Photovoltaic module detection method and system, and computer program product

Abstract

The invention provides a photovoltaic module detection method and system, and a computer program product, and relates to the field of photovoltaic technology. The photovoltaic module detection method provided by the invention comprises the following steps: acquiring a detection image of a photovoltaic module; sending the detection image to a target network address for quality judgment; receiving a returned judgment result; and based on the determination result, determining whether to carry out treatment of the next production process on the photovoltaic module.

Description

Technical Field

[0001] The present invention relates to the field of photovoltaic technology, and in particular to a photovoltaic module detection method, system and computer program product. Background Art

[0002] Photovoltaic modules (also known as solar panels) are devices that convert solar energy into electrical energy and are widely used in residential, commercial and industrial fields. In the production line of photovoltaic modules, each production process has a processing time requirement to ensure that the photovoltaic modules can be processed within the predetermined time, improve production efficiency and reduce production costs. The quality of photovoltaic modules is a key factor affecting the performance and life of photovoltaic power generation systems. Therefore, quality determination is an especially important link in the production line of photovoltaic modules. How to accurately and efficiently determine the quality is particularly critical to improving the timeliness of the photovoltaic module production line.

[0003] In view of this, some embodiments of the present specification provide a photovoltaic module detection method, system and computer program product, aiming to improve the accuracy and timeliness of photovoltaic module quality determination. Summary of the invention

[0004] One or more embodiments of the present specification provide a method, comprising: acquiring a detection image of a photovoltaic component; sending the detection image to a target network address for quality determination; receiving a returned determination result; and determining whether to perform the next production process on the photovoltaic component based on the determination result.

[0005] According to the method provided in one or more embodiments of the present specification, obtaining a detection image of a photovoltaic component includes: obtaining a first image of the photovoltaic component; processing the first image to obtain the detection image, wherein the detection image meets preset image quality requirements and the data volume of the detection image is less than the data volume of the first image.

[0006] The method provided according to one or more embodiments of the present specification further includes: performing a quality analysis on the photovoltaic component based on the first image to obtain an analysis result; and sending the analysis result to the target network address to assist the quality determination.

[0007] According to the method provided by one or more embodiments of this specification, the first image is an electroluminescence image of the photovoltaic component to be inspected.

[0008] According to the method provided by one or more embodiments of the present specification, the processing of the first image to obtain the detection image includes: compressing or intercepting the first image to obtain the detection image.

[0009] According to the method provided by one or more embodiments of the present specification, intercepting the first image to obtain the detection image includes: determining a defect area in the first image; and intercepting the first image based on the defect area to obtain the detection image including the defect area.

[0010] According to the method provided by one or more embodiments of the present specification, when there are more than two defect areas, the intercepting the first image based on the defect areas to obtain the detection image including the defect areas includes: intercepting the first image based on the two or more defect areas to obtain a single detection image, and the single detection image includes the two or more defect areas; or intercepting the first image based on each defect area in the two or more defect areas to obtain more than two detection images, and each of the two or more detection images corresponds to each defect area in the two or more defect areas.

[0011] The method provided according to one or more embodiments of the present specification further includes: marking a defect area in the detection image, and marking a corresponding defect type based on the defect area.

[0012] According to the method provided by one or more embodiments of the present specification, compressing or intercepting the first image to obtain the detection image includes: obtaining network parameters corresponding to the target network address; wherein the network parameters include at least one of packet loss rate, delay, and bandwidth; when the network parameters meet preset conditions, compressing the first image to obtain the detection image; when the network parameters do not meet the preset conditions, intercepting the first image to obtain the detection image.

[0013] The method provided according to one or more embodiments of the present specification also includes: determining the target network address from the one or more candidate network addresses based on network parameters corresponding to the one or more candidate network addresses; wherein the network parameters include at least one of packet loss rate, delay, and bandwidth.

[0014] The method provided according to one or more embodiments of the present specification further includes: when no returned determination result is received within a preset time, changing the target network address based on the network parameters, and sending the detection image to the changed target network address for quality determination.

[0015] The method provided according to one or more embodiments of the present specification further includes: when no returned determination result is received within a preset time, sending a reminder message to the target network address.

[0016] According to the method provided in one or more embodiments of the present specification, determining whether to process the photovoltaic component for the next production process based on the judgment result includes: when the judgment result indicates that the photovoltaic component is unqualified, transferring the photovoltaic component to a rework station on the production line; when the judgment result indicates that the photovoltaic component is qualified, transferring the photovoltaic component to the station corresponding to the next production process on the production line.

[0017] One or more embodiments of the present specification also provide a photovoltaic component inspection system, including: an image acquisition module for acquiring an inspection image of a photovoltaic component; a quality judgment module for sending the inspection image to a target network address for quality judgment; a result receiving module for receiving a returned judgment result; and a judgment processing module for determining whether to perform the next production process on the photovoltaic component based on the judgment result.

[0018] One or more embodiments of the present specification further provide a computer program product, including a computer program, which, when at least a portion of the computer program is executed by a processor, can implement the photovoltaic component detection method described in some embodiments of the present specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] This specification will be further explained by way of exemplary embodiments, which will be described in detail by way of the accompanying drawings. The same numbers in the drawings represent the same structures or steps.

[0020] Figure 1 It is an exemplary flow chart of a photovoltaic component detection method shown in some embodiments of this specification.

[0021] Figure 2 is an exemplary flow chart of another photovoltaic component detection method shown in some embodiments of this specification.

[0022] Figure 3 It is an exemplary flow chart of an image capture method shown in some embodiments of this specification.

[0023] Figure 4 is an exemplary block diagram of a photovoltaic assembly detection system according to some embodiments of the present specification. DETAILED DESCRIPTION

[0024] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the embodiments will be described in detail below with reference to the accompanying drawings. Obviously, the contents described below are some examples or embodiments of this specification. For ordinary technicians in this field, without paying creative work, the technical solutions or means disclosed in this specification can also be applied to other scenarios based on these technical contents.

[0025] It should be understood that the "system", "device", "unit" and / or "module" used in this specification is a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0026] Unless otherwise specified, technical terms used in this specification to describe components, elements, etc. do not refer to the singular, but may also include the plural. Generally speaking, terms such as "include", "comprise", etc. only indicate that the steps, elements, or components that have been clearly identified are included, and these steps, elements, and components do not constitute an exclusive list, such as the method or device described may also include other steps or components.

[0027] Flowcharts are used in this specification to illustrate the operation steps performed by the device or system of the relevant embodiments, but unless otherwise specified, the order used to describe these steps should not be understood as a limitation on the order in which the steps are performed. A person of ordinary skill in the art can adjust the order in which these steps are performed based on the knowledge and information conveyed by the embodiments of this specification, and the adjustment includes but is not limited to swapping the order of the order, merging multiple steps, and splitting a certain step.

[0028] The production line of photovoltaic modules can include multiple production processes, and the production process is the various process flows or operations for processing raw materials (such as silicon wafers) into final photovoltaic modules. The production process of photovoltaic modules may include, for example, welding, stacking, quality judgment, lamination, framing, junction box installation, cleaning, IV testing, finished product inspection, packaging, etc. Each process needs to be coordinated with each other, and the process treatment of photovoltaic modules must be completed according to the production rhythm requirements. The production rhythm refers to the interval time between two products that have completed the same process treatment continuously on the production line. It can also be understood that it refers to the average time required to complete the process treatment of a product. Exemplarily, the production rhythm of quality judgment can be more than ten seconds.

[0029] Quality determination is a particularly important step in the production of photovoltaic modules, which is responsible for ensuring that each photovoltaic module meets the quality standards before entering the next process. Quality determination can be achieved through electroluminescence (EL) detection technology. Inside the photovoltaic module, when electrons return to the valence band, they release energy, which is manifested as light emission, i.e. electroluminescence. In the electroluminescence test, by applying a certain voltage across the two ends of the photovoltaic module, the electrons inside the module can be excited to emit visible light. The intensity and distribution of these lights can reflect the structure and performance of the internal components. On the production line, after the photovoltaic module comes out of the previous production process (such as lamination), it passes through the assembly line to the electroluminescent detection device (or EL detection device). The EL probe in the electroluminescent detection device is pressed down to the power-on module of the photovoltaic module, and then the voltage is applied to make the photovoltaic module exhibit electroluminescence. A high-resolution camera (such as an infrared camera) is used to capture the electroluminescent image (or EL image). The electroluminescent image is analyzed to determine the quality of the photovoltaic group, so as to find out whether the photovoltaic module has defects such as hidden cracks, fragments, poor solder joints, and broken grids.

[0030] The quality determination of photovoltaic modules can be achieved through machine determination or manual determination. Machine determination can be achieved by transmitting the captured EL image to an image detection model, analyzing the electroluminescent image through the image detection model and giving a determination result to automatically identify the defects of the photovoltaic module. For manual determination, the captured electroluminescent image can be displayed on the display interface of the electroluminescent detection device for on-site technicians to view and perform manual analysis to determine the defects of the photovoltaic module.

[0031] In some embodiments, since the analysis technology of the image detection model is not mature, the automatic judgment of the EL image through the image detection model cannot meet the accuracy requirements of the quality judgment. In addition, the on-site technicians of some production lines are not experienced enough, so the judgment level of on-site technicians of production lines in various regions is uneven, and the accuracy requirements of the quality judgment of photovoltaic modules cannot be guaranteed. Therefore, it is necessary to transmit the EL image to the remote address corresponding to the experienced technician for manual judgment. In addition, the process of remotely transmitting the EL image for manual judgment must also meet the production rhythm requirements of quality judgment to ensure the stable operation of the production line.

[0032] To this end, some embodiments of the present specification propose a photovoltaic module detection method, which obtains a detection image of the photovoltaic module, sends the detection image to a target network address for quality judgment, and receives a returned judgment result, and determines whether to proceed to the next production process for the photovoltaic module based on the judgment result, so as to improve the accuracy of quality judgment of the photovoltaic module, meet the production rhythm requirements for quality judgment, and ensure the stable operation of the production line.

[0033] Figure 1 It is an exemplary flow chart of a photovoltaic component detection method shown in some embodiments of this specification. Figure 1 The process 100 shown can be executed by a computer device, for example, it can be implemented by a photovoltaic module detection system 400 deployed on the computer device. In some embodiments, the computer device can be a detection device for determining the quality of photovoltaic modules, such as an electroluminescence (or EL) detection device. The computer device can also be a desktop computer, a laptop computer, a mobile phone, etc. Figure 1 As shown, process 100 may include the following steps.

[0034] Step 110 , acquiring a detection image of the photovoltaic assembly. In some embodiments, step 110 may be implemented by the image acquisition module 410 .

[0035] In some embodiments, the detection image can be an electroluminescent image of the photovoltaic component to be detected. In some embodiments, when the photovoltaic component to be detected comes out of the previous production process (such as lamination) and reaches the electroluminescent detection device through the assembly line, the EL probe in the electroluminescent detection device is pressed down to the power-on module of the photovoltaic component to be detected, and then the voltage is applied to make the photovoltaic component to be detected present an electroluminescent phenomenon, and a high-resolution camera (such as an infrared camera) is used to photograph the photovoltaic component to be detected to obtain an electroluminescent image, and the electroluminescent image is used as the detection image. In other embodiments, the obtained electroluminescent image can be further processed to obtain a detection image. For example, when the network condition is poor, the electroluminescent image is further compressed or the electroluminescent image is intercepted to obtain the detection image, so as to reduce the amount of data transmitted when the detection image is transmitted through the network. The relevant specific instructions for further processing the electroluminescent image to obtain the detection image will be described in detail later, and the specific details can be referred to the description in the process 200 later.

[0036] Step 120 , sending the detected image to a target network address for quality determination. In some embodiments, step 120 may be implemented by the quality determination module 420 .

[0037] In some embodiments, the target network address can be determined from one or more candidate network addresses based on the network parameters corresponding to one or more candidate network addresses. The network parameters can be network parameters on the network link between the network device sending the detection image and the network device corresponding to the target network address. The network parameters can include at least one of packet loss rate, delay, and bandwidth. Packet loss rate refers to the percentage of data packets that fail to successfully reach the receiving end network address in the process of data transmission, and the packet loss rate can reflect the stability and communication quality of the network. Delay (also known as time delay or delay) refers to the time required for data to travel from the sending end network address to the receiving end network address during data transmission, and the delay can reflect the response speed of the network. Bandwidth refers to the maximum amount of data that can be transmitted per unit time during data transmission, usually in bits per second. Bandwidth reflects the transmission capacity of the network or communication channel and determines how much data can be sent and received within a certain period of time.

[0038] In some embodiments, the network addresses corresponding to one or more regions equipped with experienced technicians can be determined as candidate network addresses, and one or more candidate network addresses can be screened according to network parameters. In some embodiments, one or more candidate network addresses with a packet loss rate less than a first preset threshold can be determined as the target network address; one or more candidate network addresses with a delay less than a second preset threshold can also be determined as the target network address; one or more candidate network addresses with a bandwidth greater than a third preset threshold can also be determined as the target network address. In other embodiments, the target network address can also be determined from one or more candidate network addresses according to two or more parameters of the packet loss rate, delay, and bandwidth. For example, one or more candidate network addresses with a packet loss rate less than a first preset threshold and a delay less than a second preset threshold are determined as the target network address; or one or more candidate network addresses with a packet loss rate less than a first preset threshold and a bandwidth greater than a third preset threshold are determined as the target network address; or one or more candidate network addresses with a delay less than a second preset threshold and a bandwidth greater than a third preset threshold are determined as the target network address; or one or more candidate network addresses with a packet loss rate less than a first preset threshold, a delay less than a second preset threshold, and a bandwidth greater than a third preset threshold are determined as the target network address.

[0039] In some embodiments, the candidate network addresses may also be ranked according to network parameters. For example, each candidate network address may be prioritized in the order of packet loss rate from small to large, or each candidate network address may be prioritized in the order of delay from small to large, or each candidate network address may be prioritized in the order of bandwidth from large to small. In some embodiments, each candidate network address may also be prioritized by comprehensive calculation based on network parameters. Exemplarily, a weight may be assigned to the ranking corresponding to each network parameter (such as packet loss rate, delay, and bandwidth), and a weighted average of these network parameters may be calculated according to the weight to obtain a comprehensive score corresponding to each candidate network address, and a priority may be assigned to each candidate network address according to the comprehensive score. In some embodiments, the network parameters may be dynamically updated, and the ranking results of each candidate network address may also be updated according to the updated network parameters.

[0040] In some embodiments, after the detection image is sent to the target network address, the computer device at one end of the target network address receives the detection image and displays it in the display interface, and an experienced technician performs manual quality judgment based on the detection image displayed in the display interface, and the judgment result is sent back to the network address corresponding to the computer device that sends the detection image, so as to improve the accuracy of quality judgment of the photovoltaic module. Exemplarily, it is possible to judge whether the photovoltaic module has defects such as hidden cracks, fragments, cold welding and broken grids based on the detection image, and the judgment result is sent back to the network address corresponding to the computer device that sends the detection image. In other embodiments, after the computer device at one end of the target network address receives the detection image, the detection image can also be input into the image detection model for machine judgment, which is not specifically limited here.

[0041] In some embodiments, when the detection image is sent to the target network address, the machine information and product information of the electroluminescent image detection device for detecting the photovoltaic module can also be sent to the target network address, so as to perform statistical analysis on the equipment status and product status of the electroluminescent image detection device. Among them, the machine information can include, for example, the equipment number, equipment model, equipment manufacturer, equipment status, calibration date, software version, location information (such as the workshop or production line location where the equipment is located), etc.; the product information can include, for example, the identification of the photovoltaic module, the batch number of the photovoltaic module, etc.

[0042] Step 130 , receiving the returned determination result. In some embodiments, step 130 may be implemented by the result receiving module 430 .

[0043] In some embodiments, the determination result may include a pass and a fail, where a pass indicates that the photovoltaic component corresponding to the detection image meets the quality requirements, and a fail indicates that the photovoltaic component corresponding to the detection image does not meet the quality requirements. Further, when the determination result is a fail, the determination result may also include the reason for the fail, such as defects in the photovoltaic component corresponding to the detection image.

[0044] In some embodiments, when the returned determination result is not received within a preset time, the target network address can be replaced based on the network parameters, and the detection image can be sent to the replaced target network address for quality determination. In some embodiments, the target network address can be replaced according to the sorting results of each candidate network address. For example, the candidate network address with the highest ranking in the sorting results other than the target network address is determined as the new target network address. For the description of the sorting method of each candidate network address, reference can be made to the description in the previous step 120, which will not be repeated here.

[0045] In some embodiments, when no returned determination result is received within a preset time, a reminder message may be sent to the target network address. Exemplarily, an alarm reminder message may be sent to the target network address, and the computer device receiving the reminder message may display the alarm reminder message through a display interface or send an alarm reminder through an audio output device.

[0046] In some embodiments, when no returned judgment result is received within a preset time, the photovoltaic component to be judged can also be placed in a buffer zone. After receiving the judgment result of the photovoltaic component, it is determined whether to perform the next production process on the photovoltaic component based on the judgment result.

[0047] Step 140 , based on the determination result, determines whether to process the photovoltaic module in the next production process. In some embodiments, step 140 may be implemented by the determination processing module 440 .

[0048] In some embodiments, when the determination result indicates that the photovoltaic module is unqualified, the photovoltaic module is transferred to a rework station on the production line, wherein the rework station refers to a work area for inspecting and repairing photovoltaic modules that have not passed the quality determination. Exemplarily, when the determination result is NG, which means that the photovoltaic module is unqualified, the photovoltaic module is transferred to a rework station on the production line, and a technician re-inspects and repairs the photovoltaic module according to the defects indicated in the determination result. The repaired photovoltaic module can be sent back to the production line and quality determined according to the steps of process 100.

[0049] In some embodiments, when the determination result indicates that the photovoltaic module is qualified, the photovoltaic module is transferred to the workstation corresponding to the next production process on the production line. In some embodiments, the next production process may be a lamination process, and the lamination process may be performed by a photovoltaic module laminator to press glass, EVA, battery string, EVA, and photovoltaic module backplane into a photovoltaic module by vacuum hot pressing, and then cool and shape to form a solid whole. Exemplarily, when the determination result is OK, it means that the photovoltaic module is qualified, and the photovoltaic module is transferred to the laminator on the production line, and the photovoltaic module is subjected to a lamination process by the laminator.

[0050] In some embodiments, the electroluminescence image captured by a high-resolution camera may be further processed to obtain a detection image. Figure 2 is an exemplary flow chart of another photovoltaic component detection method shown in some embodiments of this specification. Figure 2 The process 200 shown can be executed by a computer device, for example, it can be implemented by a photovoltaic component detection system 400 deployed on the computer device. Figure 2 As shown, process 200 may include the following steps.

[0051] Step 210 , acquiring a first image of the photovoltaic assembly. In some embodiments, step 210 may be implemented by the image acquisition module 410 .

[0052] In some embodiments, the first image is an electroluminescent image of the photovoltaic component to be inspected. When the photovoltaic component to be inspected comes out of the previous production process (e.g., lamination) and reaches the electroluminescent detection device through the assembly line, the EL probe in the electroluminescent detection device is pressed down to the power-on module of the photovoltaic component to be inspected, and then the voltage is applied to make the photovoltaic component to be inspected exhibit electroluminescence, and a high-resolution camera (e.g., an infrared camera) is used to shoot the photovoltaic component to be inspected, thereby obtaining the first image.

[0053] Step 220 , processing the first image to obtain a detection image, the detection image meets the preset image quality requirement and the data volume of the detection image is less than the data volume of the first image. In some embodiments, step 220 may be implemented by the image acquisition module 410 .

[0054] In some embodiments, image quality requirements may include image resolution requirements, image pixel value requirements, image color depth requirements, peak signal-to-noise ratio of images before and after compression, structural similarity of images before and after compression, etc. The data volume of the image may be the number of bytes or bits required to store the image.

[0055] In some embodiments, the detection image can be obtained by compressing or intercepting the first image, so that the data volume of the detection image is smaller than the data volume of the first image, so that when the detection image is subsequently sent to the target network address, the bandwidth occupancy is reduced and the transmission rate of the detection image is improved. In addition, the detection image meets the preset image quality requirements to ensure that important image details in the first image are not lost, so that the quality of the photovoltaic component to be detected can be determined based on the detection image. In other embodiments, the first image can also be downsampled or subjected to related encoding processing to obtain the detection image. The detection image must meet the preset image quality requirements and the data volume of the detection image is smaller than the data volume of the first image.

[0056] In some embodiments, the compression process of the first image may be a lossless compression process. Lossless compression is a method of compressing data, and its principle is to compress the original data using the statistical redundancy of the data. The original data will not lose information during the compression process, and the original data will be restored during decompression. Exemplarily, the compression process can be implemented by an image compression model, and the image compression model can analyze the first image according to the preset quality requirements, and automatically adjust the compression rate according to the content information of different regions in the first image to compress the first image, such as compressing a complex area or an area with rich content information according to a first compression rate, and compressing a simple area or an area with relatively simple content information according to a second compression rate, wherein the first compression rate is less than the second compression rate, so as to adaptively compress the first image, and further determine whether the compressed image meets the preset quality requirements, and output the compressed image if it meets the preset quality requirements, and adjust the compression strategy (such as adjusting the compression rate of some areas) to re-compress it if it does not meet the preset quality requirements, until the compressed image meets the preset quality requirements. The preset quality requirements may be, for example, that the peak signal-to-noise ratio before and after compression is greater than a first preset value, or that the structural similarity is greater than a second preset value. In other embodiments, the first image may also be compressed using a picture compression algorithm. For example, the first image may be uniformly compressed using a picture compression algorithm according to a preset compression rate.

[0057] In some embodiments, the first image may be cropped according to a preset cropping requirement. For example, the first image may be cropped according to a preset cropping ratio to remove an edge portion of the first image that does not contain valid information. In other embodiments, the first image may be cropped according to a defective area in a specific first image. The relevant specific description of this part will be described in detail later. For details, please refer to the description of process 300 later.

[0058] In some embodiments, network parameters corresponding to the target network address may also be obtained, wherein the network parameters include at least one of packet loss rate, delay, and bandwidth. For a specific introduction to each network parameter, please refer to the description in the previous step 120, which will not be repeated here. Further, the network status may also be determined based on the network parameters. In some embodiments, when the network parameters meet the preset conditions, the first image may be compressed to obtain a detection image, and when the network parameters do not meet the preset conditions, the first image may be intercepted to obtain a detection image. The preset conditions may be determined based on at least one of packet loss rate, delay, and bandwidth. For example, the preset conditions may be any one of a packet loss rate less than a first preset threshold, a delay less than a second preset threshold, and a bandwidth greater than a third preset threshold, or any combination thereof. The network condition is judged by network parameters. When the network condition is good (for example, the network parameters meet the preset conditions), the first image is compressed, thereby reducing the data volume of the detection image while retaining as much valid information in the first image as possible. When the network condition is poor (for example, the network parameters do not meet the preset conditions), the first image is intercepted, thereby greatly reducing the data volume of the detection image while retaining the valid information in the first image. Both situations can ensure the image quality of the detection image and improve the timeliness of the detection image transmission, so as to complete the quality judgment of the photovoltaic module within the judgment beat corresponding to the quality judgment, and ensure the stable operation of the production line.

[0059] Step 230, sending the detected image to the target network address for quality determination. In some embodiments, step 230 can be implemented by the quality determination module 420. Step 230 can be a sequel to step 220, and the specific description of this step is similar to the above, and the specific description can be referred to the description in step 120 in the above text, which will not be repeated here.

[0060] Step 240 , performing quality analysis on the photovoltaic module based on the first image to obtain analysis results. In some embodiments, step 240 may be implemented by the analysis and determination module 450 .

[0061] In some embodiments, the quality analysis of the photovoltaic module can be performed based on the first image by an image detection model. The image detection model can be obtained by training the training data related to the electroluminescent image. The training data related to the electroluminescent image can include the electroluminescent sample image and the corresponding defect information. The defect information can include, for example, the defect type, the position information corresponding to the defect area, etc. After training, the image detection model can automatically output the analysis result according to the input electroluminescent image. The analysis result can include whether there is a defect in the electroluminescent image, the defect type, and the position information of the defect area corresponding to the defect area type. Among them, the position information of the defect area can be determined by the coordinate position in the preset coordinate system. Exemplarily, a preset coordinate system can be established with a fixed point on the photovoltaic module (such as the upper left corner as the center) as the origin. The position information of the defect area can be multiple coordinate positions. The area encircled by connecting the multiple coordinate positions in sequence can be used to mark the defect area. The defect type can include, for example, debris, broken grid, black spot, hidden crack and poor solder joint. For example, a first image with debris and broken grid defects is input into an image detection model, and the image detection model outputs corresponding analysis results, which may include that the photovoltaic module does not meet the quality standards, the defect types are debris and broken grid, multiple coordinate positions corresponding to the debris, and multiple coordinate positions corresponding to the broken grid.

[0062] In some embodiments, step 240 may be a sequel to step 210 , and step 220 and step 240 may be parallel steps that are executed independently of each other, and the execution order is not specific.

[0063] Step 250 , sending the analysis result to the target network address to assist in quality determination. In some embodiments, step 250 may be implemented by the analysis and determination module 450 .

[0064] In some embodiments, step 250 may be a sequel to step 240, and step 250 and step 230 may be parallel steps that are executed independently of each other, and the execution order is not particular. In other embodiments, step 250 may be executed simultaneously with step 230, that is, the detection image and the analysis result obtained by analyzing the first image may be sent to the target network address together, so that the technician corresponding to one end of the target network address can quickly understand the defect of the photovoltaic module according to the analysis result, and analyze the detection image in combination with the analysis result to quickly give a determination result.

[0065] In some embodiments, the defective area can also be marked in the detection image, and the corresponding defect type can be marked based on the defective area. In some embodiments, the detection image can be marked according to the analysis result of the first image, such as determining the vertex coordinates of the marking box according to the multiple coordinate positions corresponding to the defective area in the analysis result, determining the position and size of the marking box according to the vertex coordinates of the marking box to frame the defective area, and marking the defective area according to the defect type corresponding to the defective area in the analysis result. For example, if the analysis result includes "fragments" and four coordinate positions corresponding to the fragments, the four vertex coordinates of the marking box can be determined according to the four coordinate positions corresponding to the fragments, and the rectangular marking box can be determined according to the four vertex coordinates and the defective area can be framed, and the defect type can be marked as "fragments" at the position corresponding to the rectangular marking box (such as the position outside the rectangular marking box and close to the rectangular marking box). In some embodiments, when there are multiple defective areas, each defective area can be marked according to the above method and the corresponding defect type can be marked for each defective area. By marking the defective area in the detection image, the technicians who make the quality judgment can quickly and intuitively review the defect situation in the detection image according to the marking information in the detection image, and quickly make the quality judgment according to the defect situation.

[0066] In some embodiments, the identification of the detection image can correspond to the identification of the analysis result, so that the defect area can be marked in the corresponding detection image based on the analysis result according to the corresponding relationship. Exemplarily, the identification of the detection image can be a first barcode, and the identification of the analysis result can be a second barcode, and the first barcode corresponds to the second barcode.

[0067] In some embodiments, the steps of marking defective areas in the detection image and marking corresponding defect types based on the defective areas can be performed at the receiving end. The receiving end can be, for example, a computer device that receives the detection image. After receiving the analysis result of the first image and the detection image, the receiving end can mark the defective areas in the detection image according to the corresponding relationship between the identification of the detection image and the identification of the analysis result, and mark the corresponding defect types based on the defective areas, and display the marked detection image in the display interface of the computer device corresponding to the receiving end, so that the technician can make a quality judgment based on the detection image and the marking information in the detection image, and send the judgment result back to the computer device that sent the detection image.

[0068] In other embodiments, the step of marking defective areas in the detection image and marking corresponding defect types based on the defective areas can also be performed at the sending end. The sending end can be, for example, a computer device that sends the detection image. After the detection image is obtained in step 220 and the analysis result is obtained in step 240, before executing step 230, the defective areas can be marked in the detection image, and the corresponding defect types can be marked based on the defective areas. In step 230, the detection image marked with defective areas and defect types is sent to the target network address for quality determination.

[0069] In some embodiments, defective areas may be annotated in the first image, and corresponding defect types may be annotated based on the defective areas. The annotation method is similar to the annotation of defective areas and defect types in the detection image. For details, please refer to the description in step 250, which will not be repeated here. Furthermore, the first image annotated with defective areas and defect types may be sent to a target network address for quality determination.

[0070] Step 260, receiving the returned determination result. In some embodiments, step 260 can be implemented by the result receiving module 430. The specific description of step 260 is similar to the above, and the specific description can be referred to the above description of step 130, which will not be repeated here.

[0071] Step 270, based on the determination result, determines whether to process the photovoltaic module in the next production process. In some embodiments, step 270 can be implemented by the determination processing module 440. The specific description of step 270 is similar to the above, and the specific description can be referred to the description of step 140 above, which will not be repeated here.

[0072] Figure 3 It is an exemplary flow chart of an image capture method shown in some embodiments of this specification. Figure 3 The process 300 shown may be executed by a computer device, for example, may be implemented by an image acquisition module 410 in a photovoltaic module detection system 400 arranged on the computer device. In some embodiments, the process 300 may be a further description of step 220. Figure 3 As shown, process 300 may include the following steps.

[0073] Step 310: determine a defect area in the first image.

[0074] In some embodiments, the defective area can be determined in the first image based on the analysis result of the first image. The photovoltaic module is subjected to quality analysis based on the first image. The specific description of obtaining the analysis result can be found in the description of step 240 above, which will not be repeated here.

[0075] In some embodiments, the analysis result of the first image may include whether there are defects in the first image, the defect type, and the location information of the defect area corresponding to the defect area type. When there are defects in the first image, the defect area can be determined in the first image based on the location information of the defect area corresponding to the defect area type. Exemplarily, the location information corresponding to the defect area may include multiple coordinate positions, and the area circled by connecting the multiple coordinate positions in sequence may be determined as the defect area in the first image. In other embodiments, when there are no defects in the first image, the first image may be compressed to obtain a detection image, or the first image may be directly used as the detection image. The specific instructions for compressing the first image can refer to the description in the previous step 220, which will not be repeated here.

[0076] Step 320: intercept the first image based on the defect area to obtain a detection image including the defect area.

[0077] In some embodiments, the first image may be intercepted according to the boundary of the defective area to obtain a detection image including the defective area. Exemplarily, in order to ensure that the intercepted detection image contains the complete defective area, a certain pixel range may be extended outward according to the boundary of the defective area to obtain an extended area, and the extended pixel range may be determined according to the size and shape of the defective area, and the detection image including the defective area may be intercepted according to the boundary of the extended area to obtain.

[0078] In some embodiments, when there are more than two defective areas, the interception method of the first image can also be determined according to the data volume of the first image. When the data volume of the first image is less than the preset data volume, the first image can be intercepted to obtain a single detection image, so that the single detection image includes more than two defective areas; when the data volume of the first image is greater than or equal to the preset data volume, the first image can also be intercepted to obtain more than two detection images, and each of the more than two detection images corresponds to each of the more than two defective areas. The two interception methods are further described below.

[0079] In some embodiments, when there are more than two defective areas, the first image can be intercepted based on the two or more defective areas to obtain a single detection image, and the single detection image includes more than two defective areas. In some embodiments, for each defective area, a certain pixel range can be expanded outward according to the boundary of the defective area to obtain an extended area, and the first image can be intercepted based on the two or more extended areas to obtain a single detection image, and the single detection image includes more than two extended areas. In some embodiments, more than two defective areas or more than two extended areas can also be intercepted according to a preset shape, and the area size corresponding to the preset shape can be determined according to the boundary of each defective area or each extended area, so that the area corresponding to the preset shape includes each defective area or each extended area. Exemplarily, the preset shape can be a rectangle, and the position and size of the rectangular area can be determined according to the boundary position of each defective area or each extended area, so that the rectangular area can include each defective area or each extended area, and the rectangular area is intercepted according to the boundary of the rectangular area to obtain a detection image.

[0080] In other embodiments, when there are more than two defective areas, the first image can be intercepted based on each defective area in the more than two defective areas to obtain more than two detection images, and each detection image in the more than two detection images corresponds to each defective area in the more than two defective areas. In some embodiments, the number of intercepted detection images corresponds to the number of defective areas. Exemplarily, the number of intercepted detection images is the same as the number of defective areas, each detection image contains a defective area, and the defective areas contained in each detection image are different. In some embodiments, for each defective area, an extended area can be obtained by expanding a certain pixel range outward according to the boundary of the defective area, and the first image is intercepted according to the boundary of each extended area to obtain more than two detection images, each detection image contains an extended area, and the extended areas contained in each detection image are different.

[0081] In some embodiments, defect areas can also be marked in the detection image, and the corresponding defect types can be marked based on the defect areas. The marking method is similar to the above, and specific reference can be made to the description in step 250. The difference is that when the first image is intercepted based on each defect area in more than two defect areas to obtain more than two detection images, and each detection image in more than two detection images corresponds to each defect area in more than two defect areas, it is not necessary to determine the marking box, and the corresponding defect type can be directly marked in each detection image according to the defect type corresponding to the defect area in each detection image. Exemplarily, assuming that there are three defect areas in the first image, the first image is intercepted to obtain three detection images, and each detection image includes a defect area, then it is not necessary to determine the marking box, and the defect types corresponding to the defect areas can be directly marked in the three detection images according to the analysis results.

[0082] This specification also provides a photovoltaic module detection system. Figure 4 is an exemplary block diagram of a photovoltaic module detection system according to some embodiments of this specification. Figure 4 As shown, in some embodiments, the photovoltaic component detection system 400 may include an image acquisition module 410 , a quality determination module 420 , a result receiving module 430 , and a determination processing module 440 .

[0083] The image acquisition module 410 is used to acquire a detection image of the photovoltaic module.

[0084] The quality determination module 420 is used to send the detected image to a target network address for quality determination.

[0085] The result receiving module 430 is used to receive the returned determination result.

[0086] The determination processing module 440 is used to determine whether to perform the next production process on the photovoltaic module based on the determination result.

[0087] In some optional embodiments, the image acquisition module 410 is further used to acquire a first image of the photovoltaic assembly; process the first image to obtain a detection image, the detection image meets preset image quality requirements and the data volume of the detection image is less than the data volume of the first image.

[0088] In some optional embodiments, the photovoltaic component detection system 400 may further include an analysis and determination module 450 for performing quality analysis on the photovoltaic component based on the first image to obtain analysis results; and sending the analysis results to a target network address to assist in quality determination.

[0089] In some optional embodiments, the image acquisition module 410 is further used to compress or intercept the first image to obtain a detection image.

[0090] In some optional embodiments, the image acquisition module 410 is further used to determine a defect area in the first image; and to intercept the first image based on the defect area to obtain a detection image including the defect area.

[0091] In some optional embodiments, the image acquisition module 410 is also used to, when there are more than two defect areas, intercept the first image based on the two or more defect areas to obtain a single detection image, and the single detection image includes more than two defect areas; or intercept the first image based on each defect area in the more than two defect areas to obtain more than two detection images, and each of the more than two detection images corresponds to each defect area in the more than two defect areas.

[0092] In some optional embodiments, the photovoltaic module detection system 400 may further include a marking module for marking defective areas in the detection image and marking corresponding defect types based on the defective areas.

[0093] In some optional embodiments, the image acquisition module 410 is also used to obtain network parameters corresponding to the target network address; wherein the network parameters include at least one of packet loss rate, delay, and bandwidth; when the network parameters meet preset conditions, the first image is compressed to obtain a detection image; when the network parameters do not meet the preset conditions, the first image is intercepted to obtain a detection image.

[0094] In some optional embodiments, the photovoltaic component detection system 400 may further include an address determination module 460, which is used to determine the target network address from one or more candidate network addresses based on network parameters corresponding to one or more candidate network addresses; wherein the network parameters include at least one of packet loss rate, delay, and bandwidth.

[0095] In some optional embodiments, the address determination module 460 is also used to change the target network address based on network parameters when no returned determination result is received within a preset time, and send the detection image to the changed target network address for quality determination.

[0096] In some optional embodiments, the photovoltaic component detection system 400 may further include a reminder module 470 for sending a reminder message to a target network address when no returned determination result is received within a preset time.

[0097] In some optional embodiments, the determination processing module 440 is also used to, when the determination result indicates that the photovoltaic component is unqualified, transfer the photovoltaic component to a rework station on the production line; when the determination result indicates that the photovoltaic component is qualified, transfer the photovoltaic component to a station corresponding to the next production process on the production line.

[0098] For more information about each module, see Figures 1 to 3 The relevant description of will not be repeated here. It should be understood that Figure 4 The system and its modules shown can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented by hardware, software, or a combination of software and hardware. Among them, the hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. Those skilled in the art will understand that the above methods and systems can be implemented using computer executable instructions and / or control codes contained in a processor, such as a disk, CD or DVD-ROM, etc., and such codes are provided in a carrier medium or a memory of a programmable device. The system and its modules of this specification can not only be implemented by hardware circuits such as ultra-large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but can also be implemented by software executed by various types of processors, and can also be implemented by a combination of the above hardware circuits and software (e.g., firmware).

[0099] It should be noted that the above description of the system and its modules is only for convenience of description and does not limit this specification to the scope of the embodiments. It is understandable that for those skilled in the art, after understanding the principle of the system, it is possible to arbitrarily combine the modules without deviating from this principle to form a subsystem connected to other modules. Or some modules may be split to obtain more modules or multiple units under the module. Such variations are all within the scope of the disclosure of this specification.

[0100] Some embodiments of the present specification also provide a computer program product, including computer instructions. When at least part of the computer instructions are executed by a processor, the present specification can be implemented. Figures 1 to 3 In some embodiments, the computer program product may only involve computer instructions, which may be carried by a storage medium or a processing device. In other embodiments, the computer program product may also be a storage medium or a processing device containing the aforementioned computer instructions. The processing device may include one or more processors and a storage medium.

[0101] In some embodiments, the processor may be a combination of one or more of the following processors: a central processing unit (CPU), an application specific integrated circuit (ASIC), an application specific instruction set processor (ASIP), a graphics processing unit (GPU), a physical processing unit (PPU), a digital signal processor (DSP), a field programmable gate array (FPGA), a programmable logic device (PLD), a programmable logic controller (PLC), a reduced instruction set computer (RISC), and a microprocessor.

[0102] In some embodiments, the storage medium may include a combination of one or more of the following: a mass storage, a removable storage, a volatile read-write memory, and a read-only memory (ROM). Exemplary mass storage may include a magnetic disk, an optical disk, a solid-state hard disk, and the like. Exemplary removable storage may include a flash disk, a floppy disk, an optical disk, a memory card, a compressed hard disk, a magnetic tape, and the like. Exemplary volatile read-write memory may include a random access memory (RAM). Exemplary random access memory may include a dynamic random access memory (DRAM), a double data rate synchronous dynamic random access memory (DDRSDRAM), a static random access memory (SRAM), a thyristor random access memory (T-RAM), and a zero capacitance memory (Z-RAM), and the like. Exemplary read-only memory may include a masked read-only memory (MROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a compact disk read-only memory (CD-ROM), and a digital multifunction hard disk read-only memory, and the like.

[0103] The beneficial effects that may be brought about by the embodiments of this specification include but are not limited to: (1) by acquiring the detection image of the photovoltaic component, sending the detection image to the target network address for quality judgment, and receiving the returned judgment result, determining whether to process the photovoltaic component for the next production process based on the judgment result, so as to improve the accuracy of quality judgment of the photovoltaic component, meet the production rhythm requirements of quality judgment, and ensure the stable operation of the production line; (2) by compressing or intercepting the first image to obtain the detection image, so that the data volume of the detection image is smaller than the data volume of the first image, so that when the detection image is subsequently sent to the target network address, the bandwidth occupancy is reduced and the transmission rate of the detection image is improved. In addition, the detection image meets the preset image quality requirements to ensure that important image details in the first image are not lost, so that the quality of the photovoltaic component to be detected is judged based on the detection image, saving quality judgment time and improving quality judgment efficiency; (3) by The first image is subjected to quality analysis to obtain analysis results, and is sent to the target network address to assist in quality determination, so that the technicians at the receiving end can understand the defects of the photovoltaic module according to the analysis results and quickly make quality determinations, thereby saving quality determination time and improving quality determination efficiency; (4) By marking defective areas in the detection image, the technicians who make quality determinations can quickly and intuitively review the defects in the detection image according to the marked information in the detection image, and quickly make quality determinations according to the defects, thereby saving quality determination time and improving quality determination efficiency; (5) By setting preset conditions, when the network parameters meet the preset conditions, the first image is compressed to obtain the detection image, and when the network parameters do not meet the preset conditions, the first image is intercepted to obtain the detection image, thereby flexibly selecting the detection image acquisition method according to the network conditions, so as to complete the quality determination within the specified production cycle, while retaining the effective information of the detection image. It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects that may be produced may be any one or a combination of the above, or any other beneficial effects that may be obtained.

[0104] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of this specification. Although not explicitly stated here, those skilled in the art may make various modifications, improvements and corrections to this specification. Such modifications, improvements and corrections are taught in this specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.

Claims

1. A photovoltaic module detection method, characterized in that: include: Acquire inspection images of photovoltaic modules; Sending the detected image to a target network address for quality determination; Receive the returned judgment result; Based on the determination result, it is determined whether to perform a next production process on the photovoltaic module.

2. The method according to claim 1, characterized in that The step of acquiring a detection image of a photovoltaic module comprises: Acquiring a first image of the photovoltaic assembly; The first image is processed to obtain the detection image, where the detection image meets a preset image quality requirement and a data volume of the detection image is less than a data volume of the first image.

3. The method according to claim 2, characterized in that Also includes: Performing quality analysis on the photovoltaic module based on the first image to obtain an analysis result; The analysis result is sent to the target network address to assist the quality determination.

4. The method according to claim 2, characterized in that: The first image is an electroluminescence image of the photovoltaic component to be inspected.

5. The method according to claim 2, characterized in that: The processing of the first image to obtain the detection image comprises: The first image is compressed or intercepted to obtain the detection image.

6. The method according to claim 5, characterized in that Intercepting the first image to obtain the detection image includes: determining a defect area in the first image; The first image is intercepted based on the defect area to obtain the detection image including the defect area.

7. The method according to claim 6, characterized in that When there are more than two defective areas, intercepting the first image based on the defective areas to obtain the detection image including the defective areas includes: Based on the two or more defect areas, the first image is intercepted to obtain a single detection image, and the single detection image includes the two or more defect areas; or The first image is intercepted based on each defect area in the two or more defect areas to obtain two or more detection images, and each detection image in the two or more detection images corresponds to each defect area in the two or more defect areas.

8. The method according to claim 5, characterized in that Also includes: A defect area is marked in the detection image, and a corresponding defect type is marked based on the defect area.

9. The method according to claim 5, characterized in that The compressing or intercepting the first image to obtain the detection image comprises: Acquire a network parameter corresponding to the target network address; wherein the network parameter includes at least one of a packet loss rate, a delay, and a bandwidth; When the network parameters meet the preset conditions, compressing the first image to obtain the detection image; When the network parameters do not meet the preset conditions, the first image is intercepted to obtain the detection image.

10. The method according to claim 1, characterized in that Also includes: The target network address is determined from the one or more candidate network addresses based on network parameters corresponding to the one or more candidate network addresses; wherein the network parameters include at least one of packet loss rate, delay, and bandwidth.

11. The method according to claim 10, characterized in that Also includes: When no returned determination result is received within a preset time, the target network address is changed based on the network parameters, and the detection image is sent to the changed target network address for quality determination.

12. The method according to claim 10, characterized in that Also includes: When no returned determination result is received within a preset time, a reminder message is sent to the target network address.

13. The method according to claim 1, characterized in that The step of determining whether to perform a next production process on the photovoltaic module based on the determination result includes: When the determination result indicates that the photovoltaic component is unqualified, the photovoltaic component is transferred to a rework station on the production line; When the determination result indicates that the photovoltaic component is qualified, the photovoltaic component is transferred to a workstation corresponding to the next production process on the production line.

14. A photovoltaic module detection system, characterized in that: include: An image acquisition module, used to acquire a detection image of a photovoltaic module; A quality determination module, used for sending the detected image to a target network address for quality determination; A result receiving module is used to receive the returned judgment result; A determination processing module is used to determine whether to perform the next production process on the photovoltaic component based on the determination result.

15. A computer program product, characterized in that The invention comprises a computer program, and when at least a part of the computer program is executed by a processor, the method according to any one of claims 1 to 13 can be implemented.

Images