Perovskite laser scribing accurate regulation and control method and system based on multiple modes

Abstract

The embodiment of the invention provides a perovskite laser scribing accurate regulation and control method and system based on multiple modes. The method comprises the steps that a multispectral image of a perovskite substrate is acquired; wherein the multispectral image comprises an ultraviolet image, a visible light image and a near-infrared image; based on the ultraviolet image, the visible light image and the near-infrared image, performing fusion processing to obtain a three-dimensional fusion image; generating a laser time regulation and control diagram based on a preset light intensity query table and the three-dimensional fusion image; wherein each pixel in the laser time regulation and control diagram represents the time required for scribing the pixel depth under the fixed laser power and the fixed laser focusing degree; and based on the laser time regulation and control diagram and height distance measurement, power adjustment is carried out, and scribing is completed. According to the scheme, the consistency of the lineation depth and width can be improved, and then the lineation precision is improved.

Description

Technical Field

[0001] This application relates to the field of laser scribing technology, specifically to a method and system for precise control of perovskite laser scribing based on multimodal modes. Background Technology

[0002] Laser scribing technology is a key process in the monolithic integration fabrication of perovskite photovoltaic modules. The scribing quality, especially the accuracy of the scribing depth and width, directly determines the series efficiency and yield of the cells. However, in the actual processing of large-format perovskite substrates, the accuracy of laser scribing faces fundamental challenges introduced by the material's inherent characteristics, making it difficult for existing technologies to effectively control scribing errors.

[0003] Current mainstream technologies primarily achieve dynamic line following through visual positioning and real-time ranging. While this method can compensate for the macroscopic position and warping errors of the substrate, it suffers from a lack of error perception dimensions, resulting in poor line marking accuracy. Summary of the Invention

[0004] This application aims to provide a method and system for precise control of perovskite laser scribing based on multimodal operation, which can improve the consistency of scribing depth and width, thereby improving scribing accuracy.

[0005] The technical solution of this application is implemented as follows: In a first aspect, embodiments of this application provide a method for precise control of perovskite laser scribing based on multimodal modes, the method comprising: Acquire multispectral images of a perovskite substrate; wherein the multispectral images include ultraviolet images, visible light images, and near-infrared images; Based on the ultraviolet image, the visible light image, and the near-infrared image, a fusion process is performed to obtain a three-dimensional fused image; Based on a preset light intensity lookup table and the three-dimensional fused image, a laser time control map is generated; wherein, each pixel in the laser time control map represents the time required for the depth of the line pixel under fixed laser power and fixed laser focus. Based on the laser time control diagram and altitude measurement, the power is adjusted to complete the line marking.

[0006] In the above scheme, the step of fusing the ultraviolet image, the visible light image, and the near-infrared image to obtain a three-dimensional fused image includes: The ultraviolet image is subjected to median filtering to obtain an ultraviolet filtered image; The near-infrared image is subjected to Gaussian filtering to obtain a near-infrared filtered image; The ultraviolet filtered image, the near-infrared filtered image, and the visible light image are fused together to obtain the three-dimensional fused image; wherein, the three-dimensional fused image characterizes the perovskite's filling degree and material composition.

[0007] In the above scheme, the three-dimensional fused image includes a first feature value characterizing the perovskite's filling degree and a second feature value characterizing the material.

[0008] In the above scheme, generating a laser time-controlled map based on a preset light intensity lookup table and the three-dimensional fused image includes: Obtain the laser power and laser focus of laser scribing at different stages of the laser scribing process; The laser time control map is generated based on the light intensity lookup table, the laser power of the line drawing, the laser focus of the line drawing, and the three-dimensional fused image.

[0009] In the above scheme, generating the laser timing control map based on the light intensity lookup table, the laser power used for marking lines, the laser focus degree used for marking lines, and the three-dimensional fused image includes: Based on the laser power and laser focus, the dwell time corresponding to each of the different stages of laser scribing is obtained by querying the light intensity lookup table. Based on the laser scribing surface and the dwell time, the three-dimensional fused image is vertically mapped to generate the laser time control map.

[0010] In the above scheme, the step of obtaining the dwell time corresponding to each stage of the laser scribing process by querying the light intensity lookup table based on the scribing laser power and the scribing laser focus includes: For each laser scribing process stage, the material and fill degree corresponding to the current laser scribing position are obtained by looking up the light intensity lookup table based on the scribing laser power and the scribing laser focus. This allows for the determination of the dwell time for each laser scribing process stage.

[0011] In the above scheme, before generating the laser time-control map based on the preset light intensity lookup table and the three-dimensional fused image, the method further includes: A five-dimensional system is defined; wherein, the five-dimensional system includes laser power, laser focus, fill factor, material, and dwell time; Basic data is collected through orthogonal experiments; and the basic data is trained and fitted based on a machine learning regression model to generate a mapping relationship between dimensions. Based on the inter-dimensional mapping relationship, an initial five-dimensional data set is determined; and the initial five-dimensional data set is transformed into discretized mapping data to obtain a five-dimensional array. Based on the five-dimensional array, the light intensity lookup table is determined.

[0012] Secondly, embodiments of this application provide a multimodal perovskite laser scribing precision control system, which includes: an acquisition module, a fusion module, a generation module, and an adjustment module, wherein... The acquisition module is used to acquire multispectral images of the perovskite substrate; wherein the multispectral images include ultraviolet images, visible light images, and near-infrared images; The fusion module is used to perform fusion processing based on the ultraviolet image, the visible light image, and the near-infrared image to obtain a three-dimensional fused image; The generation module is used to generate a laser time control map based on a preset light intensity lookup table and the three-dimensional fused image; wherein, each pixel in the laser time control map represents the time required for the depth of the line pixel under fixed laser power and fixed laser focus. The adjustment module is used to adjust the power based on the laser time control diagram and altitude measurement to complete the line drawing.

[0013] Thirdly, embodiments of this application provide a multimodal perovskite laser scribing precision control device, comprising: a processor and a memory; wherein, The memory is used to store computer programs; The processor is configured to call and run the computer program from the memory to perform the method as described in the first aspect.

[0014] Fourthly, embodiments of this application provide a computer-readable storage medium storing executable instructions for causing a processor to perform the method described in the first aspect.

[0015] This application provides a method and system for precise control of perovskite laser scribing based on multimodal imaging. The method includes: acquiring a multispectral image of a perovskite substrate; wherein the multispectral image includes an ultraviolet image, a visible light image, and a near-infrared image; performing fusion processing on the ultraviolet image, the visible light image, and the near-infrared image to obtain a three-dimensional fused image; generating a laser time control map based on a preset light intensity lookup table and the three-dimensional fused image; wherein each pixel in the laser time control map represents the time required for scribing pixel depth under fixed laser power and fixed laser focus; and adjusting the power based on the laser time control map and height measurement to complete the scribing. In the above scheme, ultraviolet, visible, and near-infrared multispectral data are collected simultaneously to comprehensively capture information on substrate physical defects, film boundaries, and material uniformity. Then, through dynamic filtering optimization and dual 8-bit feature quantization fusion, the physical fill degree and material laser intensity requirements are transformed into a structured three-dimensional fused image, establishing a basis for error correlation. Based on a five-dimensional lookup table of "laser power-focus degree-fill degree-material-time," a pixel-level laser time control map is generated to achieve precise matching of multi-dimensional parameters. Finally, through real-time linkage between laser head movement and the control map, combined with closed-loop feedback correction from coaxial detection, the scribing time and parameters are dynamically adjusted to solve errors such as excessive depth / shallowness and excessive width / narrowness, ensuring the scribing accuracy and consistency of large-format substrates. This improves the consistency of scribing depth and width, thereby enhancing scribing accuracy. Attached Figure Description

[0016] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the specification, serve to explain the technical solutions of this application. Obviously, the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0017] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0018] Figure 1 A flowchart illustrating a method for precise control of perovskite laser scribing based on multimodal operation, provided in an embodiment of this application; Figure 2 A schematic diagram of laser timing adjustment for the production of a multimodal perovskite laser scribing precision control method provided in this application embodiment; Figure 3A schematic diagram of a multimodal perovskite laser scribing precision control system provided in this application embodiment; Figure 4 This is a schematic diagram of a multimodal perovskite laser scribing precision control device provided in an embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the specific technical solutions of this application will be further described in detail below with reference to the accompanying drawings of the embodiments of this application. The following embodiments are used to illustrate this application, but are not intended to limit the scope of this application.

[0020] Unless otherwise defined, all technical and scientific terms used in this application have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. The terminology used in this application is for the purpose of describing embodiments of this application only and is not intended to be limiting of this application.

[0021] In the following description, references to "some embodiments," "this embodiment," "this application embodiment," and examples, etc., describe a subset of all possible embodiments. However, it is understood that "some embodiments" may be the same subset or different subset of all possible embodiments and may be combined with each other without conflict.

[0022] If the application documents contain similar descriptions such as "first / second", the following explanation shall be added: In the following description, the terms "first / second / third" are used only to distinguish similar objects and do not represent a specific order of objects. It is understood that "first / second / third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0023] Current traditional laser marking technology mainly achieves dynamic following of lines through visual positioning and real-time ranging, but it has the following problems: First, there is a lack of error perception: existing systems can only perceive the geometric information of the substrate (position, height), while lacking perception of another decisive factor in scribing accuracy—the material properties of the perovskite thin film. Inevitably, during the film preparation process, there are compositional inhomogeneities, thickness fluctuations, differences in crystallinity, and microscopic defects, which can lead to localized changes in the absorption efficiency of various laser energies. Current technology cannot detect these changes, and therefore cannot predict the resulting random errors in scribing depth and morphology.

[0024] Second, the control logic is disconnected from the error sources: Existing control strategies are based on "fixed laser parameters" or "focusing only based on height," and their underlying logic assumes that the material's response to energy is uniform. This contradicts the inherent inhomogeneity of perovskite materials. When the laser scans regions with different characteristics, fixed energy parameters inevitably lead to systematic errors such as excessively deep (damaging the underlying layer), excessively shallow (ineffective isolation), or inconsistent widths of the lines, severely affecting the battery's electrical isolation and connection performance.

[0025] Third, the unpredictability and retrospective nature of errors: Since the material state cannot be perceived before processing, the existing technology is essentially a "trial and error" process. All scribing errors caused by material inhomogeneity can only be discovered after scribing is completed or even in the subsequent electrical testing. It is impossible to predict in advance and prevent in the process, resulting in extremely high waste of materials and production costs.

[0026] Based on this, embodiments of this application provide a method for precise control of perovskite laser scribing based on multimodal modes. Figure 1 A flowchart illustrating a method for precise control of perovskite laser scribing based on multimodal modes, provided in this application embodiment, will be combined with... Figure 1 The steps shown are explained.

[0027] S101. Acquire a multispectral image of the perovskite substrate; wherein the multispectral image includes an ultraviolet image, a visible light image, and a near-infrared image.

[0028] In some embodiments of this application, a multimodal perovskite laser scribing precision control method is adapted to perovskite solar cell scenarios.

[0029] In some embodiments of this application, a multimodal perovskite laser scribing precision control method is adapted to a multimodal perovskite laser scribing precision control system.

[0030] In some embodiments of this application, multispectral images of a perovskite substrate are obtained, wherein the multispectral images include ultraviolet images, visible light images, and near-infrared images.

[0031] For example, ultraviolet images can be acquired using a 405nm ultraviolet camera to obtain physical defects inside the perovskite substrate, such as pinholes and cracks; visible light images can be acquired using a 532nm visible light camera to obtain the reflectivity differences and physical deformation of different functional films, providing a benchmark for subsequent scribing positions and multi-light fusion calibration; near-infrared images can be acquired using a 900-1700nm near-infrared camera to analyze the uniformity of the perovskite substrate material composition.

[0032] S102. Based on the ultraviolet image, visible light image and near-infrared image, perform fusion processing to obtain a three-dimensional fused image.

[0033] In some embodiments of this application, a median filter is applied to an ultraviolet image to obtain an ultraviolet filtered image; a Gaussian filter is applied to a near-infrared image to obtain a near-infrared filtered image; and a fusion process is performed based on the ultraviolet filtered image, the near-infrared filtered image, and the visible light image to obtain a three-dimensional fused image; wherein, the three-dimensional fused image characterizes the filling degree and material of the perovskite.

[0034] In some embodiments of this application, the three-dimensional fused image includes a first feature value characterizing the filling degree of perovskite and a second feature value characterizing the material; wherein, the first feature value of the three-dimensional fused image is the high 8 bits of each pixel, representing physical characteristics, ranging from [0, 255], the larger the pixel value at each pixel position, the greater the filling degree of perovskite, and vice versa, the greater the void; the second feature value of the three-dimensional fused image is the low 8 bits of each pixel, representing the material, ranging from [0, 255], from small to large, reflecting the laser intensity required for laser scribing, wherein 128 is the standard laser scribing intensity for perovskite.

[0035] S103. Based on the preset light intensity lookup table and the three-dimensional fusion image, generate a laser time control map; wherein, each pixel in the laser time control map represents the time required for the depth of the line pixel under fixed laser power and fixed laser focus.

[0036] In some embodiments of this application, the preset light intensity lookup table is a five-dimensional table, with dimensions being laser power, laser focus, material, perovskite fill degree (hereinafter referred to as fill degree), and dwell time. Specifically, the lookup method involves determining the required scribing time for pixel-level scribing depth based on a fixed laser power and a fixed laser focus, for a specific material (including pure perovskite) and perovskite fill degree.

[0037] In some embodiments of this application, the laser power and laser focus corresponding to different stages of laser scribing are obtained; a laser time control map is generated based on the light intensity lookup table, the laser power, the laser focus, and the three-dimensional fused image.

[0038] S104. Based on the laser time control map and altitude measurement, adjust the power to complete the line marking.

[0039] In some embodiments of this application, dual power adjustments are performed based on the height ranging and laser time control diagram during the laser marking process to complete the marking.

[0040] For example, for each stage of laser scribing P1-P3, when the laser head moves to a certain coordinate (i,j) on the substrate, the laser time control diagram of the corresponding stage is queried in real time. With the help of height measurement, the power is adjusted through a dual protection mechanism to complete the scribing and ensure that the scribing is not too deep or too shallow, too wide or too narrow.

[0041] Understandably, simultaneous acquisition of ultraviolet, visible, and near-infrared multispectral data comprehensively captures information on substrate physical defects, film boundaries, and material uniformity. Dynamic filtering optimization and dual 8-bit feature quantization fusion then transform the physical fill degree and material laser intensity requirements into a structured three-dimensional fused image, establishing a foundation for error correlation. Based on a five-dimensional lookup table of "laser power-focusing degree-filling degree-material-time," a pixel-level laser time control map is generated, achieving precise matching of multi-dimensional parameters. Finally, through real-time linkage between laser head movement and the control map, combined with closed-loop feedback correction from coaxial detection, the scribing time and parameters are dynamically adjusted to resolve errors such as excessive depth / shallowness and excessive width / narrowness, ensuring the scribing accuracy and consistency of large-format substrates. This improves the consistency of scribing depth and width, thereby enhancing scribing accuracy.

[0042] In some embodiments of this application, S102 can be implemented by S201-S203, as follows: S201. Perform median filtering on the ultraviolet image to obtain an ultraviolet filtered image.

[0043] S202. Perform Gaussian filtering on the near-infrared image to obtain a near-infrared filtered image.

[0044] S203. Based on the ultraviolet filtered image, near-infrared filtered image and visible light image, a fusion processing is performed to obtain a three-dimensional fused image; wherein, the three-dimensional fused image characterizes the filling degree and material of the perovskite.

[0045] For example, a median filter is applied to an ultraviolet image to obtain an ultraviolet filtered image, the purpose of which is to remove pseudo-defects in physical defects; a Gaussian filter is applied to a near-infrared image to obtain a near-infrared filtered image, the purpose of which is to smooth the uniformity of chemical composition; and the ultraviolet filtered image, the near-infrared filtered image and the visible light image are fused to obtain a three-dimensional fused image.

[0046] In some embodiments of this application, S103 can be implemented by S301 and S302, as follows: S301. Obtain the laser power and laser focus of laser scribing at different stages of the laser scribing process.

[0047] S302. Based on the light intensity lookup table, the laser power of the scribing line, the laser focus of the scribing line, and the three-dimensional fusion image, generate a laser time control map.

[0048] In some embodiments of this application, based on the laser power and laser focus of the laser scribing, the dwell time corresponding to each stage of the laser scribing process is obtained by querying a light intensity lookup table; based on the laser scribing surface and the dwell time, the three-dimensional fused image is vertically mapped to generate a laser time control map.

[0049] In some embodiments of this application, for the scribing positions corresponding to different stages of laser scribing, the material and fill degree corresponding to the current laser scribing position are looked up through a light intensity lookup table based on the scribing laser power and scribing laser focus, so as to obtain the dwell time corresponding to each stage of laser scribing.

[0050] For example, the laser power and laser focus data for each stage of laser scribing (P1-P2-P3 process) are obtained from P1 to P3. Based on a preset light intensity lookup table, the laser power and laser focus data for each stage of P1-P3, and the 3D fusion image, a laser time control map for each stage of P1-P3 is generated. For any stage in P1-P3, the corresponding laser time control map is generated by: vertically mapping the 3D fusion image onto the laser scribing surface (e.g., ...). Figure 2 As shown in the figure, a two-dimensional laser time control map (i.e., laser time control map) is obtained by querying and calculating the light intensity lookup table. Each pixel in the laser time control map represents the time required to reach the pixel depth of the line under fixed laser power and fixed laser focus. The specific calculation formula is as follows: Among them, T i,j This represents the dwell time at position (i,j). str and fov Indicates laser power and laser focus. fill and texture The values ​​represent fill intensity and material properties, i, j, and n represent the coordinates of the 3D fusion map, and Z represents the line depth. Here, map is a five-dimensional array, with the following square brackets [] representing the data for four of its dimensions, and the final result retrieved is the fifth dimension.

[0051] In some embodiments of this application, before S103, the method further includes: A five-dimensional system was established, which includes laser power, laser focus, fill factor, material properties, and dwell time. Basic data were collected through orthogonal experiments; and the basic data were trained and fitted based on a machine learning regression model to generate mapping relationships between dimensions. Based on the mapping relationship between dimensions, an initial five-dimensional data set is determined; and the initial five-dimensional data set is transformed into discretized mapped data to obtain a five-dimensional array. A lookup table for light intensity is determined based on a five-dimensional array.

[0052] For example, laser power (unit: W) P represents the intensity of the laser output energy, and the experimental range can be set to P∈[Pmin,Pmax], for example, 1W to 10W, based on the absorption characteristics of the perovskite film; laser focus (unit: μm) D represents the laser spot diameter, which affects the energy density, and the experimental range can be set to D∈[Dmin,Dmax], for example, 10μm to 50μm; fill factor (dimensionless) F represents the physical coverage of the perovskite film in a unit pixel area, quantized by the high 8-bit feature value of the 3D fusion image, ranging from 0 to 255; material (dimensionless) M represents the laser absorption coefficient corresponding to the chemical composition and crystallization state of the perovskite film, quantized by the low 8-bit feature value of the 3D fusion image, ranging from 0 to 255; dwell time (unit: ms) T represents the duration of laser action on a single pixel, directly affecting the scribing depth.

[0053] A five-dimensional light intensity lookup table can be constructed through the following steps: design a five-dimensional system (laser power, laser focus, fill power, material, and residence time); collect basic data through orthogonal experiments; train and fit the experimental data using a machine learning regression model to generate inter-dimensional mapping relationships to supplement data not covered in the experiments; store the discretized mapping data as a five-dimensional array to achieve fast indexing and retrieval. The specific implementation process is as follows: An orthogonal experimental design was employed, with L(n,m,k) as the experimental group, where n is the number of experiments, m is the number of levels, and k is the number of factors (5 factors in this case). For example, L(32,4,45) was chosen, and four levels were set for laser power, laser focus, fill power, and material properties, respectively, to conduct combined experiments. The experimental samples were perovskite substrates with different fill powers and material characteristics. Actual laser scribing was performed, and the scribing depth and width were measured. The optimal dwell time for each set of parameters was recorded, forming the basic dataset {P,D,F,M,T}.

[0054] The basic data is trained using a machine learning regression model (such as Support Vector Regression (SVR), Random Forest Regression, or a neural network). The input features are laser power, laser focus, fill factor, and material, and the output is the dwell time.

[0055] Five-fold cross-validation was used during model training to optimize hyperparameters and ensure the model's generalization ability. After training, the mapping relationship from {P,D,F,M} to T is obtained, i.e., T = func(P,D,F,M).

[0056] Based on the above mapping relationship, uniform sampling is performed within the experimental parameter range to generate a dense five-dimensional data set. For example, each dimension is discretized by step size (e.g., laser power step size 0.1W, focus step size 1μm, fill size and material step size 1), and the corresponding dwell time is calculated to form an initial five-dimensional data table. To improve query efficiency, continuous data is discretized into a multi-dimensional array structure, i.e., a five-dimensional lookup table.

[0057] It is understood that this application has the following beneficial effects: First, it solves the problem of multimodal data discontinuity and improves information integrity and correlation: By synchronously acquiring data from multiple band cameras and using hardware triggering control, it achieves full-dimensional information coverage of physical defects, film boundaries, and material uniformity of perovskite substrates, avoiding the limitations of single data acquisition; at the same time, it ensures accurate alignment of spatial / temporal coordinates of multi-source data, establishing a reliable data foundation for subsequent fusion and regulation, and eliminating the information misalignment risks of traditional multi-step acquisition.

[0058] Second, optimize image processing accuracy, retain effective features and achieve structured fusion: adopt a dynamically adaptable filtering algorithm (such as adjusting filtering parameters according to defect size and composition distribution) to effectively eliminate false defects and retain true features, avoiding feature distortion of traditional fixed filtering; through dual 8-bit pixel quantization mapping, the physical fill degree and material laser intensity requirements are structuredly fused, so that the fused image can directly support parameter adjustment, solving the problem of information invalidation of traditional pixel superposition.

[0059] Third, achieve pixel-level precise parameter control to adapt to regional differences: construct a five-dimensional lookup table of "power-focus-fill-material-time", combine the pixel-level features of the three-dimensional fused image to achieve differentiated scribing time control in different regions; avoid the limitations of traditional fixed parameters or single-dimensional lookup tables, solve the problem of over-scribing and under-scribing caused by uneven fill and material on large-format substrates, and improve the consistency of scribing depth and width.

[0060] Based on the above embodiments, a method for precise control of perovskite laser scribing based on multimodal modes is provided. This application also provides a system for precise control of perovskite laser scribing based on multimodal modes, such as... Figure 3 As shown, Figure 3 This is a schematic diagram of a multimodal perovskite laser scribing precision control system provided in an embodiment of this application. The multimodal perovskite laser scribing precision control system 3 includes: an acquisition module 301, a fusion module 302, a generation module 303, and an adjustment module 304, wherein... The acquisition module 301 is used to acquire multispectral images of the perovskite substrate; wherein, the multispectral images include ultraviolet images, visible light images, and near-infrared images; The fusion module 302 is used to perform fusion processing based on the ultraviolet image, the visible light image, and the near-infrared image to obtain a three-dimensional fused image; The generation module 303 is used to generate a laser time control map based on a preset light intensity lookup table and the three-dimensional fusion image; wherein, each pixel in the laser time control map represents the time required for the depth of the line pixel under fixed laser power and fixed laser focus. The adjustment module 304 is used to adjust the power based on the laser time control diagram and the height distance measurement to complete the line drawing.

[0061] In some embodiments of this application, the fusion module 302 is further configured to perform median filtering on the ultraviolet image to obtain an ultraviolet filtered image; perform Gaussian filtering on the near-infrared image to obtain a near-infrared filtered image; and perform fusion processing on the ultraviolet filtered image, the near-infrared filtered image, and the visible light image to obtain the three-dimensional fused image; wherein the three-dimensional fused image characterizes the filling degree and material of the perovskite.

[0062] In some embodiments of this application, the generation module 303 is further configured to obtain the laser power and laser focus corresponding to different stages of laser scribing; and generate the laser time control map based on the light intensity lookup table, the laser power, the laser focus and the three-dimensional fused image.

[0063] In some embodiments of this application, the generation module 303 is further configured to query the light intensity lookup table based on the laser power and the laser focus of the laser scribing process to obtain the dwell time corresponding to each stage of the laser scribing process; and to perform vertical mapping on the three-dimensional fused image based on the laser scribing surface and the dwell time to generate the laser time control map.

[0064] In some embodiments of this application, the generation module 303 is further configured to, for the scribing positions corresponding to different stages of laser scribing, query the material and fill degree corresponding to the current laser scribing position through the light intensity lookup table based on the scribing laser power and the scribing laser focus, and obtain the dwell time corresponding to each stage of laser scribing.

[0065] In some embodiments of this application, the acquisition module 301 is further configured to determine a five-dimensional system before generating a laser time-controlled map based on a preset light intensity lookup table and the three-dimensional fused image; wherein the five-dimensional system includes laser power, laser focus, fill factor, material, and dwell time; basic data is collected through orthogonal experiments; and the basic data is trained and fitted based on a machine learning regression model to generate inter-dimensional mapping relationships; an initial five-dimensional data set is determined based on the inter-dimensional mapping relationships; and the initial five-dimensional data set is converted into discretized mapping data to obtain a five-dimensional array; and the light intensity lookup table is determined based on the five-dimensional array.

[0066] Based on the above embodiments, a method for precise control of perovskite laser scribing based on multimodal modes is provided. This application also provides a device for precise control of perovskite laser scribing based on multimodal modes, such as... Figure 4 As shown, Figure 4 This is a schematic diagram of a multimodal perovskite laser scribing precision control device provided in an embodiment of this application. The multimodal perovskite laser scribing precision control device 4 includes a processor 401 and a memory 402. The memory 402 is used to store computer programs; the processor 401 is used to call and run the computer programs from the memory to execute a multimodal perovskite laser scribing precision control method as described in the above embodiment.

[0067] In the embodiments of this application, the processor 401 described above can be at least one of the following: Application-Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field-Programmable Gate Array (FPGA), Central Processing Unit (CPU), Controller, Microcontroller, and Microprocessor. It is understood that for different devices, the electronic device used to implement the above processor function can also be other types, and the embodiments of this application do not specifically limit it.

[0068] This application provides a computer-readable storage medium storing a computer program for implementing, when executed by a processor, a multimodal perovskite laser scribing precision control method as described in any of the above embodiments.

[0069] For example, the program instructions corresponding to the multimodal perovskite laser scribing precision control method in this embodiment can be stored on storage media such as optical discs, hard disks, and USB flash drives. When the program instructions corresponding to the multimodal perovskite laser scribing precision control method in the storage media are read or executed by an electronic device, the multimodal perovskite laser scribing precision control method described in any of the above embodiments can be realized.

[0070] Furthermore, in the embodiments of this application, the functional modules can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional module.

[0071] If the integrated unit is implemented as a software functional module and is not sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this embodiment, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the method of this embodiment. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0072] It should be understood that the phrases "one embodiment," "an embodiment," or "some embodiments" mentioned throughout the specification mean that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, "in one embodiment," "in one embodiment," or "in some embodiments" appearing throughout the specification do not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in the various embodiments of this application, the sequence numbers of the above-described processes do not imply a sequential order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application. The sequence numbers of the embodiments in this application are merely for descriptive purposes and do not represent the superiority or inferiority of the embodiments. The descriptions of the various embodiments above tend to emphasize the differences between the various embodiments; their similarities or commonalities can be referred to mutually, and for the sake of brevity, these will not be repeated here.

[0073] The modules described above as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules. They may be located in one place or distributed across multiple network units. Some or all of the modules may be selected to achieve the purpose of this embodiment according to actual needs.

[0074] In addition, each functional module in the various embodiments of this application can be integrated into one processing unit, or each module can be a separate unit, or two or more modules can be integrated into one unit; the integrated modules can be implemented in hardware or in the form of hardware plus software functional units.

[0075] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, read-only memory (ROM), magnetic disks, or optical disks.

[0076] The methods disclosed in the several method embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments.

[0077] The features disclosed in the several product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.

[0078] The features disclosed in the several method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method or device embodiments.

[0079] The above description is merely an implementation method of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the protection scope of this application.

Claims

1. A method for precise control of perovskite laser scribing based on multimodal operation, characterized in that, The method includes: Acquire multispectral images of a perovskite substrate; wherein the multispectral images include ultraviolet images, visible light images, and near-infrared images; Based on the ultraviolet image, the visible light image, and the near-infrared image, a fusion process is performed to obtain a three-dimensional fused image; Based on a preset light intensity lookup table and the three-dimensional fused image, a laser time control map is generated; wherein, each pixel in the laser time control map represents the time required for the depth of the line pixel under fixed laser power and fixed laser focus. Based on the laser time control diagram and altitude measurement, the power is adjusted to complete the line marking.

2. The method according to claim 1, characterized in that, The process of fusing the ultraviolet image, the visible light image, and the near-infrared image to obtain a three-dimensional fused image includes: The ultraviolet image is subjected to median filtering to obtain an ultraviolet filtered image; The near-infrared image is subjected to Gaussian filtering to obtain a near-infrared filtered image; The ultraviolet filtered image, the near-infrared filtered image, and the visible light image are fused together to obtain the three-dimensional fused image; wherein, the three-dimensional fused image characterizes the perovskite's filling degree and material composition.

3. The method according to claim 1, characterized in that, The 3D fused image contains a first feature value characterizing the perovskite's fillability and a second feature value characterizing the material.

4. The method according to claim 1, characterized in that, The generation of a laser time-controlled map based on a preset light intensity lookup table and the three-dimensional fused image includes: Obtain the laser power and laser focus of laser scribing at different stages of the laser scribing process; The laser time control map is generated based on the light intensity lookup table, the laser power of the line drawing, the laser focus of the line drawing, and the three-dimensional fused image.

5. The method according to claim 4, characterized in that, The process of generating the laser time-controlled map based on the light intensity lookup table, the laser power of the line drawing, the laser focus of the line drawing, and the three-dimensional fused image includes: Based on the laser power and laser focus, the dwell time corresponding to each of the different stages of laser scribing is obtained by querying the light intensity lookup table. Based on the laser scribing surface and the dwell time, the three-dimensional fused image is vertically mapped to generate the laser time control map.

6. The method according to claim 5, characterized in that, The process involves using the laser intensity lookup table, based on the laser power and laser focus, to obtain the dwell time for each stage of the laser scribing process, including: For each laser scribing process stage, the material and fill degree corresponding to the current laser scribing position are obtained by looking up the light intensity lookup table based on the scribing laser power and the scribing laser focus. This allows for the determination of the dwell time for each laser scribing process stage.

7. The method according to claim 1, characterized in that, Before generating the laser time-controlled map based on the preset light intensity lookup table and the three-dimensional fused image, the method further includes: A five-dimensional system is defined; wherein, the five-dimensional system includes laser power, laser focus, fill factor, material, and dwell time; Basic data is collected through orthogonal experiments; and the basic data is trained and fitted based on a machine learning regression model to generate a mapping relationship between dimensions. Based on the inter-dimensional mapping relationship, an initial five-dimensional data set is determined; and the initial five-dimensional data set is transformed into discretized mapping data to obtain a five-dimensional array. Based on the five-dimensional array, the light intensity lookup table is determined.

8. A multimodal perovskite laser scribing precision control system, characterized in that, The multimodal perovskite laser scribing precision control system includes: an acquisition module, a fusion module, a generation module, and an adjustment module, wherein... The acquisition module is used to acquire multispectral images of the perovskite substrate; wherein the multispectral images include ultraviolet images, visible light images, and near-infrared images; The fusion module is used to perform fusion processing based on the ultraviolet image, the visible light image, and the near-infrared image to obtain a three-dimensional fused image; The generation module is used to generate a laser time control map based on a preset light intensity lookup table and the three-dimensional fused image; wherein, each pixel in the laser time control map represents the time required for the depth of the line pixel under fixed laser power and fixed laser focus. The adjustment module is used to adjust the power based on the laser time control map and altitude measurement to complete the line drawing.

9. A multimodal perovskite laser scribing precision control device, characterized in that, include: Processor and memory, of which, The memory is used to store computer programs; The processor is configured to call and run the computer program from the memory to perform the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores executable instructions for causing a processor to execute, thereby implementing the method of any one of claims 1 to 7.

Images