Whole-process intelligent sensing data analysis system and method thereof

Abstract

The invention provides a whole-process intelligent perception data analysis system and method, the system comprises a data acquisition module, an image processing module, a defect identification module, a data analysis module and a control module, the data acquisition module is used for acquiring defect data of a solar cell, a machine number and working data of each machine; the image processing module comprises an image preprocessing unit, an edge optimization unit and a morphological processing unit; a near-infrared laser, an area array visible light, a PL sensor and an EL sensor are arranged in the defect identification module; the data analysis module is used for integrating and analyzing various kinds of collected data and calculating the defect rate within 24 hours; the control module adjusts production parameters, optimizes the technological process and the like according to the data analysis result, and when the control module detects abnormity, the control module sends early warning information to an operator.

Description

Technical Field

[0001] This invention relates to a whole-process intelligent sensing data analysis system and method. Background Technology

[0002] With the global energy structure transformation and the increasing awareness of environmental protection, solar energy, as a clean and efficient renewable energy source, has received widespread attention. As the core component of a solar power generation system, the quality of solar cells directly affects the performance and lifespan of the entire system.

[0003] However, traditional solar cell testing methods have many limitations. Currently, most manufacturers have not established standardized evaluation systems, making it impossible to achieve comprehensive monitoring and statistical analysis of surface and internal defects in solar cells across the entire factory.

[0004] As the automation level of solar cell production continues to increase, the tools and process steps involved in cell contact increase, making quality control more complex. The lack of comprehensive monitoring and data analysis capabilities makes it difficult to detect and handle potential quality problems in a timely manner.

[0005] Therefore, a full-process intelligent sensing data analysis system and method are needed to solve the problem that manufacturers lack comprehensive monitoring and data analysis capabilities in existing technologies. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide a full-process intelligent sensing data analysis system and method. This system establishes a defect data and work data acquisition mechanism, enabling full-process monitoring of the production process. This effectively improves the controllability and quality management level of the production process, allowing potential quality problems to be detected and addressed in a timely manner, and reducing downtime and maintenance costs caused by quality issues. This objective is achieved as follows: This invention proposes a full-process intelligent sensing data analysis system, comprising a data acquisition module, an image processing module, a defect identification module, a data analysis module, and a control module. The data acquisition module collects defect data from solar cells, including the number of microcracks (PL), black edges (PL), scratches, corner blackenings (PL), radon aluminum winding plating cycles, and contamination. It also collects machine numbers and operating data for each machine, including the number of slurry leaks, grid line transfers, grid breaks, wire thickenings, scratches, and missing solder joints. The image processing module includes an image preprocessing unit, an edge optimization unit, and a morphological processing unit. The defect identification module integrates near-infrared laser, area array visible light, PL sensor, and EL sensor. The data analysis module integrates and analyzes the collected data, calculating the 24-hour defect rate. The control module adjusts production parameters and optimizes the process flow based on the data analysis results. When the control module detects an anomaly, it sends a warning message to the operator.

[0007] Furthermore, the image preprocessing unit performs preliminary processing on the acquired image to remove small-area noise; the edge optimization unit optimizes the edges of the segmented sub-images using morphological methods to reduce the impact of edge gaps on subsequent defect identification; the morphological processing unit truncates the pixels of the top, bottom, left, and right edges of the battery sub-image with a preset width to reduce interference caused by miscutting.

[0008] Furthermore, the defect identification module processes the acquired images using a hybrid image processing algorithm to obtain key defect features, and uses a multimodal data fusion method to classify, locate, and label defects such as cracks, broken grids, scratches, dents, poor welding, and hot spots.

[0009] Furthermore, the full-process intelligent sensing data analysis system uses a full-process intelligent sensing data analysis method, which specifically includes: S1, a data acquisition module for collecting defect data of solar cells, including the number of microcracks, black edges, scratches, corner blackenings, radon aluminum winding plating, and dirt accumulation, as well as machine numbers and working data for each machine, including the number of slurry leaks, grid line transfers, grid breaks, wire thickenings, scratches, and missing solder joints; S2, an image processing module for preprocessing the collected image data; S3, a defect identification module for identifying defects in the preprocessed images; S4, a data analysis module for integrating and analyzing the collected data and calculating the 24-hour defect rate; S5, a control module for adjusting production parameters and optimizing the process flow based on the data analysis results, and sending a warning message to the operator when the control module detects an anomaly.

[0010] Furthermore, the steps in S2 specifically include: S201, obtaining the pixel value sum of the entire image through horizontal and vertical projection of the battery cell image, and the pixel value sum having an extreme point at the edge position or gap position. If the extreme point is higher than a certain set threshold, it is regarded as the edge position of the battery cell or the gap position of the sub-cell. S202. Divide the image of the entire solar cell into multiple sub-images, accumulate the pixel values ​​and store them as a vector, and invert the vector to find the position of the maximum value as the edge position, so as to obtain the coordinates of the top, bottom and left and right edges of the solar cell. S203. Threshold segmentation is performed on the segmented sub-image to convert it into a binary image. Therefore, the gray values ​​of the 10 surrounding pixels are set to 0. S204. Optimize the edges of the segmented sub-images using morphological methods, specifically including: determining the horizontal edges using a peak function and setting the minimum interval between peaks as the width of the sub-image; determining the vertical edges using a peak function and setting the minimum interval between peaks as the length of the sub-image; and removing small-area noise through morphological processing.

[0011] Furthermore, the steps in S4 specifically include: S401, performing statistical analysis on the collected defect data, including defect occurrence frequency and production process data; S402. Perform correlation analysis between image features and defect data to obtain comprehensive features; S403. Calculate the 24-hour defect rate based on comprehensive characteristics and generate a quality report; S404. Conduct data analysis based on the quality report.

[0012] Compared with the prior art, the beneficial effects of the present invention are: the present invention establishes a defect data and working data collection mechanism, realizes full-process monitoring of the production process, and effectively improves the controllability of the production process and the level of quality management; Through real-time monitoring and data analysis, potential quality issues can be detected and addressed in a timely manner, reducing downtime and maintenance costs caused by quality problems. Detailed Implementation

[0013] To enhance understanding of the present invention, the present invention will be further described in detail below with reference to embodiments. These embodiments are only used to explain the present invention and do not constitute a limitation on the scope of protection of the present invention.

[0014] This invention proposes a full-process intelligent sensing data analysis system, comprising a data acquisition module, an image processing module, a defect identification module, a data analysis module, and a control module. The data acquisition module collects defect data from solar cells, including the number of microcracks (PL), black edges (PL), scratches, corner blackenings (PL), radon aluminum winding plating cycles, and contamination counts, as well as the machine number and operating data for each machine, including the number of slurry leaks, grid line transfers, grid breaks, wire thickness issues, scratches, and missing solder joints. The image processing module includes an image preprocessing unit, an edge optimization unit, and a morphological processing unit. The defect identification module integrates near-infrared laser, area array visible light, PL sensor, and EL sensor. The data analysis module integrates and analyzes the collected data. The system analyzes and calculates the defect rate over 24 hours. Based on the data analysis results, the control module adjusts production parameters and optimizes the process flow. When the control module detects an anomaly, it sends a warning message to the operator. The image preprocessing unit performs preliminary processing on the acquired images to remove small-area noise. The edge optimization unit optimizes the edges of the segmented sub-piece images using morphological methods to reduce the impact of edge gaps on subsequent defect identification. The morphological processing unit truncates the pixels of the top, bottom, left, and right edges of the battery sub-pieces with a preset width to reduce interference caused by miscutting. The defect identification module processes the acquired images using a hybrid image processing algorithm to obtain key defect features and uses a multimodal data fusion method to classify, locate, and label defects such as cracks, broken grids, scratches, dents, poor welding, and hot spots.

[0015] Through the solution in this embodiment, the system of the present invention can simultaneously detect multiple types of defects such as PL microcracks, PL black edges, scratches, PL corner black, radon aluminum immersion plating, and dirt, breaking through the limitation of existing equipment that can only detect a single type of defect, and meeting the diverse detection needs in complex production processes; and by using image processing technology, through the effective fusion of multimodal data, it effectively removes interfering defects and noise in the image, improving the precision and accuracy of defect detection.

[0016] Another embodiment of the present invention proposes a full-process intelligent sensing data analysis method, which specifically includes: S1, a data acquisition module for collecting defect data of solar cells, including the number of microcracks, black edges, scratches, corner blackening, radon aluminum winding plating, and dirt accumulation, as well as the machine number and working data of each machine, including the number of slurry leakage, grid line transfer, grid breakage, wire thickening, scratches, and missing solder joints; S2, an image processing module for preprocessing the collected image data; S3, a defect identification module for identifying defects in the preprocessed images; S4, a data analysis module for integrating and analyzing the collected data and calculating the 24-hour defect rate; S5, a control module for adjusting production parameters and optimizing the process flow based on the data analysis results, and sending a warning message to the operator when the control module detects an anomaly.

[0017] The steps in S2 specifically include: S201, obtaining the sum of pixel values ​​of the entire image through horizontal and vertical projection of the solar cell image. Extreme points appear at edge or gap locations in the sum of pixel values. If an extreme point exceeds a certain set threshold, it is considered an edge location of the solar cell or a gap location of a sub-cell. S202, dividing the entire solar cell image into multiple sub-cell images, storing the sum of pixel values ​​as a vector, and inverting the vector to find the location of the maximum value, thus obtaining the coordinates of the top, bottom, left, and right edges of the solar cell. S203, performing thresholding on the segmented sub-cell images to convert them into binary images, therefore setting the grayscale values ​​of the surrounding 10 pixels to 0. S204, optimizing the edges of the segmented sub-cell images using morphological methods, specifically including: determining the horizontal edges using a peak function, setting the minimum interval between peaks to the width of the sub-cell image; determining the vertical edges using a peak function, setting the minimum interval between peaks to the length of the sub-cell image; and removing small-area noise through morphological processing.

[0018] The steps in S4 specifically include: S401, performing statistical analysis on the collected defect data, including defect occurrence frequency and production process data; S402, performing correlation analysis between image features and defect data to obtain comprehensive features; S403, calculating the 24-hour defect rate based on the comprehensive features and generating a quality report; S404, performing data analysis based on the quality report.

[0019] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A full-process intelligent sensing data analysis system, characterized in that, The system includes a data acquisition module, an image processing module, a defect identification module, a data analysis module, and a control module. The data acquisition module collects defect data from solar cells, including the number of microcracks (PL), black edges (PL), scratches, corner blackenings (PL), radon aluminum winding plating cycles, and dirt accumulations. It also collects machine numbers and operating data for each machine, including the number of slurry leaks, grid line transfers, grid breaks, wire thickenings, scratches, and missing solder joints. The image processing module includes an image preprocessing unit, an edge optimization unit, and a morphological processing unit. The defect identification module incorporates near-infrared laser, area array visible light, PL, and EL sensors. The data analysis module integrates and analyzes the collected data, calculating the 24-hour defect rate. The control module adjusts production parameters and optimizes the process flow based on the data analysis results. When the control module detects an anomaly, it sends a warning message to the operator.

2. The intelligent sensing data analysis system for the entire process according to claim 1, characterized in that, The image preprocessing unit performs preliminary processing on the acquired image to remove small-area noise; the edge optimization unit optimizes the edges of the segmented sub-images using morphological methods to reduce the impact of edge gaps on subsequent defect identification; the morphological processing unit truncates the pixels of the top, bottom, left, and right edges of the battery sub-image with a preset width to reduce interference caused by miscutting.

3. The intelligent sensing data analysis system for the entire process according to claim 2, characterized in that, The defect identification module processes the acquired images using a hybrid image processing algorithm to obtain key defect features, and uses a multimodal data fusion method to classify, locate, and label defects such as cracks, broken grids, scratches, dents, poor welding, and hot spots.

4. The intelligent sensing data analysis system for the entire process according to claim 3, characterized in that, The entire process intelligent sensing data analysis system uses an entire process intelligent sensing data analysis method, which specifically includes: S1, a data acquisition module for collecting defect data of solar cells, including the number of microcracks, black edges, scratches, corner blackening, radon aluminum winding plating, and dirt accumulation, as well as machine number and working data for each machine, including the number of slurry leakage, grid line transfer, grid breakage, wire thickening, scratches, and missing solder joints; S2, an image processing module for preprocessing the collected image data; S3, a defect identification module for identifying defects in the preprocessed images; S4, a data analysis module for integrating and analyzing the collected data and calculating the 24-hour defect rate; S5, a control module for adjusting production parameters and optimizing the process flow based on the data analysis results, and sending a warning message to the operator when the control module detects an anomaly.

5. The intelligent sensing data analysis system for the entire process according to claim 4, characterized in that, The steps in S2 specifically include: S201, obtaining the pixel value sum of the entire image through horizontal and vertical projection of the battery cell image, and the pixel value sum has extreme points at the edge position or gap position. If the extreme point is higher than a certain set threshold, it is regarded as the edge position of the battery cell or the gap position of the sub-cell. S202. Divide the image of the entire solar cell into multiple sub-images, accumulate the pixel values ​​and store them as a vector, and invert the vector to find the position of the maximum value as the edge position, so as to obtain the coordinates of the top, bottom and left and right edges of the solar cell. S203. Threshold segmentation is performed on the segmented sub-image to convert it into a binary image. Therefore, the gray values ​​of the 10 surrounding pixels are set to 0. S204. Optimize the edges of the segmented sub-images using morphological methods, specifically including: determining the horizontal edges using a peak function and setting the minimum interval between peaks as the width of the sub-image; determining the vertical edges using a peak function and setting the minimum interval between peaks as the length of the sub-image; and removing small-area noise through morphological processing.

6. The intelligent sensing data analysis system for the entire process according to claim 5, characterized in that, The steps in S4 specifically include: S401, performing statistical analysis on the collected defect data, including defect occurrence frequency and production process data; S402. Perform correlation analysis between image features and defect data to obtain comprehensive features; S403. Calculate the 24-hour defect rate based on comprehensive characteristics and generate a quality report; S404. Conduct data analysis based on the quality report.