Perovskite-coated foreign object detection system and method

Abstract

The application relates to a perovskite coating foreign matter detection system and method, a plurality of 3D visual detectors are arranged side by side above a bearing platform along the width direction of a substrate, the scanning visual fields of adjacent detectors partially overlap, and the total coverage width is not less than the substrate width. A processing assembly fits a substrate plane according to three-dimensional point cloud data and identifies a target area protruding relative to the plane. According to the scheme, the overall coverage scanning of a large-width coating substrate is realized through the side-by-side and partially-overlapping arrangement of the plurality of 3D visual detectors, the problem of insufficient visual field of a single sensor and the precision reduction caused by light attenuation under a long span are solved. Moreover, the fitting of a plane and the identification of protruding foreign matters in combination with point cloud data can reliably detect tiny particles with a height greater than or equal to 50 microns, avoid the interference of the tiny particles with a coating die, and guarantee the coating quality and equipment safety.

Description

Technical Field

[0001] This application relates to the field of solar cell technology, and in particular to a perovskite coating foreign object detection system and method. Background Technology

[0002] Currently, with the rapid development of science and technology, perovskite solar cells are gradually being developed and applied due to their high theoretical conversion efficiency. In the industrial production of perovskite solar cells, slot coating is the core process for preparing a uniform, large-area perovskite photosensitive layer. As production capacity increases, the coating area is also gradually increasing, which, while improving coating efficiency, also places more stringent requirements on coating cleanliness.

[0003] During the coating process, the gap between the coating die and the glass substrate is extremely small, as small as 50 μm. If hard protruding foreign objects are attached to the surface of the glass substrate during the cleaving, transfer or loading process, when the height of the foreign object exceeds 50 μm, it will directly interfere rigidly with the die, causing scratches on the die, coating line breaks, film layer Mura defects, etc., which may lead to the scrapping of the die and the shutdown of the production line in severe cases.

[0004] Therefore, there is an urgent need for a high-precision foreign object identification scheme to identify larger foreign object particles before coating. Summary of the Invention

[0005] Therefore, it is necessary to propose a foreign object detection system and method for perovskite coating to accurately identify large-sized foreign objects on the glass substrate before coating, so as to ensure the safe and stable coating process.

[0006] This application provides a foreign object detection system for perovskite coating, including a support platform, a coating die, a vision recognition component, and a processing component. The support platform is used to place the substrate to be coated and to move the substrate along the coating direction. The coating die is fixedly disposed above the support platform. The vision recognition component includes multiple 3D vision detectors, which are fixedly disposed side by side above the support platform along the width direction of the substrate to be coated. The scanning field of view of two adjacent 3D vision detectors at least partially overlaps, and the scanning coverage width of the multiple 3D vision detectors is greater than or equal to the width of the substrate to be coated, and the width direction is perpendicular to the coating direction. The 3D vision detectors are used to scan the substrate to be coated as it moves toward the coating die along the coating direction to determine three-dimensional point cloud data. Each 3D vision detector is connected to the processing component, which is used to fit the substrate plane according to the three-dimensional point cloud data, determine the geometric parameters of the target area protruding relative to the substrate plane, and determine the foreign object detection result according to the geometric parameters.

[0007] In one embodiment, a plurality of the 3D vision detectors are arranged coaxially in a straight line along the width direction of the substrate to be coated, and the mounting height of each 3D vision detector is consistent.

[0008] In one embodiment, the plurality of the 3D vision detectors are arranged at equal intervals.

[0009] In one embodiment, the overlap width of the scanning field of view of two adjacent 3D vision detectors is greater than or equal to 1 mm.

[0010] In one embodiment, the distance between the visual recognition component and the coating die head is greater than or equal to a preset spacing, wherein the preset spacing is greater than or equal to the product of the moving speed of the substrate to be coated and the processing time, and the processing time represents the total processing time from the start of a single acquisition to the completion of foreign object recognition by the processing component.

[0011] In one embodiment, the scanning coverage width is determined based on the total scanning field width of the 3D vision detector, the overlap width of the scanning field ranges of two adjacent 3D vision detectors, and the substrate positioning error compensation parameters.

[0012] In one embodiment, the substrate positioning error compensation parameters include angular error parameters and horizontal positioning error parameters.

[0013] In one embodiment, the included angle error parameter is determined based on the length of the substrate to be coated, the width of the substrate to be coated, and a preset positioning error angle.

[0014] This application also provides a method for detecting foreign objects in perovskite coating based on the above-mentioned perovskite coating foreign object detection system, comprising: acquiring three-dimensional point cloud data of the substrate to be coated; fitting and determining the substrate plane based on the three-dimensional point cloud data, and determining the geometric parameters of the target area protruding relative to the substrate plane; and determining the foreign object detection result based on the geometric parameters.

[0015] In one embodiment, the geometric parameters include the protrusion height, and determining the foreign object detection result based on the geometric parameters includes: if there is a target protrusion height greater than or equal to a preset height threshold, determining the location information of the target area corresponding to the target protrusion height; and outputting the foreign object detection result based on the location information and the target protrusion height.

[0016] The aforementioned perovskite coating foreign object detection system and method involves arranging multiple 3D vision detectors side-by-side along the width of the substrate above a support platform. The scanning fields of adjacent detectors partially overlap, ensuring the total coverage width is not less than the substrate width. The processing component fits the substrate plane based on 3D point cloud data and identifies target areas protruding relative to that plane. This approach, through the side-by-side arrangement of multiple 3D vision detectors with partially overlapping fields of view, achieves comprehensive coverage scanning of a wide-width coating substrate, solving the problems of insufficient field of view of a single sensor and decreased accuracy due to light attenuation over long spans. Furthermore, by combining point cloud data with plane fitting and identification of protruding foreign objects, it can reliably detect microparticles with a height greater than or equal to 50 micrometers, preventing interference with the coating die and ensuring coating quality and equipment safety. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the conventional technology, the drawings used in the description of the embodiments or the conventional technology will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the perovskite coating foreign object detection system in one embodiment of this application;

[0019] Figure 2 This is a three-dimensional point cloud image in one embodiment of this application;

[0020] Figure 3 This is a schematic diagram of three-dimensional point cloud number fitting in one embodiment of this application;

[0021] Figure 4 This is a schematic diagram of a foreign object detection method for perovskite coating in one embodiment of the application;

[0022] Figure 5 This is a schematic diagram of the foreign object detection method for perovskite coating in another embodiment of the application. Detailed Implementation

[0023] To facilitate understanding of this application, a more complete description will be provided below with reference to the accompanying drawings. Preferred embodiments of this application are shown in the drawings. However, this application can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a thorough and complete understanding of the disclosure of this application.

[0024] Please see Figure 1This application provides a foreign object detection system for perovskite coating, including a support platform (not shown), a coating die 110, a vision recognition component 120, and a processing component 130. The support platform is used to place the substrate 200 to be coated and to move the substrate 200 along the coating direction. The coating die 110 is fixedly disposed above the support platform. The vision recognition component 120 includes multiple 3D vision detectors 101, which are fixedly disposed side by side above the support platform along the width direction of the substrate 200 to be coated. The scanning field of view of two adjacent 3D vision detectors 101 is... The scanning coverage width of the multiple 3D vision detectors 101 is greater than or equal to the width of the substrate 200 to be coated, and the width direction is perpendicular to the coating direction. The 3D vision detectors 101 are used to scan the substrate 200 to be coated as it moves toward the coating die head 110 along the coating direction to determine three-dimensional point cloud data. Each 3D vision detector 101 is connected to a processing component 130. The processing component 130 is used to fit the substrate plane according to the three-dimensional point cloud data, determine the geometric parameters of the target area protruding relative to the substrate plane, and determine the foreign object detection result according to the geometric parameters.

[0025] In the embodiments of this application, the length can be understood as the distance in the direction parallel to the coating direction, while the width can be understood as the distance on the substrate plane in the direction perpendicular to the coating direction.

[0026] The support platform refers to the component used to place the substrate 200 to be coated (such as a 2.5m long glass) and to move the substrate 200 along the coating direction. The substrate 200 to be coated is the glass substrate that needs to be coated with perovskite. The coating die 110 is also known as the slit coating die 110, and the minimum gap between it and the substrate 200 to be coated can be 50um. If the height of foreign objects exceeds this gap, interference will occur.

[0027] The visual recognition component 120 is an assembly consisting of multiple 3D vision detectors 101 mounted side-by-side, used for scanning the substrate surface. Each 3D vision detector 101 is a high-precision 3D scanning camera based on the triangulation principle. Since a single detector has a small field of view, multiple detectors are needed to achieve wide-span coverage. The scanning field of view, i.e., the width that a single 3D vision detector 101 can cover on the coated substrate in a single scan, is usually related to height and is configured according to the actual scenario, only needing to balance scanning accuracy. The processing component 130 refers to the computing unit that receives the 3D point cloud data collected by each 3D vision detector 101 and performs image processing and logical judgments such as plane fitting, protrusion recognition, and height measurement.

[0028] 3D point cloud data refers to 3D image data synthesized by the 3D vision detector 101 at a certain acquisition frequency. It contains the spatial coordinate information of each point on the substrate surface. For details, please refer to [reference needed]. Figure 2 The substrate plane refers to the ideal plane obtained by fitting the point cloud data through the processing component 130, which serves as a reference plane for calculating the height of the foreign object. For example, see the attached document. Figure 3 The target area refers to the region on the substrate surface that is significantly raised relative to the fitted plane, i.e., the location of the suspected foreign object, for example... Figure 3 The diagram shows areas A and B. Geometric parameters include quantifiable shape and size information such as the object's height, length, and width.

[0029] In some embodiments, the foreign object identification results output by the processing component 130 include whether there is a foreign object exceeding a set height threshold, as well as the location coordinates, height, length and width dimensions of the foreign object, and can generate a LOG report to the host computer to determine whether to stop coating and issue an alarm.

[0030] A support platform is used to place the substrate 200 to be coated and to move it along the coating direction. A coating die 110 is fixedly mounted above the support platform, maintaining a minimum gap of 50 μm between it and the substrate. The visual recognition component 120 consists of multiple 3D vision detectors 101, which are fixedly mounted side-by-side above the support platform along the width direction of the substrate 200. The scanning fields of adjacent 3D vision detectors 101 are designed to at least partially overlap, such that the total scanning coverage width of the multiple detectors is greater than or equal to the width of the substrate 200 to be coated.

[0031] As the substrate 200 to be coated moves toward the coating die 110 along the coating direction, the 3D vision detector 101 scans the substrate surface to acquire three-dimensional point cloud data. The processing component 130 receives the data from each detector, first fits the substrate plane based on the point cloud data, then identifies the target area that protrudes relative to the plane, and measures the geometric parameters of the area (such as the height, length and width of the foreign object).

[0032] Finally, the processing component 130 determines the presence of dangerous foreign objects based on whether the geometric parameters exceed a set threshold and outputs the foreign object detection result. This solution solves the problem of a sharp drop in accuracy caused by light attenuation and the triangulation principle when a single sensor covers a span of more than 2.4m by using multiple sensors spliced ​​together to cover a long-span substrate. It can reliably detect tiny foreign objects with a height ≥50μm, providing early warning before coating, avoiding interference between foreign objects and the coating die 110, and preventing scratches or Mura defects on the die.

[0033] The aforementioned perovskite coating foreign object detection system arranges multiple 3D vision detectors 101 side-by-side along the width of the substrate above the support platform. The scanning fields of adjacent detectors partially overlap, ensuring that the total coverage width is not less than the substrate width. The processing component 130 fits the substrate plane based on 3D point cloud data and identifies target areas protruding relative to this plane. This solution, by arranging multiple 3D vision detectors 101 side-by-side with partially overlapping fields of view, achieves comprehensive coverage scanning of a large-width coating substrate, solving the problems of insufficient field of view of a single sensor and decreased accuracy due to light attenuation over long spans. Furthermore, by combining point cloud data with plane fitting and identifying protruding foreign objects, it can reliably detect microparticles with a height greater than or equal to 50 micrometers, avoiding interference with the coating die 110 and ensuring coating quality and equipment safety.

[0034] In one embodiment, a plurality of 3D vision detectors 101 are arranged coaxially in a straight line along the width direction of the substrate 200 to be coated, and the mounting height of each 3D vision detector 101 is consistent.

[0035] Linear coaxial arrangement refers to the fact that the mounting positions of multiple 3D vision detectors 101 are distributed in a straight line along the width direction of the substrate, and the optical axes of each 3D vision detector 101 are parallel to each other and all point perpendicularly to the support platform. Consistent mounting height means that each 3D vision detector 101 is at the same vertical distance from the support platform (or substrate surface).

[0036] In this embodiment, multiple 3D vision detectors 101 are arranged coaxially in a straight line along the width direction of the substrate 200 to be coated, and the mounting height of each 3D vision detector 101 is the same. For example, the mounting height can be set to 185 mm, but the specific height needs to be determined in conjunction with the type of 3D vision detector 101.

[0037] This layout ensures that the 3D point cloud data acquired by each sensor has a unified spatial reference coordinate system, and the overlapping areas of adjacent sensors are not offset in the height direction, facilitating subsequent point cloud stitching and planar fitting. If the installation height is inconsistent or the arrangement is not linear, it will cause height reference deviations on the same substrate surface in different sensor coverage areas, resulting in distortion of the fitted substrate plane and affecting the measurement accuracy of the foreign object height.

[0038] In the above scheme, multiple 3D vision detectors 101 are arranged coaxially along the width direction and installed at the same height, which ensures that the spatial reference is consistent when each sensor collects data, avoids measurement errors caused by installation misalignment, thereby improving the consistency and accuracy of foreign object height detection, and providing a reliable three-dimensional point cloud basis for subsequent plane fitting and protrusion judgment.

[0039] In one embodiment, a plurality of 3D vision detectors 101 are arranged at equal intervals.

[0040] Equal spacing refers to maintaining the same installation distance (center distance) between two adjacent 3D vision detectors 101. For example, in one embodiment, the spacing between two adjacent 3D vision detectors 101 can be configured as 41mm. Equal spacing ensures a uniform overlap width between adjacent sensors, simplifying the installation and debugging process and allowing for consistent setting of overlap parameters in the image stitching algorithm. Furthermore, equal spacing, combined with the defined overlap area, enables stable and complete full-width coverage with a minimum number of sensors, while also facilitating batch calibration and maintenance.

[0041] The above scheme arranges the 3D vision detectors 101 at equal intervals, which can make the width of the overlapping area uniformly distributed, simplifying the installation process and splicing algorithm. At the same time, it avoids local scanning blind spots or data redundancy caused by uneven spacing, and further ensures the consistency and stability of foreign object recognition capability across the entire width range.

[0042] In one embodiment, the overlap width of the scanning field of view of two adjacent 3D vision detectors 101 is greater than or equal to 1 mm.

[0043] The overlap width refers to the width of the overlapping area between two adjacent 3D vision detectors 101 in the width direction of the substrate. The overlap area is mainly used to compensate for errors caused by part size, installation angle, etc. during the installation process. In one embodiment, the overlap width can be determined according to the scanning field of view of the 3D vision detector 101. For example, the overlap width can be configured as 5% of the scanning field of view width.

[0044] Setting the lower limit of the overlap width to 1mm can effectively prevent scanning gaps between adjacent sensors caused by factors such as machining tolerances, assembly deviations or temperature deformation, and ensure that each point in the width direction of the substrate to be coated is covered by at least one sensor.

[0045] The above solution has an overlap width of more than or equal to 1mm between adjacent sensor fields of view, which can effectively compensate for the scanning gap caused by installation tolerances, part size and angle deviations, and ensure that at least one sensor completely covers any position on the substrate surface, eliminating the omission of foreign objects due to splicing gaps. It is especially suitable for high-precision and large-span coating inspection needs.

[0046] In one embodiment, the distance between the visual recognition component 120 and the coating die head 110 is greater than or equal to a preset spacing, wherein the preset spacing is greater than or equal to the product of the moving speed of the substrate 200 to be coated and the processing time, and the processing time represents the total processing time from the start of a single acquisition to the completion of foreign object recognition by the processing component 130.

[0047] The preset spacing refers to the horizontal distance between the visual recognition component 120 (i.e., the 3D vision detector 101 closest to the coating die 110) and the coating die 110. For example, in one embodiment, the preset spacing can be configured as 450 mm. The moving speed refers to the maximum speed at which the substrate 200 to be coated moves along the coating direction. The processing time is the total processing time required from the start of a single acquisition until the processing component 130 completes foreign object recognition and outputs the result. For example, it may include visual processing time T, visual image processing time t1, and visual communication time t2; correspondingly, the total processing time is T + t1 + t2.

[0048] In this embodiment, the distance between the visual recognition component 120 and the coating die head 110 is greater than or equal to a preset spacing, which is at least equal to the product of the moving speed of the substrate 200 to be coated and the processing time. This spacing ensures that when the substrate moves at its highest speed, the entire process, from the sensor scanning the foreign object, the processing component 130 completing image synthesis, plane fitting, protrusion recognition, height measurement, and logical judgment, to issuing a stop signal or alarm, can be completed before the substrate reaches the coating die head 110.

[0049] In the above scheme, the distance between the visual recognition component 120 and the coating die head 110 is greater than or equal to the product of the moving speed and the processing time, ensuring that there is sufficient margin before the coating reaches the die head. This enables the system to complete the entire process of scanning, synthesizing 3D images, identifying foreign objects, and outputting alarms, achieving early identification and early intervention, and effectively preventing physical interference between foreign objects and the die head.

[0050] In one embodiment, the scanning coverage width is determined based on the total scanning field width of the 3D vision detector 101, the overlap width of the scanning field ranges of two adjacent 3D vision detectors 101, and the substrate positioning error compensation parameters.

[0051] The scanning coverage width refers to the total width that can be effectively scanned in the width direction of the substrate after all 3D vision detectors 101 are arranged side by side, and it needs to be greater than or equal to the width of the substrate 200 to be coated. The total scanning field of view width refers to the sum of the scanning field of view widths of all 3D vision detectors 101, i.e., H×W (H is the number of 3D vision detectors 101, and W is the scanning field of view width of a single 3D vision detector). The substrate positioning error compensation parameter is used to correct the increased scanning coverage requirement caused by inaccurate positioning of the substrate on the carrier platform, including angular error and horizontal positioning error.

[0052] The scanning coverage width needs to be determined based on the total scanning field of view width of the 3D vision detector 101, the overlap width of two adjacent detectors, and the substrate positioning error compensation parameter. For example, in one embodiment, the constraint formula should be satisfied: K ≤ H × W - (H-1) × w1 + N, where N is the substrate positioning error compensation parameter, H represents the number of 3D vision detectors 101, W represents the scanning field of view width of a single 3D vision detector 101, w1 represents the overlap width of the scanning field of view ranges of two adjacent 3D vision detectors 101, and K represents the width of the substrate 200 to be coated. This formula indicates that the effective coverage width is equal to the sum of the field of view widths of all sensors, minus the width lost due to overlap, plus the error compensation. If only the ideal case (no positioning error) is considered, then N = 0.

[0053] However, due to angular deviations and horizontal positioning accuracy issues in actual production, additional coverage width is required to compensate for these deviations. This feature ensures that even if the substrate is placed at an angle or not fully against the edge, its entire width still falls within the scanning coverage area, avoiding missed edge detection due to positioning errors.

[0054] The above scheme calculates the scanning coverage width based on the total field of view of the sensor, the overlap width, and the substrate positioning error. It can dynamically compensate for the parallel deviation and positioning accuracy error that may occur during the substrate positioning process, ensuring that even if the substrate has a certain angle or horizontal offset, foreign objects can still be completely covered and identified, thus improving the system's adaptability to actual working conditions.

[0055] In one embodiment, the substrate positioning error compensation parameters include angular error parameters and horizontal positioning error parameters.

[0056] The included angle error parameter refers to the angular deviation caused by the non-parallelism between the substrate 200 to be coated and the camera connection line, and the compensation amount for the coverage width caused by this angular deviation. The horizontal positioning error parameter is the positioning accuracy error of the substrate along the width direction when it is positioned near the edge.

[0057] When the glass (substrate 200 to be coated) is transferred to the marble countertop (i.e., the support platform), although it is positioned along the edge, it may still not be parallel to the camera line, creating an angle. This angle will cause an additional requirement w3 for the scanning coverage width, which is the angle error parameter. Simultaneously, the horizontal positioning error parameter (reference value 0.05mm) will also affect the sensor quantity assessment. Together, these two parameters constitute the substrate positioning error compensation parameter N. Without angle error compensation, when the substrate is skewed, its actual projected length in the width direction will change, potentially causing the substrate edge to exceed the scanning range of the outermost sensor. By introducing these two compensation parameters, the system can reserve margins in advance when designing the number and layout of sensors, ensuring that positioning fluctuations in actual production will not cause missed foreign object detection.

[0058] The above solution introduces angular error and horizontal positioning error as compensation parameters, making the number and layout design of sensors more consistent with the positioning fluctuations in actual production, avoiding missed detection in edge areas caused by substrate tilting or transport deviation, thereby ensuring the integrity of foreign object detection on long-distance coated substrates without increasing the number of sensors.

[0059] In one embodiment, the included angle error parameter is determined based on the length of the substrate 200 to be coated, the width of the substrate 200 to be coated, and a preset positioning error included angle.

[0060] The preset positioning error angle refers to the angular deviation value preset based on the actual positioning accuracy of the device. For example, in one embodiment, the formula for calculating the angle error parameter w3 is: w3=|L×cosQ°+M×sinQ°|, where L is the length of the substrate 200 to be coated, M is the width of the substrate 200 to be coated, and Q is the preset positioning error angle.

[0061] When the substrate is skewed, the offset of its four corner points in the width direction is related to the sine of the substrate length L and the included angle Q. For a substrate with a width of 2500mm, even an included angle of only ±1° will result in an offset of approximately 43.6mm in the length direction (2500×sin1°≈43.6mm). This offset is much larger than the sensor's field of view and must be compensated for by increasing the number of sensors or adjusting their layout. This characteristic quantifies the impact of the included angle on the coverage area, making system design more data-driven and avoiding insufficient coverage or excessive redundancy caused by empirical estimation.

[0062] The above scheme calculates the included angle error parameter by combining the substrate length, width and preset positioning error angle. It can quantify the impact of substrate rotation on the scanning coverage area, reasonably expand the sensor layout width, and ensure that even if the substrate is slightly tilted, its four corners are still within the effective scanning area, avoiding missed detection of edge foreign objects.

[0063] In a more detailed embodiment, to achieve coating of a 2500mm wide substrate 200, 26 3D vision detectors 101 are typically installed side by side with a spacing of 41mm between them and a height of 185mm from the support platform. In this way, the surface of the 2500mm wide substrate can be scanned in one go.

[0064] Please see Figure 4 This application also provides a method for detecting foreign matter in perovskite coating based on the above-mentioned perovskite coating foreign matter detection system, including steps 402, 404 and 406.

[0065] Step 402: Obtain the three-dimensional point cloud data of the substrate to be coated.

[0066] Step 404: Determine the substrate plane based on the fitting of the three-dimensional point cloud data, and determine the geometric parameters of the target area that protrudes relative to the substrate plane.

[0067] Step 406: Determine the foreign object detection result based on the geometric parameters.

[0068] This embodiment provides a method for detecting foreign objects in perovskite coatings, based on the perovskite coating foreign object detection system shown in the above embodiment and accompanying drawings. First, three-dimensional point cloud data of the substrate 200 to be coated is acquired; then, the substrate plane is determined by fitting the three-dimensional point cloud data; next, a target area protruding relative to the substrate plane is identified; then, the geometric parameters of this target area are measured, including height, length, and width. Finally, the presence of foreign objects is determined based on the geometric parameters, and the detection result is output.

[0069] The above solution utilizes multiple 3D vision detectors 101 arranged side-by-side with partially overlapping fields of view to achieve comprehensive scanning of a wide-width coating substrate. This solves the problems of insufficient field of view of a single sensor and decreased accuracy due to light decay over long spans. Furthermore, by combining point cloud data with plane fitting and identifying protruding foreign objects, it can reliably detect tiny particles with a height greater than or equal to 50 micrometers, preventing interference with the coating die 110 and ensuring coating quality and equipment safety.

[0070] Please see Figure 5 In one embodiment, the geometric parameters include the protrusion height, and step 406 includes steps 502 and 504.

[0071] Step 502: If there is a target protrusion height greater than or equal to a preset height threshold, determine the location information of the target area corresponding to the target protrusion height.

[0072] Step 504: Output the foreign object detection result based on the location information and the height of the target protrusion.

[0073] The protrusion height refers to the vertical distance of the target area relative to the fitted substrate plane. The preset height threshold refers to the maximum allowable height of foreign objects set by the system. For example, it can be set based on the minimum gap (50µm) between the coating die 110 and the glass. Exceeding this value is considered the presence of dangerous foreign objects. The location information is the coordinates of the foreign object on the substrate surface. For example, it can include position x (along the width direction) and position y (along the coating direction). The foreign object detection results can include whether the limit is exceeded, the location of the foreign object, its height h, length and width dimensions, etc., and can generate a LOG report to the host computer to determine whether to stop coating and issue an alarm.

[0074] If a target protrusion with a height greater than or equal to a preset height threshold (e.g., 50µm) exists, the location information of the target area corresponding to the protrusion is determined, and then the foreign object detection result is output based on the location information and the protrusion height. For example, in one embodiment, if the foreign object size exceeds the preset limit, the device stops coating, triggers an alarm, and personnel intervene for inspection; if the foreign object size does not exceed the preset limit, only a log record is retained.

[0075] This method implements a tiered response mechanism. For hazardous foreign objects with a height ≥50µm, the system not only outputs the height value but also provides precise x and y coordinates, facilitating quick location and removal by operators. For tiny protrusions below a preset height threshold, only a log is recorded to avoid frequent downtime and reduced production efficiency. Furthermore, the output location information can guide subsequent automatic cleaning or marking equipment.

[0076] The above solution automatically outputs the position information and height value of the protrusion when the height of the protrusion is greater than or equal to the preset threshold, and triggers an alarm or logs it. This achieves a graded response mechanism of stopping the machine if the height exceeds the threshold and recording the log if the height does not exceed the threshold. This avoids frequent shutdowns due to small foreign objects, which affect efficiency, and effectively prevents dangerous foreign objects from damaging the coating die head 110, thus balancing production safety and continuity.

[0077] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0078] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A foreign matter detection system for perovskite coatings, characterized in that, include: A support platform is used to place the substrate to be coated and to move the substrate along the coating direction. A coating die head is fixedly mounted above the support platform; A visual recognition component includes multiple 3D vision detectors, which are fixedly arranged side-by-side above the support platform along the width direction of the substrate to be coated. The scanning field of view of two adjacent 3D vision detectors at least partially overlaps, and the scanning coverage width of the multiple 3D vision detectors is greater than or equal to the width of the substrate to be coated, with the width direction perpendicular to the coating direction. The 3D vision detectors are used to scan the substrate to be coated as it moves toward the coating die along the coating direction to determine three-dimensional point cloud data. The processing component is connected to each of the 3D vision detectors. The processing component is used to fit the substrate plane according to the three-dimensional point cloud data, determine the geometric parameters of the target area protruding relative to the substrate plane, and determine the foreign object detection result according to the geometric parameters.

2. The foreign matter detection system for perovskite coatings according to claim 1, characterized in that, The plurality of 3D vision detectors are arranged coaxially in a straight line along the width direction of the substrate to be coated, and the mounting height of each 3D vision detector is consistent.

3. The foreign matter detection system for perovskite coatings according to claim 1, characterized in that, Multiple 3D vision detectors are arranged at equal intervals.

4. The foreign matter detection system for perovskite coatings according to claim 1, characterized in that, The overlap width of the scanning field of view of two adjacent 3D vision detectors is greater than or equal to 1 mm.

5. The foreign matter detection system for perovskite coatings according to claim 1, characterized in that, The distance between the visual recognition component and the coating die head is greater than or equal to a preset spacing, wherein the preset spacing is greater than or equal to the product of the moving speed of the substrate to be coated and the processing time, and the processing time represents the total processing time from the start of a single acquisition to the completion of foreign object recognition by the processing component.

6. The foreign matter detection system for perovskite coatings according to any one of claims 1-5, characterized in that, The scanning coverage width is determined based on the total scanning field width of the 3D vision detector, the overlap width of the scanning field ranges of two adjacent 3D vision detectors, and the substrate positioning error compensation parameters.

7. The foreign matter detection system for perovskite coatings according to claim 6, characterized in that, The substrate positioning error compensation parameters include the included angle error parameter and the horizontal positioning error parameter.

8. The foreign matter detection system for perovskite coatings according to claim 7, characterized in that, The included angle error parameter is determined based on the length of the substrate to be coated, the width of the substrate to be coated, and the preset positioning error angle.

9. A method for detecting foreign matter in perovskite coatings based on the perovskite coating foreign matter detection system according to any one of claims 1-8, characterized in that, include: Obtain the three-dimensional point cloud data of the substrate to be coated; The substrate plane is determined by fitting the three-dimensional point cloud data, and the geometric parameters of the target region that protrudes relative to the substrate plane are determined. The foreign object detection result is determined based on the geometric parameters.

10. The method for detecting foreign matter in perovskite coatings according to claim 9, characterized in that, The geometric parameters include the protrusion height, and determining the foreign object detection result based on the geometric parameters includes: If there is a target protrusion height greater than or equal to a preset height threshold, determine the location information of the target area corresponding to the target protrusion height; Based on the location information and the height of the target protrusion, output the foreign object detection result.

Images