Visual light source self-adaptive laser processing positioning system and method for photovoltaic cell

Abstract

The invention relates to a photovoltaic cell visual light source self-adaptive laser processing positioning system and method, and the system comprises a visual collection processing module which is used for collecting a photovoltaic cell image and recognizing and grabbing MARK points at four corners of a photovoltaic cell, and comprises a failure judgment unit which is used for judging whether the MARK points in the four corners of the photovoltaic cell fail or not; the automatic switching module is used for triggering an automatic switching instruction when MARK point feature capturing fails or an error exceeds a limit; the three-color light adaptive light source module is in communication connection with the visual acquisition and processing module; the coordinates of the MARK points at the four corners of the photovoltaic cell are captured through the visual acquisition and processing module, multiple sets of MARK point capturing schemes and corresponding three-color light source parameter information are stored in the preset visual scheme library, and when the identification of the cell fails, the system automatically calls the visual schemes in the preset scheme library, so that the identification efficiency of the cell is improved. According to the shape characteristics of the MARK points and the appearance characteristics of the battery piece, a proper visual scheme is matched, the work of manual troubleshooting is replaced, the positioning failure rate is reduced, and the working efficiency is improved.

Description

Technical Field

[0001] This application relates to the field of photovoltaic cell manufacturing technology, and in particular to a photovoltaic cell visual light source adaptive laser processing positioning system and method. Background Technology

[0002] In the field of photovoltaic cell manufacturing, laser processing technology is the core process that determines the photoelectric conversion efficiency and reliability of cells. Its processing accuracy needs to be controlled within ±5μm. The premise of achieving this accuracy is based on the precise coordinate positioning reference of the cell. The industry generally uses vision cameras to identify and capture the MARK points at the four corners of the cell. By converting the pixel coordinates of the MARK points to the world coordinates of the cell, a positioning reference is provided for the laser processing system.

[0003] However, during the production of photovoltaic cells, variations in process parameters lead to differences in the thickness and composition of the surface passivation film (such as SiNx, AlOx), resulting in different reflectance spectral characteristics. Simultaneously, the substrate color (such as N-type silicon, P-type silicon) and back electrode structure (such as the poly-Si stack in BC cells) also vary significantly. This diversity in appearance makes it extremely difficult for visual light sources and recognition algorithms with fixed parameters (such as a single wavelength and fixed brightness) to detect different batches and models of cells due to insufficient feature contrast. Currently, production lines generally rely on manual experience for switching light sources and adjusting algorithm parameters, which is not only inefficient but also heavily dependent on operator skill, becoming a bottleneck for achieving fully automated production. Summary of the Invention

[0004] To address the problems of high failure rates, low efficiency, and reliance on manual intervention in existing photovoltaic cell positioning systems, which are unable to adapt to the varied appearances of photovoltaic cells, this application provides a photovoltaic cell vision light source adaptive laser processing positioning system.

[0005] The photovoltaic cell visual light source adaptive laser processing positioning system provided in this application adopts the following technical solution: A photovoltaic cell vision light source adaptive laser processing positioning system includes: The visual acquisition and processing module is used to acquire images of photovoltaic cells, identify and capture MARK points on the four corners of the photovoltaic cells. It includes a failure judgment unit, which is used to trigger an automatic switching command when the MARK point feature capture fails or the error exceeds the limit. The three-color adaptive light source module is communicatively connected to the vision acquisition and processing module. It is used to store the light source parameters corresponding to the preset vision scheme for each set of MARK point capture, and to adjust the light source color and brightness after receiving the switching command. A preset vision scheme library is connected to the vision acquisition and processing module to store multiple vision grasping schemes for different battery cell appearances and different shaped MARK points, as well as the corresponding three-color light source parameters. The intelligent matching and decision-making module is connected to the visual acquisition and processing module and the preset visual scheme library, respectively. It is used to extract key features from the acquired failed images and perform intelligent matching based on the extracted features and the "feature-scheme" mapping relationship stored in the preset visual scheme library to determine the optimal visual scheme.

[0006] The above technical solution is used to capture the coordinates of the MARK points at the four corners of the photovoltaic cell through a visual acquisition and processing module. Multiple MARK point capture schemes and corresponding three-color light source parameter information are stored in a preset visual scheme library. When the cell recognition fails, the system automatically calls the visual scheme in the preset scheme library and matches the appropriate visual scheme according to the shape characteristics of the MARK point and the shape characteristics of the cell. This replaces the manual inspection work, reduces the positioning failure rate, and improves work efficiency.

[0007] Optionally, the system further includes a coordinate calculation, transformation, and transmission module, which is communicatively connected to the vision acquisition and processing module. This module is used to calculate the world coordinates of the battery cell based on the pixel coordinates of the identified MARK points, convert the world coordinates into processing coordinates for laser processing, and transmit them to the laser processing control module.

[0008] The above technical solution achieves seamless connection and high-precision conversion between visual positioning information and laser processing commands. The captured MARK point positions based on image pixels are converted into absolute coordinates in the real world through a calibration algorithm, thereby generating processing coordinates that the laser head can directly recognize, ensuring the accuracy of positioning.

[0009] Optionally, the system further includes a laser processing control module, which is communicatively connected to the coordinate calculation, conversion and transmission module, for receiving laser processing coordinate data transmitted from the coordinate calculation, conversion and transmission module, and controlling the laser to perform high-precision scanning processing on the battery cell.

[0010] The above technical solution enables the calculated coordinate data to be executed directly for high-quality, high-precision scanning and processing, thereby improving processing efficiency.

[0011] Optionally, the failure to capture MARK points refers to the fact that the number of MARK points identified in three consecutive frames of images is less than three.

[0012] The above technical solution is used to quantify the failure of the capture process, and to determine the capture failure when the number of MARK points identified in three consecutive frames of images is less than 3.

[0013] Optionally, the error exceeding the limit refers to the MARK point positioning error being greater than 10μm.

[0014] The above technical solutions are used to prevent unqualified positioning results from entering the processing stage, thus ensuring the reliability of the final processing accuracy.

[0015] Optionally, the light source of the three-color adaptive light source module is a red, blue, and green three-color adjustable light source, with a color adjustment wavelength range of 450-660nm and a brightness adjustment level of 0-255 levels.

[0016] The above technical solution is used to limit the specific parameters of the three-color light source in order to improve the recognition success rate and clarity of the MARK point under complex appearance conditions.

[0017] Optionally, the battery cell appearance information stored in the preset visual scheme library includes the color and film thickness of the battery cell.

[0018] The above technical solution uses the color and film thickness of the solar cells as key information in the solar cell solution library to improve the accuracy and efficiency of intelligent matching.

[0019] Optionally, the intelligent matching and decision-making module includes a feature extraction unit and a scheme matching unit; The feature extraction unit is used to extract key features of the battery cell, including the overall average gray level of the battery cell, the contrast between the MARK point region and the background, and image texture features. The scheme matching unit performs intelligent matching based on the extracted key features and the "feature-scheme" mapping relationship stored in the scheme library.

[0020] By employing the aforementioned technical solutions, and through quantitative analysis and precise mapping of image features, the system can quickly identify the optimal solution, significantly improving decision-making speed and efficiency, and achieving automated matching.

[0021] Optionally, the intelligent matching and decision-making module further includes a learning and updating unit, which records each successfully matched scheme and the corresponding battery cell features, and adds the corresponding scheme to the preset visual scheme library after the battery cell is successfully located, so as to realize the system's self-learning evolution.

[0022] Through the above technical solution, the positioning system is equipped with the function of self-learning during operation, enabling it to continuously accumulate successful solutions and expand the solution library autonomously. As a result, it has stronger adaptability and a higher positioning and processing success rate when dealing with new types of battery cells in the future.

[0023] Optionally, the three-color adaptive light source module includes a light source parameter mapping unit and a light source real-time adjustment unit; The light source parameter mapping unit is used to store the optimal light source color and brightness parameters corresponding to each visual scheme; The real-time light source adjustment unit is used to synchronously output a light source control signal when receiving a switching command.

[0024] By adopting the above technical solution, it can be ensured that the light source parameters are executed accurately and synchronously each time the scheme is switched, so that the system can operate efficiently and smoothly.

[0025] This invention also discloses an adaptive laser processing positioning method for photovoltaic cells using a visual light source, which is applied to the aforementioned positioning system and includes the following steps: S1: Photovoltaic cell loading, the vision camera of the vision acquisition and processing module uses the initial scheme and corresponding light source parameters to capture the MARK points at the four corners of the cell; S2: Determine whether the MARK point can be successfully identified and captured. If yes, proceed to S3. Otherwise, trigger the automatic switching command, switch the vision scheme, call the preset vision scheme library, match the optimal vision scheme according to the shape characteristics of the MARK point and the color and film thickness of the battery cell, and synchronously adjust the three-color light source parameters of the three-color light adaptive light source module. S3: The vision camera captures the MARK point, and the coordinate calculation, conversion and transmission module calculates the world coordinates of the battery cell based on the pixel coordinates and converts the world coordinates into laser processing coordinates; S4: The laser processing control module receives the laser processing coordinates and completes high-precision scanning of the laser pattern.

[0026] Optionally, in step S1, the photovoltaic cell loading includes placing the photovoltaic cells on the workbench surface, with the axis of the vision camera of the vision acquisition and processing module perpendicular to the workbench surface, and the lower end of the vision camera being 300mm from the surface of the cells.

[0027] By adopting the above technical solution, the installation position and orientation of the vision camera are clearly defined, setting the optimal imaging benchmark for the system and ensuring that the field of view of the acquired image is fixed and free from perspective distortion. This provides a stable physical basis for the high-precision and repeatable identification and positioning of subsequent MARK points, effectively eliminating measurement errors caused by installation distance or angle deviations.

[0028] In summary, this application includes at least one of the following beneficial technical effects: This invention uses a visual acquisition and processing module to capture the coordinates of the MARK points at the four corners of a photovoltaic cell and stores multiple MARK point capture schemes and corresponding three-color light source parameter information in a preset visual scheme library. When cell recognition fails, the system automatically calls the visual scheme in the preset scheme library and matches a suitable visual scheme based on the shape characteristics of the MARK points and the shape characteristics of the cell, replacing the work of manual inspection, reducing the positioning failure rate and improving work efficiency. Attached Figure Description

[0029] Figure 1 This is a system structure diagram of the present invention; Figure 2 This is a schematic diagram of the workflow of the present invention; Figure 3 This is a schematic diagram of the photovoltaic cell arrangement when the system of the present invention is working.

[0030] Explanation of reference numerals in the attached figures: 1. Preset visual scheme library; 2. Intelligent matching and decision-making module; 3. Three-color adaptive light source module; 4. Visual acquisition and processing module; 5. Coordinate calculation, transformation and transmission module; 6. Laser processing control module. Detailed Implementation

[0031] The present application will be further described in detail below with reference to the accompanying drawings.

[0032] like Figure 1-3 As shown in the figure, this embodiment discloses a photovoltaic cell visual light source adaptive laser processing positioning system, which is mainly used for visual positioning before laser processing of photovoltaic cells. It includes a visual acquisition and processing module 4, a three-color light adaptive light source module 3, a preset visual scheme library 1, an intelligent matching and decision module 2, a coordinate calculation, transformation and transmission module 5, and a laser processing control module 6.

[0033] The visual acquisition and processing module 4 is used to acquire images of photovoltaic cells and identify and capture MARK points at the four corners of the photovoltaic cells. More specifically, the hardware of the visual acquisition and processing module 4 includes a high-resolution image sensor, an industrial camera (vision camera) supporting high-speed continuous shooting, a wide-field fixed-focus lens adapted to different specifications of photovoltaic cells such as 182mm and 210mm, and a laser protective lens. The software includes an image preprocessing unit with noise reduction and distortion correction functions, a MARK point recognition unit with multiple preset visual capture schemes (such as grayscale matching scheme, edge contour matching scheme, and contrast enhancement matching scheme), and a failure judgment unit that automatically triggers a switching mechanism when the "MARK point recognition success rate < threshold (such as 95%)". This failure judgment unit is used to trigger an automatic switching command when MARK point feature capture fails or the error exceeds the limit. Here, MARK point feature capture failure means that the number of MARK points recognized in 3 consecutive frames of images is less than 3, and the error exceeds the limit means that the MARK point positioning error is greater than 10μm. When the visual acquisition and processing module 4 triggers the switching command, it calls the preset visual scheme library 1.

[0034] The three-color adaptive light source module 3 is communicatively connected to the vision acquisition and processing module 4. It stores the light source parameters corresponding to each set of preset visual schemes for MARK point capture and adjusts the light source color and brightness upon receiving a switching command. The light source of the three-color adaptive light source module 3 is a red, blue, and green adjustable light source with a color adjustment wavelength range of 450-660nm and a brightness adjustment level of 0-255. Specifically, the three-color adaptive light source module 3 includes a light source parameter mapping unit and a light source real-time adjustment unit. The light source parameter mapping unit stores the optimal light source color and brightness parameters corresponding to each visual scheme; the light source real-time adjustment unit synchronously outputs a light source control signal upon receiving a switching command.

[0035] The preset visual scheme library 1 is communicatively connected to the visual acquisition and processing module 4. Essentially, it is a relational database of "feature-scheme-light source parameters," used to store multiple visual grasping schemes for different battery cell appearances and different shaped MARK points, as well as the corresponding three-color light source parameters. The stored battery cell appearance information includes the battery cell's color and film thickness.

[0036] The intelligent matching and decision-making module 2 is communicatively connected to the visual acquisition and processing module 4 and the preset visual scheme library 1, respectively. It is not a simple traversal principle, but is used to extract key features from the failed images and perform intelligent matching based on the extracted features and the "feature-scheme" mapping relationship stored in the preset visual scheme library 1 to determine the optimal visual scheme. For example, if the battery cell is found to be dark and has low contrast, the scheme using red light and high brightness will be matched first.

[0037] The intelligent matching and decision-making module 2 includes a feature extraction unit and a scheme matching unit. The feature extraction unit is used to extract key features of the battery cell, including the overall average gray level of the battery cell, the contrast between the MARK point region and the background, and image texture features. The scheme matching unit performs intelligent matching based on the extracted key features and the "feature-scheme" mapping relationship stored in the scheme library.

[0038] In addition, in other embodiments, the intelligent matching and decision-making module 2 further includes a learning and updating unit, which records each successfully matched scheme and the corresponding battery cell features, and adds the corresponding scheme to the preset visual scheme library 1 after the battery cell is successfully located, so as to realize the system's self-learning evolution. The coordinate calculation, transformation, and transmission module 5 is communicatively connected to the vision acquisition and processing module 4. It is used to calculate the world coordinates of the battery cell based on the pixel coordinates of the identified MARK points, convert the world coordinates into the processing coordinates for laser processing, and transmit them to the laser processing control module 6.

[0039] The laser processing control module 6 is communicatively connected to the coordinate calculation, conversion and transmission module 5, and is used to receive the laser processing coordinate data transmitted from the coordinate calculation, conversion and transmission module 5, and control the laser to perform high-precision scanning processing on the battery cell.

[0040] When the number of MARK points identified by the vision camera in three consecutive frames is less than 3, or the positioning error of the MARK points is greater than 10μm, automatic switching will be triggered. The system will automatically call the preset vision scheme library 1 and match the optimal vision grasping scheme according to the appearance characteristics of the battery cell and the shape characteristics of the MARK points. After matching the vision scheme, the preset corresponding light source parameters will be called simultaneously. The three-color light source will complete the color and brightness switching within 500ms. After switching the vision scheme, the vision camera will re-identify and grasp the MARK points for secondary verification. If the identification is successful and the positioning error of the MARK points is qualified, the coordinate calculation process will begin. If it fails, the next set of preset vision schemes will be called again until success is achieved. (In this example, a maximum of 3 schemes can be called for a single battery cell. If all 3 schemes fail, a new vision scheme matching the battery cell needs to be added to the preset vision scheme library 1.)

[0041] The positioning method of this photovoltaic cell vision light source adaptive laser processing positioning system includes the following steps: S1: Photovoltaic cell loading, the vision camera of vision acquisition and processing module 4 uses the initial scheme and corresponding light source parameters to capture the MARK points of the four corners of the cell; S2: Determine whether the MARK point can be successfully identified and captured. If yes, proceed to S3. Otherwise, trigger the automatic switching command, switch the vision scheme, call the preset vision scheme library 1, match the optimal vision scheme according to the shape characteristics of the MARK point and the color and film thickness of the battery cell, and synchronously adjust the three-color light source parameters of the three-color light adaptive light source module 3. S3: The vision camera captures the MARK point, and the coordinate calculation, conversion and transmission module 5 calculates the world coordinates of the battery cell based on the pixel coordinates and converts the world coordinates into laser processing coordinates. S4: The laser processing control module 6 receives the laser processing coordinates and completes high-precision scanning of the laser pattern.

[0042] The following examples of different specifications illustrate this solution in detail.

[0043] In this embodiment, 182mm×182mm battery cells are selected. The processing platform for the vision camera and laser processing control module 6 is constructed as follows: Figure 3 As shown, the following installation requirements must be met: the axis of the vision camera must be perpendicular to the worktable surface, and the lower end of the camera lens must be 300mm away from the surface of the battery cell; the three-color ring light source must be coaxially mounted on the outside of the lens.

[0044] The visual scheme in this embodiment was determined through orthogonal experiments and error verification. The specific process is as follows: 100 identical battery cells were selected, and images were acquired using a vision camera. Algorithm parameters such as grayscale threshold, edge detection threshold, and light source color and brightness were adjusted. The success rate and positioning error of MARK point recognition under different parameter combinations were recorded. An orthogonal experimental design was adopted (the factors are algorithm threshold, light source color, and light source brightness). The optimal parameter combination with a recognition success rate ≥99% and a positioning error of less than 10μm was selected as the preset visual scheme for this type of battery cell.

[0045] This embodiment takes laser grooving of electrodes in the poly-Si stacked region on the back of a 182mm BC battery as an example to illustrate the system workflow in detail. The complete implementation process of this embodiment is as follows: S1: BC battery loading and initial positioning triggering: The BC battery is loaded onto the worktable using a vacuum suction cup and the battery cells are vacuum-adsorbed. The deviation between the center of the battery cell and the center of the worktable is ≤±1mm. The host computer sends a "positioning start command". The color and brightness of the three-color ring light source are set according to the preset vision scheme 1 (SiNx film scheme). The vision camera acquires 3 consecutive frames of MARK point images at the four corners of the BC battery cell according to the preset vision scheme 1 (SiNx film scheme).

[0046] S2: Initial scheme validity judgment. The visual acquisition and processing module 4 processes the image according to the grayscale threshold matching algorithm (threshold 130-190) of the preset visual scheme: Because the back of the BC battery is a poly-Si stack (grayscale value 100-120, lower than the lower threshold), only 2 MARK points are identified, and the recognition success rate is 50% (<75% threshold). The positioning error cannot be calculated. The failure judgment unit determines that "the preset visual scheme 1 (SiNx film scheme) is failed" and generates a "scheme switching trigger signal". The optimal visual scheme matching and light source adjustment (adapting to the poly-Si layer) involves the system calling a preset scheme library and matching the schemes in the preset visual scheme library 1 with the BC battery film layer to determine the desired scheme 2 (poly-Si layer scheme). The system sends the "scheme 2 light source color and brightness parameter command" to the three-color light source. After receiving the command, the three-color adaptive light source module 3 completes the switching of the light source color within 500ms: synchronously switching from "green light 525nm / 190 level" to "red light 650nm / 210 level", and feeding back the "light source ready signal".

[0047] S3: Secondary verification and coordinate calculation of MARK points. After receiving the "light source ready signal", the system triggers the vision camera to acquire three consecutive frames of BC solar cell MARK point images. The vision acquisition and processing module 4 processes the images according to scheme 2 to obtain the pixel coordinates of the four MARK points. The world coordinates of the solar cell are calculated by the coordinate transformation and transmission module: the center coordinates of the solar cell (X=0.45mm, Y=0.65mm), the rotation angle θ=0.06°, and the positioning error 3.2μm. The calculated world coordinates of the solar cell are then converted into laser processing coordinates.

[0048] S4: Coordinate transmission and laser grooving on the back of the BC battery. The laser processing control module 6 receives the laser processing coordinates sent by the coordinate calculation, conversion and transmission module 5, determines the offset and rotation angle of the laser pattern, and begins high-precision laser scanning processing of the battery cell.

[0049] After actual testing, the overall recognition success rate of this processing positioning system was increased to over 99.5%, the entire process required no manual intervention, and the average positioning cycle (including adaptive switching) was only 2.1 seconds.

[0050] Comparative example: The traditional system using a fixed green light source and a single grayscale algorithm has an overall recognition success rate of 78%, and on average, each failure requires 120 seconds of manual intervention and debugging.

[0051] After comparison, this positioning system has achieved significant improvements in recognition success rate, production efficiency, and automation level. It has increased the recognition success rate from 78% to over 99.5%, and reduced the processing time due to recognition failure from 2 minutes of manual debugging to 2.1 seconds of automatic completion, greatly ensuring the continuous, stable, and efficient operation of the production line.

[0052] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A photovoltaic cell visual light source adaptive laser processing positioning system, characterized in that, include: The visual acquisition and processing module (4) is used to acquire images of photovoltaic cells, identify and capture MARK points on the four corners of the photovoltaic cells. It includes a failure judgment unit, which is used to trigger an automatic switching command when the MARK point feature capture fails or the error exceeds the limit. The three-color adaptive light source module (3) is connected to the visual acquisition and processing module (4) for storing the light source parameters corresponding to the preset visual scheme for each set of MARK point capture, and adjusting the light source color and brightness after receiving the switching instruction; A preset visual scheme library (1) is connected to the visual acquisition and processing module (4) for storing multiple visual grasping schemes for different battery cell appearances and different shaped MARK points, and storing the three-color light source parameters of the corresponding schemes; The intelligent matching and decision-making module (2) is connected to the visual acquisition and processing module (4) and the preset visual scheme library (1) respectively. It is used to extract key features from the failed images and perform intelligent matching based on the extracted features and the "feature-scheme" mapping relationship stored in the preset visual scheme library (1) to determine the optimal visual scheme. The coordinate calculation, transformation and transmission module (5) is connected in communication with the vision acquisition and processing module (4) to calculate the world coordinates of the battery cell based on the pixel coordinates of the identified MARK point, convert the world coordinates into the processing coordinates of the laser processing, and transmit them to the laser processing control module (6). The laser processing control module (6) is communicatively connected to the coordinate calculation, conversion and transmission module (5) and is used to receive the laser processing coordinate data transmitted by the coordinate calculation, conversion and transmission module (5) and control the laser to perform high-precision scanning processing on the battery cell.

2. The photovoltaic cell visual light source adaptive laser processing positioning system according to claim 1, characterized in that, The failure to capture MARK points refers to the fact that the number of MARK points identified in three consecutive frames of images is less than 3.

3. The photovoltaic cell visual light source adaptive laser processing positioning system according to claim 1, characterized in that, The error exceeding the limit refers to a MARK point positioning error greater than 10μm.

4. The photovoltaic cell visual light source adaptive laser processing positioning system according to claim 1, characterized in that, The light source of the three-color adaptive light source module (3) is a red, blue and green three-color adjustable light source with a color adjustment wavelength range of 450-660nm and a brightness adjustment level of 0-255.

5. The photovoltaic cell visual light source adaptive laser processing positioning system according to claim 1, characterized in that, The battery cell appearance information stored in the preset visual scheme library (1) includes the color and film thickness of the battery cell.

6. The photovoltaic cell visual light source adaptive laser processing positioning system according to claim 1, characterized in that, The intelligent matching and decision-making module (2) includes a feature extraction unit and a scheme matching unit; The feature extraction unit is used to extract key features of the battery cell, including the overall average gray level of the battery cell, the contrast between the MARK point region and the background, and image texture features. The scheme matching unit performs intelligent matching based on the extracted key features and the "feature-scheme" mapping relationship stored in the scheme library.

7. The photovoltaic cell visual light source adaptive laser processing positioning system according to claim 6, characterized in that, The intelligent matching and decision-making module (2) also includes a learning and updating unit, which records each successfully matched scheme and the corresponding battery cell features, and adds a corresponding scheme to the preset visual scheme library (1) after the battery cell is successfully located, so as to realize the self-learning evolution of the system.

8. The photovoltaic cell visual light source adaptive laser processing positioning system according to claim 1, characterized in that, The tri-color adaptive light source module (3) includes a light source parameter mapping unit and a light source real-time adjustment unit; The light source parameter mapping unit is used to store the optimal light source color and brightness parameters corresponding to each visual scheme; The real-time light source adjustment unit is used to synchronously output a light source control signal when receiving a switching command.

9. A method for adaptive laser processing positioning of photovoltaic cells using a visual light source, which applies the positioning system described in any one of claims 1-8, characterized in that, Includes the following steps: S1: Photovoltaic cell loading, the vision camera of the vision acquisition and processing module (4) uses the initial scheme and corresponding light source parameters to capture the MARK points of the four corners of the cell; S2: Determine whether the MARK point can be successfully identified and captured. If yes, proceed to S3. Otherwise, trigger the automatic switching instruction, switch the vision scheme, call the preset vision scheme library (1), match the optimal vision scheme according to the shape characteristics of the MARK point and the color and film thickness of the battery cell, and synchronously adjust the three-color light source parameters of the three-color light adaptive light source module (3). S3: The vision camera captures the MARK point, and the coordinate calculation, conversion and transmission module (5) calculates the world coordinates of the battery cell based on the pixel coordinates and converts the world coordinates into laser processing coordinates. S4: The laser processing control module (6) receives the laser processing coordinates and completes high-precision scanning of the laser pattern.

10. The adaptive laser processing positioning method for photovoltaic cell visual light source according to claim 9, characterized in that, In step S1, the photovoltaic cell loading includes placing the photovoltaic cell on the workbench surface, with the vision camera axis of the vision acquisition and processing module (4) perpendicular to the workbench surface, and the lower end of the vision camera being 300mm from the surface of the cell.

Images