power station; array data is extracted from the logical diagram of the photovoltaic power station to obtain the row and column numbers of each photovoltaic array and the array vertex coordinates of each photovoltaic array; when the management object has identified the vector base points of the map interface corresponding to the electronic map data, the row and column numbers and the array vertex coordinates are marked on the map interface, and the map interface is extracted using a preset extraction application to obtain a vector diagram of the photovoltaic array of the target station.

[0095] In an exemplary embodiment, the above-mentioned extraction module is also used to perform image preprocessing on the dual-light image to obtain a first image; and use a preset deep learning algorithm to extract features from the first image to obtain multiple real-time feature information, wherein the real-time feature information includes at least one of the following: crack size of the photovoltaic array, crack shape of the photovoltaic array, and crack direction of the photovoltaic array.

[0096] In an exemplary embodiment, the above-mentioned identification module is also used to determine the defect types and defect locations corresponding to multiple real-time feature information respectively; wherein the defect type includes at least one of the following: photovoltaic array hot spots, photovoltaic array multiple spots, photovoltaic array missing, photovoltaic array fragmentation, photovoltaic array abnormality, photovoltaic array surface abnormality; use the sample feature information corresponding to the sample type and sample position to screen the defect type and defect location; determine the defect information based on the screening results.

[0097] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0098] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0099] S1, extracting multiple real-time feature information from the dual-light image corresponding to the target plant station;

[0100] S2, identifying the difference between the sample feature information in the preset defect feature library and the plurality of real-time feature information to obtain defect information;

[0101] S3. Determine the target coordinates corresponding to the defect information in the photovoltaic array vector diagram, wherein the target coordinates are defect locations where defects exist in the located target station.

[0102] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0103] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0104] An embodiment of the present application further provides a computer program product, including a computer program, and when the computer program is executed by a processor, the steps in any one of the above method embodiments are performed.

[0105] An embodiment of the present application further provides another computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0106] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0107] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0108] S1, extracting multiple real-time feature information from the dual-light image corresponding to the target plant station;

[0109] S2, identifying the difference between the sample feature information in the preset defect feature library and the plurality of real-time feature information to obtain defect information;

[0110] S3. Determine the target coordinates corresponding to the defect information in the photovoltaic array vector diagram, wherein the target coordinates are defect locations where defects exist in the located target station.

[0111] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0112] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0113] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0114] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for determining a defect location, characterized in that: include: Extract multiple real-time feature information from the bi-optical image corresponding to the target plant; Identify the difference between the sample feature information in the preset defect feature library and the plurality of real-time feature information to obtain defect information; Determine the target coordinates corresponding to the defect information in the photovoltaic array vector diagram, wherein the target coordinates are defect locations where defects exist in the located target station.

2. The method according to claim 1, characterized in that Before extracting a plurality of real-time feature information from the bi-optical image corresponding to the target plant, the method further includes: Obtain electronic map data corresponding to the target station; Determine a photovoltaic array vector diagram of the target station based on the electronic map data; According to the photovoltaic array vector diagram, a monitoring operation track of the unmanned detection equipment is set, and a real-time image of the unmanned detection equipment is received; wherein the real-time image is an image collected by running the monitoring operation track, and the real-time image at least includes: a real-time infrared area image of the target station and a real-time visible light image of the target station; A bi-optic image is determined based on the real-time image.

3. The method according to claim 1, characterized in that After determining the target coordinates corresponding to the photovoltaic array vector diagram of the defect information, the method further includes: Adding attribute data to the target coordinates, wherein the attribute data at least includes: the photovoltaic material attribute data corresponding to the associated detailed defect content and the target coordinates; The attribute data, the photovoltaic array vector diagram, the target coordinates and the electronic map data corresponding to the target station are integrated to generate a defect distribution diagram.

4. The method according to claim 2, characterized in that: Determining a photovoltaic array vector diagram of a target station based on the electronic map data includes: Acquire a photovoltaic power station logic diagram corresponding to the target station, wherein the photovoltaic power station logic diagram contains the distribution positions of all photovoltaic arrays in the target station; Extracting array data from the photovoltaic power station logic diagram to obtain row and column numbers of each photovoltaic array and array vertex coordinates of each photovoltaic array; When the management object has identified the vector base points of the map interface corresponding to the electronic map data, the row and column numbers and the array vertex coordinates are identified on the map interface, and the map interface is extracted using a preset extraction application to obtain the photovoltaic array vector diagram of the target site.

5. The method according to claim 1, characterized in that Extract multiple real-time feature information from the bi-optical image corresponding to the target plant, including: Performing image preprocessing on the bi-optical image to obtain a first image; A preset deep learning algorithm is used to perform feature extraction on the first image to obtain multiple real-time feature information, wherein the real-time feature information includes at least one of the following: crack size of the photovoltaic array, crack shape of the photovoltaic array, and crack direction of the photovoltaic array.

6. The method according to claim 1, characterized in that Identifying the difference between the sample feature information in the preset defect feature library and the plurality of real-time feature information to obtain defect information includes: Determine the defect types and defect locations corresponding to the multiple real-time feature information respectively; wherein the defect type includes at least one of the following: photovoltaic array hot spots, photovoltaic array multiple spots, photovoltaic array missing, photovoltaic array fragmentation, photovoltaic array abnormality, photovoltaic array surface abnormality; Using the sample feature information corresponding to the sample type and the sample position to screen the defect type and the defect position; Defect information is determined based on the screening results.

7. A device for determining a defect position, characterized in that: include: An extraction module, used to extract multiple real-time feature information from the bi-optical image corresponding to the target plant; An identification module, used to identify the difference between the sample feature information in the preset defect feature library and the plurality of real-time feature information, and obtain defect information; The determination module is used to determine the target coordinates corresponding to the defect information in the photovoltaic array vector diagram, wherein the target coordinates are the defect positions of the defects in the located target station.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the method according to any one of claims 1 to 6 when executed.

9. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the method according to any one of claims 1 to 6 through the computer program.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images