Ink-jet printing control method and device of perovskite battery and computer equipment

Abstract

The invention discloses an ink-jet printing control method and device for a perovskite battery and computer equipment, and the method comprises the steps: obtaining the time sequence characteristics of a current ink droplet in a process of falling to a perovskite battery substrate, and the spreading morphological characteristics of the current ink droplet on the perovskite battery substrate, the environmental characteristics are obtained in the process that current ink droplets are ejected by an ink-jet printer to form a film on a perovskite cell substrate; performing defect identification according to the time sequence features, the spreading morphological features and the environment features to obtain a current defect type; matching in a preset parameter adjustment rule according to the current defect type, and determining a target parameter adjustment strategy; and the printing process parameters are adjusted based on the target parameter adjustment strategy, and subsequent ink-jet printing is carried out based on the adjusted printing process parameters. Therefore, the problem of film forming defects caused by environmental fluctuation is solved, the uniformity of the perovskite cell film is improved, and the defect rate is effectively reduced.

Description

Technical Field

[0001] This invention relates to the field of battery manufacturing technology, and in particular to an inkjet printing control method, apparatus, and computer equipment for perovskite batteries. Background Technology

[0002] Inkjet printing technology has shown great potential in the preparation of thin films in the field of perovskite solar cells due to its advantages such as high precision, non-contact operation and high material utilization. Precise control in the inkjet printing process plays an extremely important role in ensuring the performance of perovskite solar cells.

[0003] The relevant technologies employ various monitoring techniques, such as laser scattering and electrical detection, to achieve precise control of inkjet printing. However, these technologies suffer from poor film uniformity, and the defect rate of inkjet printing needs to be reduced. Summary of the Invention

[0004] This invention aims to at least partially solve one of the technical problems in related technologies. To this end, this invention proposes an inkjet printing control method, apparatus, and computer equipment for perovskite solar cells, addressing the problem of film formation defects caused by environmental fluctuations, improving the uniformity of perovskite solar cell films, and effectively reducing the defect rate.

[0005] To achieve the above objectives, a first aspect of the present invention proposes an inkjet printing control method for perovskite solar cells. The method includes: acquiring the temporal characteristics of a current ink droplet falling onto a perovskite solar cell substrate, the spreading morphology characteristics of the ink droplet on the current perovskite solar cell substrate, and the environmental characteristics from the time the current ink droplet is ejected by the inkjet printer to the time it forms a film on the perovskite solar cell substrate; identifying defects based on the temporal characteristics, the spreading morphology characteristics, and the environmental characteristics to obtain a current defect type; matching the current defect type with preset parameter adjustment rules to determine a target parameter adjustment strategy; wherein the preset parameter adjustment rules include a mapping relationship between defect types and parameter adjustment strategies; and adjusting printing process parameters based on the target parameter adjustment strategy to perform subsequent inkjet printing based on the adjusted printing process parameters.

[0006] According to one embodiment of the present invention, obtaining the temporal features, the spreading morphology features, and the environmental features includes: obtaining an image sequence of the current ink droplet falling onto the perovskite solar cell substrate, an image of the spreading morphology of the ink droplet on the current perovskite solar cell substrate, and an environmental parameter sequence from the time the current ink droplet is ejected by the inkjet printer to the time sequence of film formation on the perovskite solar cell substrate; extracting features from the image sequence and the spreading morphology image using an ink droplet feature extraction model to obtain the temporal features corresponding to the image sequence and the spreading morphology features corresponding to the spreading morphology image; and extracting features from the environmental parameter sequence using an environmental feature extraction model to obtain the environmental features.

[0007] According to one embodiment of the present invention, the step of identifying defects based on the temporal features, the spreading morphology features, and the environmental features to obtain the current defect type includes: inputting the temporal features, the spreading morphology features, and the environmental features into a pre-trained defect type classification model, performing defect classification, and outputting the current defect type.

[0008] According to one embodiment of the present invention, the image sequence includes a first image at a first time moment and a second image at a second time moment; during the acquisition of the image sequence, the inkjet printing control method further includes: calculating the volume difference of the current ink droplet based on the acquired first image and second image to obtain the absolute value of the volume difference of the current ink droplet; if the absolute value of the volume difference is greater than a preset volume difference threshold, adjusting the pulse voltage amplitude of the inkjet printer according to preset voltage amplitude increment data.

[0009] According to one embodiment of the present invention, the environmental parameter sequence includes a first environmental parameter at a third time and a second environmental parameter at a fourth time; in the process of acquiring the environmental parameter sequence, the inkjet printing control method further includes: calculating the environmental parameter difference based on the acquired first environmental parameter and second environmental parameter to obtain environmental parameter difference data; wherein, the environmental parameter difference data includes at least one of the absolute value of the temperature difference in the inkjet printing chamber, the absolute value of the humidity difference in the inkjet printing chamber, and the absolute value of the ink droplet evaporation intensity difference; if the environmental parameter difference data includes the absolute value of the temperature difference, and If the absolute value of the temperature difference is greater than a preset temperature difference threshold, the temperature of the inkjet printing chamber is adjusted according to the preset temperature data; if the environmental parameter difference data includes the absolute value of the humidity difference, and the absolute value of the humidity difference is greater than a preset humidity difference threshold, the humidity of the inkjet printing chamber is adjusted according to the preset humidity data, and / or, the inert gas flux is adjusted according to the first preset flux increment data; if the environmental parameter difference data includes the absolute value of the ink droplet evaporation intensity difference, and the absolute value of the ink droplet evaporation intensity difference is greater than a preset intensity difference threshold, the inert gas flux is adjusted according to the second preset flux increment data.

[0010] According to one embodiment of the present invention, the defect types include pinhole defects, crack defects, satellite droplet defects, edge warping defects, and uneven stacking defects; wherein, the parameter adjustment strategy corresponding to the pinhole defect type includes reducing the temperature of the perovskite solar cell substrate according to preset temperature increment data, and / or increasing the humidity of the inkjet printing chamber according to preset humidity increment data; the parameter adjustment strategy corresponding to the crack defect type includes increasing the flow rate gradient of the inert gas according to preset flow rate increment data, and / or adjusting the temperature of the central region or the edge region according to a preset temperature difference between the central region and the edge region of the perovskite solar cell substrate; the parameter adjustment strategy corresponding to the satellite droplet defect type includes... The inkjet printer's inkjet voltage is reduced according to preset voltage increment data, and / or the inkjet printer's pulse width is increased according to a first preset time increment data; the parameter adjustment strategy corresponding to the edge warping defect type includes reducing the ink ejection volume of the inkjet printer's edge nozzle channels according to preset ink volume increment data, and / or delaying the inkjet time of the edge nozzle channels according to a second preset time increment data; the parameter adjustment strategy corresponding to the uneven stacking defect type includes performing grayscale printing calibration according to preset droplet volume, and / or adjusting the inkjet printer's inkjet trigger time according to preset timing compensation data; wherein, the preset timing compensation data is determined based on the deviation between the actual landing position and the expected landing position of the current ink droplet.

[0011] According to one embodiment of the present invention, after obtaining the current defect type, the method further includes: determining the defect cause corresponding to the current defect type based on the correspondence between the defect type and the defect cause; and performing matching according to the current defect type in a preset parameter adjustment rule, including: performing matching according to the current defect type and the corresponding defect cause in the preset parameter adjustment rule; wherein the preset parameter adjustment rule includes a mapping relationship between the defect type and the corresponding defect cause and the parameter adjustment strategy.

[0012] According to one embodiment of the present invention, each current defect type has its own priority; if multiple current defect types are obtained, before matching according to the current defect types in the preset parameter adjustment rules to determine the target parameter adjustment strategy, the inkjet printing control method further includes: determining a target defect type according to the priority of each current defect type; the step of matching according to the current defect data in the preset parameter adjustment rules to determine the target parameter adjustment strategy includes: matching according to the target defect type in the preset parameter adjustment rules to determine the target parameter adjustment strategy.

[0013] To achieve the above objectives, a second aspect of the present invention provides an inkjet printing control device for perovskite solar cells. The device includes: a feature acquisition module, configured to acquire the temporal characteristics of a current ink droplet falling onto a perovskite solar cell substrate, the spreading morphology characteristics of the ink droplet on the current perovskite solar cell substrate, and environmental characteristics from the moment the current ink droplet is ejected by the inkjet printer to the moment it forms a film on the perovskite solar cell substrate; a defect identification module, configured to identify defects based on the temporal characteristics, the spreading morphology characteristics, and the environmental characteristics to obtain a current defect type; an adjustment strategy determination module, configured to match the current defect type with preset parameter adjustment rules to determine a target parameter adjustment strategy; wherein the preset parameter adjustment rules include a mapping relationship between defect types and parameter adjustment strategies; and a parameter adjustment module, configured to adjust printing process parameters based on the target parameter adjustment strategy to perform subsequent inkjet printing based on the adjusted printing process parameters.

[0014] To achieve the above objectives, a third aspect of the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the inkjet printing control method described in any of the preceding embodiments.

[0015] According to multiple embodiments provided by this invention, real-time dynamic perception of ink droplet defects is achieved through full-process real-time monitoring of the printing process. This overcomes the limitations of post-process detection in related technologies, shifting the defect identification window from after film formation to during the film formation process, thus allowing time for timely correction of defect causes. Furthermore, by monitoring multi-dimensional features during the printing process, it solves the problem of limited accuracy in defect identification caused by measuring only a single-dimensional parameter in related technologies, providing a clear target for subsequent parameter adjustments. Automated decision-making for defect repair is achieved based on preset parameter adjustment rules, thereby realizing intelligent closed-loop control of the inkjet printing process. This ensures the printing system is always in an optimal state adapting to dynamic changes, significantly improving the uniformity of perovskite films and effectively reducing the defect rate, providing key technical support for efficient and stable perovskite solar cell manufacturing.

[0016] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating an inkjet printing control method provided according to an embodiment of this specification.

[0018] Figure 2a This is a flowchart illustrating a feature acquisition method provided according to an embodiment of this specification.

[0019] Figure 2b This is a schematic diagram of a closed-loop control system for inkjet printing of a three-terminal perovskite solar cell based on environmental perception and droplet monitoring, provided according to an embodiment of this specification.

[0020] Figure 3 This is a structural block diagram of an inkjet printing control device provided according to an embodiment of this specification.

[0021] Figure 4 This is a structural block diagram of a computer device provided according to an embodiment of this specification. Detailed Implementation

[0022] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0023] Inkjet printing technology, with its advantages of high precision, non-contact operation, and high material utilization, has shown great potential in the preparation of thin films for perovskite solar cells. However, perovskite inks typically contain varying concentrations of precursors (such as PbI2 / DMSO) and complex additives, resulting in significantly different viscosity (approximately 10 cp) and surface tension (30-50 mN / m) compared to traditional inks. This can easily lead to printhead clogging, satellite droplets, or multiple droplet merging. Large droplet volume deviations (>±5%) directly result in uneven film thickness, forming pinholes or coffee ring effects, reducing cell efficiency. Furthermore, droplets may deviate from their trajectory during flight due to factors such as airflow and static electricity (with deviations reaching tens of micrometers). The high evaporation rate of perovskite inks further alters droplet morphology (such as shrinkage or deformation), leading to blurred pattern edges or interlayer misalignment, affecting the structural accuracy of the device.

[0024] The unique physical properties of perovskite inks (such as high viscosity, easy crystallization, and rapid solvent evaporation) and the dynamic complexity of the inkjet process make precise control of the droplet state a key challenge.

[0025] For inkjet printing of perovskite solar cell thin films, various monitoring techniques are employed to achieve precise control of inkjet printing. For example: (1) Offline detection technology: After inkjet printing is completed, the printed film is analyzed and detected to evaluate the printing effect; (2) Laser scattering technology: The principle of laser scattering is used to measure ink droplets during the inkjet printing process; (3) Acoustic sensing technology (piezoelectric sensor): By installing a piezoelectric sensor at the printhead, the vibration of the printhead is monitored to indirectly obtain relevant information; (4) Electrical detection technology (substrate electrode measurement of droplet capacitance): Electrodes are set on the substrate, and the droplet situation is monitored by measuring the droplet capacitance.

[0026] However, these monitoring methods have obvious limitations: (1) Offline detection can only achieve post-process analysis and cannot adjust the printing process parameters in real time during the printing process. Moreover, the detection cost is high and the efficiency is low. (2) Laser scattering technology can only measure a single dimension (such as volume) and cannot obtain full-field information on the shape or flight trajectory of ink droplets. (3) Piezoelectric sensors can only monitor nozzle vibration and have no ability to sense the landing point of ink droplets. They are also susceptible to environmental noise interference. For example, when multiple nozzles work in parallel, crosstalk is easily generated between the piezoelectric sensors of different nozzles. (4) Electrical detection technology measures the landing point capacitance through the substrate electrode. It is only applicable to conductive inks and cannot monitor the state of ink droplets during the flight process.

[0027] Furthermore, related technologies primarily rely on open-loop inkjet printing systems, lacking effective closed-loop control technology. That is, fixed process parameters (such as printhead voltage, jetting frequency, substrate temperature, and ambient temperature and humidity baselines) are preset based on experience or prior experiments. The printhead then continuously jets perovskite ink according to these parameters, maintaining consistency throughout the printing process. However, the spreading and drying of perovskite ink on the substrate are affected by the substrate's hydrophilicity / hydrophobicity and ambient temperature and humidity. The open-loop printing mechanism in these technologies cannot effectively feed back the monitored data to real-time adjustments of the process parameters. This results in poor film uniformity and a high defect rate, severely restricting the large-scale production and performance consistency of perovskite devices.

[0028] To address film formation defects caused by environmental fluctuations, improve the uniformity of perovskite solar cell films, and effectively reduce the defect rate, it is necessary to propose an inkjet printing control method, apparatus, and computer equipment for perovskite solar cells. The inkjet printing control method provided in this specification is applied to an inkjet printing control system (hereinafter referred to as the control system). The control system acquires multi-dimensional features in real time during the inkjet printing process, including: the time-series characteristics of ink droplets ejected from the inkjet printer head and falling onto the perovskite solar cell substrate (e.g., flight speed, flight trajectory, morphological changes, etc.); the spreading morphological characteristics of ink droplets on the perovskite solar cell substrate (e.g., edge irregularity, number of satellite droplets, thickness uniformity, etc.); and real-time environmental characteristics during the printing process.

[0029] The control system identifies defects based on real-time acquired multi-dimensional features using preset algorithms or specific models. This determines whether ink droplets exhibit defects such as satellite droplets, pinholes, cracks, or uneven stacking. If defects are identified, current defect data is generated. This data may include information such as the defect type, defect severity, and cause.

[0030] The control system matches the identified current defect type with preset parameter adjustment rules to determine the target parameter adjustment strategy for the current defect situation. This rule contains the mapping relationship between defect types and corresponding effective parameter adjustment strategies. Based on the target parameter adjustment strategy, the control system automatically corrects printing process parameters (such as inkjet voltage, printing chamber temperature, jetting frequency, and substrate temperature) to ensure that droplet morphology and film quality meet expectations during subsequent printing processes, thus achieving closed-loop adaptive control of perovskite solar cell inkjet printing.

[0031] Therefore, by monitoring the entire printing process in real time, real-time dynamic perception of ink droplet defects is achieved, overcoming the limitations of post-process detection in related technologies. The defect identification window is moved from after film formation to during the film formation process, allowing time for timely correction of defect causes. Furthermore, by monitoring multi-dimensional features during the printing process, the problem of limited accuracy in defect identification due to measuring only a single-dimensional parameter in related technologies is solved, providing clear targets for subsequent parameter adjustments. Based on preset parameter adjustment rules, the decision-making for defect repair is automated, thereby achieving intelligent closed-loop control of the inkjet printing process. This ensures the printing system is always in an optimal state adapting to dynamic changes, significantly improving the uniformity of perovskite films and effectively reducing the defect rate, providing key technological support for efficient and stable perovskite solar cell manufacturing.

[0032] This specification provides an inkjet printing control method for perovskite solar cells, referring to... Figure 1 As shown, the inkjet printing control method may include the following steps.

[0033] S110. Obtain the temporal characteristics of the current ink droplet falling onto the perovskite solar cell substrate, the spreading morphology characteristics of the ink droplet on the current perovskite solar cell substrate, and the environmental characteristics from the time the current ink droplet is ejected by the inkjet printer to the time it forms a film on the perovskite solar cell substrate.

[0034] S120. Based on temporal characteristics, spreading morphology characteristics, and environmental characteristics, defect identification is performed to obtain the current defect type.

[0035] S130. Match the current defect type with the preset parameter adjustment rules to determine the target parameter adjustment strategy; wherein, the preset parameter adjustment rules include the mapping relationship between defect type and parameter adjustment strategy.

[0036] S140. Adjust the printing process parameters based on the target parameter adjustment strategy, so as to perform subsequent inkjet printing based on the adjusted printing process parameters.

[0037] The printing process parameters include the printing parameters of the inkjet printer and the environmental parameters.

[0038] Specifically, the control system acquires multi-dimensional feature data in real time during the printing process, including: the temporal characteristics of the current ink droplet falling onto the perovskite solar cell substrate, such as flight speed, flight trajectory, and morphological changes, which can help understand the dynamic situation of the ink droplet during its movement; the spreading morphological characteristics of the ink droplet on the current perovskite solar cell substrate, such as edge irregularity, number of satellite droplets, and thickness uniformity, which can reflect the deposition and diffusion of ink droplets on the substrate surface; and the environmental characteristics from the time the current ink droplet is ejected by the inkjet printer to the time it forms a film on the perovskite solar cell substrate, such as the temperature and humidity of the printing chamber, substrate temperature, and inert gas flow rate, which can reflect the evaporation rate and spreading speed of the ink droplet.

[0039] The control system utilizes acquired temporal characteristics, droplet spreading characteristics, and environmental characteristics to perform comprehensive analysis through specific algorithms or models to determine whether a defect exists and, if so, the current defect type. The control system then matches the identified defect type with preset parameter adjustment rules to determine a target parameter adjustment strategy for that defect type. These preset parameter adjustment rules establish a mapping relationship between various possible defect types and corresponding parameter adjustment strategies. For example, if insufficient droplet spreading is detected, according to the preset rules, it may be necessary to adjust parameters such as the inkjet printer's ejection pressure and ink volume.

[0040] The control system adjusts the current printing process parameters based on a defined target parameter adjustment strategy. The adjusted printing process parameters optimize subsequent processes such as ink droplet ejection, flight, and spreading, thereby improving the quality of subsequent inkjet printing, reducing the subsequent defect rate, and improving the quality of film formation on perovskite solar cell substrates.

[0041] For example, the temporal characteristics can be obtained by image recognition and feature analysis of the flight time sequence images of ink droplets falling onto the perovskite solar cell substrate; the spreading morphology characteristics can be obtained by image recognition and feature analysis of the images of the thin film printed on the perovskite solar cell substrate; and the environmental characteristics can be obtained by feature analysis of the temperature and humidity of the printing chamber, the substrate temperature, and the volatility signal collected by the laser interferometer gas phase sensor.

[0042] For example, preset parameter adjustment rules can be implemented as follows: If the defect type is uneven ink droplet spreading on the substrate with severe coffee ring effect, the corresponding parameter adjustment strategy can be to adjust the substrate temperature or adjust the inkjet volume of the nozzles at the edge of the printhead; if the defect type is satellite droplets (i.e., satellite droplet defect type), the corresponding parameter adjustment strategy can be to reduce the inkjet printer's inkjet voltage according to preset voltage increment data, and / or, increase the inkjet printer's pulse width according to a first preset time increment data, and / or, reduce the piezoelectric ceramic amplitude; if the defect type is a crack defect, the corresponding parameter adjustment strategy can be to increase the inert gas flow rate gradient according to preset flow rate increment data, and / or, adjust the temperature of the central or edge region according to the preset temperature difference between the central and edge regions of the perovskite solar cell substrate; if the defect type is ink droplet positioning deviation, the corresponding parameter adjustment strategy can be... If the defect type is abnormal droplet volume, the corresponding parameter adjustment strategy can be to adjust the ink ejection volume or the ink ejection frequency of the printhead; if the defect type is pinhole defect, the corresponding parameter adjustment strategy can be to reduce the temperature of the perovskite solar cell substrate according to preset temperature increment data, and / or to increase the humidity of the inkjet printing chamber according to preset humidity increment data; if the defect type is edge warping defect, the corresponding parameter adjustment strategy can be to reduce the ink ejection volume of the edge nozzle channel of the inkjet printer according to preset ink volume increment data, and / or to delay the ink ejection time of the edge nozzle channel according to a second preset time increment data; if the defect type is uneven stacking defect, the corresponding parameter adjustment strategy can be to perform grayscale printing calibration according to preset droplet volume, and / or to adjust the inkjet trigger time of the inkjet printer according to preset timing compensation data, etc.

[0043] It should be noted that this manual does not specify the methods for obtaining temporal characteristics, spreading morphology characteristics, and environmental characteristics, nor does it specify the rules for adjusting preset parameters.

[0044] The method for determining the target parameter adjustment strategy by matching the current defect type within the preset parameter adjustment rules can be any of the following: physical model-based method, deep learning-based method, rule-based expert system, etc. This specification does not specify any particular method.

[0045] In the above embodiments, real-time monitoring of the entire printing process enables real-time dynamic perception of ink droplet defects, overcoming the limitations of post-process detection in related technologies. This shifts the defect identification window from after film formation to during the film formation process, allowing time for timely correction of defect causes. Furthermore, by monitoring multi-dimensional features during the printing process, it addresses the problem of limited accuracy in defect identification caused by measuring only a single-dimensional parameter in related technologies, providing clear targets for subsequent parameter adjustments. Automated decision-making for defect repair based on preset parameter adjustment rules achieves intelligent closed-loop control of the inkjet printing process. This ensures the printing system is always in an optimal state adapting to dynamic changes, significantly improving the uniformity of perovskite films and effectively reducing the defect rate, providing key technological support for efficient and stable perovskite solar cell manufacturing.

[0046] In addition, the inkjet printing control method provided in this manual is applicable to perovskite inks with different formulations (such as different concentrations and viscosities) and diverse substrate types. By dynamically adjusting parameters to accommodate differences in material properties, it expands the application scenarios of inkjet printing technology.

[0047] In some implementations, reference Figure 2a As shown, obtaining the temporal characteristics of the current ink droplet falling onto the perovskite solar cell substrate, the spreading morphology characteristics of the ink droplet on the current perovskite solar cell substrate, and the environmental characteristics from the current ink droplet being ejected by the inkjet printer to the formation of a film on the perovskite solar cell substrate can include the following steps.

[0048] S210. Acquire the image sequence of the current ink droplet falling onto the perovskite solar cell substrate, the image of the ink droplet spreading pattern on the current perovskite solar cell substrate, and the environmental parameter sequence from the current ink droplet being ejected by the inkjet printer to the formation of a film on the perovskite solar cell substrate.

[0049] S220. Use the ink droplet feature extraction model to extract features from the image sequence and the spreading morphology image to obtain the temporal features corresponding to the image sequence and the spreading morphology features corresponding to the spreading morphology image; use the environmental feature extraction model to extract features from the environmental parameter sequence to obtain the environmental features.

[0050] The image sequence includes multiple images arranged in chronological order to reflect the dynamic characteristics of the ink droplet during its flight.

[0051] Spread morphology images are static morphological images of ink droplets after they fall onto a substrate, containing static information such as the ink droplet diffusion range, thickness distribution, film morphology, and satellite droplets.

[0052] The environmental parameter sequence includes various parameters related to the printing environment recorded in chronological order, such as printing chamber temperature and humidity, substrate temperature, inert gas flux, volatility, etc.

[0053] Specifically, during the printing process, the control system can acquire image sequences of the current ink droplets falling onto the perovskite solar cell substrate through an ink droplet observer and a high-speed camera, and acquire images of the current ink droplet spreading morphology on the perovskite solar cell substrate through a high-speed camera. It can also acquire relevant environmental parameter sequences through temperature and humidity sensors, laser interferometric gas phase sensors, etc.

[0054] The control system inputs the acquired image sequences and spreading morphology images into a pre-trained droplet feature extraction model to extract temporal features reflecting the dynamic changes of ink droplets from the image sequences, and spreading morphology features describing the spreading of ink droplets on the perovskite solar cell substrate from the spreading morphology images. The control system also inputs the acquired environmental parameter sequences into a pre-trained environmental feature extraction model to extract environmental features that comprehensively reflect the impact of environmental conditions on the printing process.

[0055] The ink droplet feature extraction model is obtained through model training based on a reference image sequence of a reference ink droplet falling onto a reference battery substrate and a reference spreading pattern image of the ink droplet on the reference battery substrate. Specifically, it includes: acquiring a reference image sequence of a reference ink droplet falling onto a reference battery substrate and a reference spreading pattern image of the reference ink droplet on the reference battery substrate; constructing image sample data based on the reference image sequence and the reference spreading pattern image; wherein the image sample data corresponds to image feature labels containing reference temporal features and reference spreading pattern features; inputting the image sample data into a first initial feature extraction model to obtain the initial image features output by the first initial feature extraction model; training the first initial feature extraction model based on the initial image features and image feature labels to obtain the trained ink droplet feature extraction model. The reference ink droplet is an ink droplet ejected by an inkjet printer under specific, standard, and controllable conditions for model training, feature analysis, or as a comparison benchmark, representing an ink droplet sample under ideal conditions or with typical characteristics. The reference cell substrate is used to receive reference ink droplets, serving as a carrier for droplet spreading and film formation, and as a perovskite cell substrate for model training, performance evaluation, or as a benchmark.

[0056] The environmental feature extraction model is trained based on a reference environmental parameter sequence, which can include environmental parameters from the ejection of the reference ink droplet to the film formation on the reference battery substrate. Specifically, the training process includes: acquiring the reference environmental parameter sequence from the ejection of the reference ink droplet to the film formation on the reference battery substrate; constructing environmental sample data based on the reference environmental parameter sequence; wherein the environmental sample data corresponds to environmental feature labels; inputting the environmental sample data into a second initial feature extraction model to obtain the initial environmental features output by the second initial feature extraction model; and training the second initial feature extraction model based on the initial environmental features and environmental feature labels to obtain the trained environmental feature extraction model.

[0057] For example, refer to Figure 2b As shown, a closed-loop control system for inkjet printing of three-terminal perovskite solar cells based on environmental perception and droplet monitoring is provided. The system includes a capacitive digital hygrometer, a miniature infrared thermal sensor, a laser interferometric gas phase sensor, a droplet observer, a reflector, a CCD (Charge-Coupled Device) high-speed camera, and an FPGA (Field Programmable Gate Array) industrial computer.

[0058] Among them, a miniature infrared thermal sensor is used to collect the internal temperature of the printer (or the internal temperature of the printing chamber) and the surface temperature of the stage (i.e., the surface temperature of the battery substrate) as a reference for adjusting ink droplet viscosity; a capacitive digital hygrometer is used to collect the internal humidity of the printer (or the internal humidity of the printing chamber) for predicting solvent evaporation rate; and a laser interferometric gas phase sensor is used to collect the evaporation intensity for predicting ink droplet drying state. Therefore, the environmental parameter sequence includes the electrical signals from the hygrometer, the infrared thermal sensor (including the internal temperature of the printer and the surface temperature of the stage), and the gas phase sensor.

[0059] In real-time online monitoring of ink droplet status, the ink droplet observer monitors the volume and generation frequency of ink droplets during the printhead ejection phase, and the speed, trajectory, and morphology during the ink droplet flight phase. The reflector optimizes ink droplet imaging quality and improves signal capture efficiency; it is an auxiliary device to the ink droplet observer. The reflector enables optical path design and supports coaxial or backlight illumination to ensure clear imaging of ink droplet edges (resolution ≤2μm). The CCD altitude camera enhances the capture and monitoring of ink droplet speed, trajectory, and morphology during the flight phase, and monitors the ink droplet spreading diameter and satellite droplet number during the substrate deposition phase. Therefore, the image sequence includes continuous images of the ink droplet flight phase acquired by the ink droplet observer and the CCD altitude camera, while the spreading morphology images include substrate surface images acquired by the CCD synchronous high-speed camera during the substrate deposition phase. In this specification, the image sequence and spreading morphology images can be collectively referred to as ink droplet monitoring images.

[0060] Furthermore, the ink droplet observer can employ a high-frequency pulsed light source (such as an LED or laser with a pulse width ≤ 1 μs) to avoid motion blur. A CCD synchronous high-speed camera (frame rate ≥ 50000 fps) is used in conjunction with a pulsed LED / laser light source (wavelength 850 nm) to capture the transient morphology of ink droplets. Multiple CCD high-speed cameras (such as the Phantom v2512, frame rate ≥ 50000 fps) are combined to enhance morphology capture and monitoring.

[0061] The industrial control computer (ICC) acquires the real-time monitored environmental parameter sequence and ink droplet monitoring images, and performs data analysis and defect identification. The ICC includes a synchronization control unit that utilizes an FPGA / high-speed DSP (Digital Signal Processor) to ensure that ink droplet excitation, the ink droplet observer, and the CCD high-speed camera are under global clock control. This achieves nanosecond-level synchronization of the printhead trigger signal, light source flash, and camera exposure, with an image processing delay of less than 0.5ms and a closed-loop response time of less than 2ms. By combining environmental sensing monitoring signals, ink droplet monitoring images, and a trained AI model, the ICC enables closed-loop control of ink droplet printing and inkjet commands.

[0062] It should be noted that, in some feasible implementations, the aforementioned closed-loop control system can also employ a triple-mode solution combining a high-speed camera, laser scattering, and piezoelectric sensing. This integrates optical, acoustic, and electrical signals to enhance monitoring dimensionality and reliability. Furthermore, the relevant sensors can be integrated into the nozzle (such as a MEMS micro-piezoelectric array) to achieve embedded miniaturization and reduce external interference.

[0063] The aforementioned closed-loop control system serves as the process control center for perovskite layer inkjet printing. Through a "sensing-analysis-execution" closed loop, it intercepts ink droplet anomalies at the nanoscale, fundamentally solving the pain points of high-precision solution-based manufacturing.

[0064] In some feasible ways, the droplet feature extraction model and the environment feature extraction model can be different network modules in the same feature extraction model.

[0065] For example, a PyTorch-based Backbone network can be used to extract environmental signal features and high-dimensional features of ink droplet monitoring images, respectively. The Backbone adopts a dual-channel feature extraction network structure (implemented in PyTorch), including an image feature extraction network branch (i.e., the ink droplet feature extraction model) and an environmental signal feature extraction network branch (i.e., the environmental feature extraction model).

[0066] The input to the image feature extraction network branch is a sequence of ink droplet flight images (i.e., the image sequence in this specification) plus a static image of the spread-out pattern (i.e., the spread-out pattern image in this specification). The network structure can employ 3D-CNN (3D-Convolutional Neural Networks, such as C3D) to process the ink droplet flight sequence images, extracting dynamic features (i.e., time-series features, such as trajectory shift trends and abrupt morphological changes); and 2D-CNN (2D-Convolutional Neural Networks, such as ResNet-34) to process the static image of the spread-out pattern, extracting static features (i.e., spread-out pattern features, such as edge irregularity and the number of satellite droplets). The output of this branch is a multidimensional image feature vector.

[0067] The input to the environmental signal feature extraction network branch includes time-series data on humidity, temperature, and volatile matter. The network structure can employ a combination of 1D-CNN (1D-Convolutional Neural Networks) and LSTM (Long Short-Term Memory). 1D-CNN can be used to extract local fluctuation features (such as sudden increases / decreases in humidity), while LSTM can capture long-term temporal dependencies (such as the lag correlation between temperature and volatile matter). The output of this network branch is a multi-environment feature vector (i.e., environmental features, including information on environmental stability and coupling relationships, where coupling relationships refer to the synergistic effects between humidity, temperature, and volatile matter).

[0068] It should be noted that the description of the training process of the ink droplet feature extraction model and the environment feature extraction model is similar to the description of the implementation process of the ink droplet feature extraction model and the environment feature extraction model mentioned above, and will not be repeated here.

[0069] In some cases, the spreading morphology image may also include static images of ink droplets spreading on the perovskite solar cell substrate at multiple time points. The specific structures of the ink droplet feature extraction model and the environmental feature extraction model can be determined according to actual application requirements, etc., and are not limited in this specification.

[0070] In some implementations, defect identification based on temporal features, morphological features, and environmental features to obtain current defect data may include: inputting temporal features, morphological features, and environmental features into a pre-trained defect type classification model, classifying defects, and outputting the current defect type.

[0071] The defect type classification model is trained based on the reference temporal features corresponding to the reference image sequence, the reference spreading morphology features corresponding to the reference spreading morphology image, the reference environmental features corresponding to the reference environmental parameter sequence, and the defect type label corresponding to the reference ink droplet.

[0072] Specifically, the control system can use a pre-trained defect type classification model to perform defect analysis and classification on the temporal features, spread morphology features and environmental features obtained by the aforementioned method, and finally obtain the currently existing defect type output by the model, i.e., the current defect type.

[0073] The defect type classification model is trained based on reference temporal features corresponding to the reference image sequence, reference spreading morphology features corresponding to the reference spreading morphology image, reference environmental features corresponding to the reference environmental parameter sequence, and defect type labels corresponding to the reference ink droplets. Specifically, it includes: reference temporal features corresponding to the reference image sequence, reference spreading morphology features corresponding to the reference spreading morphology image, and reference environmental features corresponding to the reference environmental parameter sequence; constructing feature sample data based on the reference temporal features, reference spreading morphology features, and reference environmental features; wherein, the feature sample data corresponds to defect type labels; inputting the feature sample data into the initial defect classification model to obtain the initial defect type output by the initial defect classification model; training the initial defect classification model based on the initial defect type and defect type labels to obtain the trained defect type classification model.

[0074] For example, taking the aforementioned image feature extraction network branch and environmental signal feature extraction network branch as examples, the image feature extraction network branch outputs a multi-dimensional image feature vector containing temporal features and unfolding morphological features, while the environmental signal feature extraction network branch outputs a multi-environment feature vector. The control system inputs the multi-dimensional image feature vector and the multi-environment feature vector into a pre-trained defect type classification network (e.g., MM-DTCN (Multi-Modal Deep Temporal Convolution Network)), and the defect type classification network outputs the identified current defect type.

[0075] Specifically, the implementation process of MM-DTCN is as follows: The input layer receives two types of features from the backbone (including multi-dimensional image feature vectors and multi-environment feature vectors). The fusion layer fuses the two types of features through a cross-attention mechanism. The feature concatenation and normalization module concatenates the output features of the two sub-modules into a multi-dimensional fused feature, which is then normalized through a LayerNorm layer. The classification output layer performs defect identification based on the normalized fused features and outputs the identified defect type.

[0076] The fusion layer can be designed to include an image-environment attention submodule and an environment-image attention submodule. The image-environment attention submodule mainly uses multi-dimensional image feature vectors as "queries" and multi-environment feature vectors as "key-value pairs" to calculate the weight distribution of environment features on image features; the environment-image attention submodule mainly uses multi-environment feature vectors as "queries" and multi-dimensional image feature vectors as "key-value pairs" to calculate the weight distribution of image features on environment features.

[0077] In some other feasible ways, the current defect data includes the current defect type and the predicted probability corresponding to the current defect type.

[0078] For example, the defect types include pinhole defects, crack defects, satellite drop defects, edge warping defects, and uneven stacking defects. Taking the MM-DTCN mentioned above as an example, the classification output layer can be designed to include 5 neurons (corresponding to the 5 types of defects), which are activated by the Softmax function to output the probability values ​​of each type of defect.

[0079] Specifically, a lightweight fully connected network can be designed to output the probability of defect types, including a first fully connected layer, a second fully connected layer, and an output activation layer. The first fully connected layer is used to compress the fused features, introducing non-linearity through the ReLU activation function, while adding a Dropout layer to suppress overfitting. The second fully connected layer is used to map the compressed features to a 5-dimensional output. The output activation layer uses the Softmax function to transform the 5-dimensional output into a probability distribution (summing to 1), directly outputting each defect type and its corresponding predicted probability.

[0080] If the defect types include pinhole defects, crack defects, satellite drop defects, and uneven spreading defects, then a classification output layer with four neurons can be designed, which can be flexibly set according to the actual application scenario or requirements.

[0081] Furthermore, in some feasible ways, after classifying defects based on temporal features, morphological features, and environmental features using a defect type classification model, multiple defect types identified by the defect type classification model can be obtained, each defect type having its own prediction probability; if the prediction probability of a specified defect type is greater than or equal to a preset probability threshold, the defect type classification model determines the specified defect type as the current defect type and outputs it; wherein, the specified defect type is any one of the multiple defect types identified by the defect type classification model.

[0082] Alternatively, based on the predicted probabilities of the multiple identified defect types, the defect type classification model can determine the defect type corresponding to the highest predicted probability as the current defect type and output it.

[0083] It should be noted that the description of the training process of the defect type classification model is similar to the description of the implementation process of the defect type classification model described above, and will not be repeated here.

[0084] In some cases, the droplet feature extraction model, the environmental feature extraction model, and the defect type classification model can be different network modules in the same defect recognition model. The specific structure of the defect type classification model can be determined according to actual application requirements, etc., and is not limited in this specification.

[0085] In some implementations, the image sequence includes a first image at a first time and a second image at a second time; during the acquisition of the image sequence, the inkjet printing control method may further include: calculating the volume difference of the current ink droplet based on the acquired first and second images to obtain the absolute value of the volume difference of the current ink droplet; if the absolute value of the volume difference is greater than a preset volume difference threshold, adjusting the pulse voltage amplitude of the inkjet printer according to preset voltage amplitude increment data.

[0086] In the process of acquiring image sequences by the image acquisition device, one of the first and second moments is the current moment (or the latest moment), and the other moment is located before the current moment.

[0087] Specifically, when continuously acquiring images of the current ink droplet falling onto the perovskite solar cell substrate using an image acquisition device, assuming the second moment is the current moment and the first moment is before the second moment, the image acquired by the image acquisition device at the current moment is recorded as the second image, and the control system can obtain the first image at the first moment from the images acquired by the image acquisition device at previous moments.

[0088] The control system uses image processing algorithms to analyze and process the first and second images. These algorithms may include image segmentation and feature extraction, aiming to accurately identify the outline and boundaries of the ink droplets from the images, and then calculate the volume of the ink droplets at the first and second moments. Then, by comparing the volumes of the ink droplets at these two moments, the absolute value of the volume difference is obtained, reflecting the volume stability of the ink droplets during flight.

[0089] The control system compares the calculated absolute value of the volume difference with a preset volume difference threshold. When the absolute value of the volume difference exceeds the preset threshold, it indicates that the current droplet volume change exceeds the allowable range, which may adversely affect print quality. In this case, the control system adjusts the pulse voltage amplitude of the inkjet printer according to preset voltage amplitude increment data. The pulse voltage amplitude is one of the important factors affecting droplet ejection speed and volume. By adjusting the pulse voltage amplitude, the droplet ejection characteristics can be changed, thereby restoring the droplet volume to the normal range.

[0090] Furthermore, when the absolute value of the volume difference is greater than the preset volume difference threshold, and the ink droplet volume corresponding to the second image is smaller than the ink droplet volume corresponding to the first image, the pulse voltage amplitude of the inkjet printer is increased according to the preset voltage amplitude increment data; when the absolute value of the volume difference is greater than the preset volume difference threshold, and the ink droplet volume corresponding to the second image is greater than the ink droplet volume corresponding to the first image, the pulse voltage amplitude of the inkjet printer is decreased according to the preset voltage amplitude increment data.

[0091] For example, the preset voltage amplitude increment data can be pre-calibrated experimentally. Assume the absolute value of the volume difference is calculated by calculating the rate of volume change, the preset volume difference threshold is set to 2%, and the preset voltage amplitude increment data is 5%. When the calculated absolute value of the volume difference... When the value is greater than 2%, reduce the pulse voltage amplitude of the inkjet printer by 5%.

[0092] For example, the preset voltage amplitude increment data can also be determined by looking up a table or by rules (such as fuzzy control rules).

[0093] In some feasible ways, the first image is adjacent to the second image.

[0094] Furthermore, the control system controls the inkjet printer to eject ink within a preset droplet volume range. During image sequence acquisition, if the volume of the current droplet exceeds the preset droplet volume range (e.g., 20 ± 0.5 pL) based on the acquired image, the control system can adjust the pulse voltage amplitude of the inkjet printer according to a certain voltage amplitude increment to ensure that the droplet volume remains within the allowable range. The method for obtaining the voltage amplitude increment here is similar to the method for obtaining the preset voltage amplitude increment data mentioned above; however, the voltage amplitude increment here may be the same as or different from the preset voltage amplitude increment data.

[0095] It should be noted that the preset volume difference threshold and preset voltage amplitude increment data can be set according to actual application scenarios or requirements, and are not specifically limited in this manual.

[0096] In some embodiments, the environmental parameter sequence includes a first environmental parameter at a third time point and a second environmental parameter at a fourth time point. During the acquisition of the environmental parameter sequence, the inkjet printing control method may further include: calculating environmental parameter differences based on the acquired first and second environmental parameters to obtain environmental parameter difference data; wherein the environmental parameter difference data includes at least one of the absolute values ​​of the temperature difference in the inkjet printing chamber, the absolute values ​​of the humidity difference in the inkjet printing chamber, and the absolute values ​​of the ink droplet evaporation intensity difference; if the environmental parameter difference data includes the absolute value of the temperature difference, and the absolute value of the temperature difference is greater than a preset temperature difference threshold, the temperature of the inkjet printing chamber is adjusted according to the preset temperature data; if the environmental parameter difference data includes the absolute value of the humidity difference, and the absolute value of the humidity difference is greater than a preset humidity difference threshold, the humidity of the inkjet printing chamber is adjusted according to the preset humidity data, and / or, the inert gas flux is adjusted according to the first preset flux increment data; if the environmental parameter difference data includes the absolute value of the ink droplet evaporation intensity difference, and the absolute value of the ink droplet evaporation intensity difference is greater than a preset intensity difference threshold, the inert gas flux is adjusted according to the second preset flux increment data.

[0097] In the process of collecting environmental parameter sequences using relevant sensing devices, one of the third and fourth moments is the current moment (or the latest moment), and the other moment is located before the current moment.

[0098] The absolute value of the temperature difference in the inkjet printing chamber refers to the absolute difference between the temperatures inside the printing chamber of the inkjet printer at the third and fourth moments. Temperature changes affect the physical properties of ink, such as viscosity and surface tension, which in turn affect the ejection and formation of ink droplets.

[0099] The absolute value of the humidity difference in the inkjet printing chamber refers to the absolute value of the difference in humidity inside the printing chamber of the inkjet printer at two different times. Humidity has a significant impact on the drying speed and morphology of ink droplets; improper humidity may lead to uneven diffusion of ink droplets on the substrate or the formation of cracks.

[0100] The absolute value of the difference in ink droplet evaporation intensity reflects the change in ink droplet evaporation intensity between two moments. Ink droplet evaporation intensity is related to factors such as gas composition and airflow velocity in the printing environment; abnormal evaporation intensity can affect the size and shape of ink droplets.

[0101] Specifically, when collecting environmental parameter sequences using relevant sensing devices, assuming the fourth moment is the current moment and the third moment is before the fourth moment, the environmental parameters collected at the current moment are recorded as the second environmental parameters. The control system can obtain the first environmental parameters at the third moment from the environmental parameters collected at previous moments. The first environmental parameters include at least one of the following: the temperature of the inkjet printing chamber, the humidity of the inkjet printing chamber, and the ink droplet evaporation intensity. The parameters in the second environmental parameters are the same as those in the first environmental parameters.

[0102] The control system compares and analyzes the specific parameters included in the first and second environmental parameters, and calculates the absolute value of the difference between the corresponding parameters, which is the environmental parameter difference data. Therefore, the environmental parameter difference data includes at least one of the absolute values ​​of the temperature difference in the inkjet printing chamber, the absolute value of the humidity difference in the inkjet printing chamber, and the absolute value of the difference in ink droplet evaporation intensity.

[0103] If the environmental parameter difference data includes the absolute value of the temperature difference in the inkjet printing chamber, the control system compares this absolute value with a preset temperature difference threshold. When the absolute value of the temperature difference is greater than the preset temperature difference threshold, the control system adjusts the temperature of the inkjet printing chamber according to the preset temperature data to minimize the impact of abnormal temperature changes as quickly as possible. Furthermore, when the absolute value of the temperature difference is greater than the preset temperature difference threshold, if the inkjet printing chamber temperature in the second environmental parameter is lower than the inkjet printing chamber temperature in the first environmental parameter, the inkjet printing chamber temperature is increased according to the preset temperature data; if the inkjet printing chamber temperature in the second environmental parameter is greater than the inkjet printing chamber temperature in the first environmental parameter, the inkjet printing chamber temperature is decreased according to the preset temperature data.

[0104] If the environmental parameter difference data includes the absolute value of the humidity difference in the inkjet printing chamber, the control system compares this absolute value with a preset humidity difference threshold. When the absolute value of the humidity difference is greater than the preset humidity difference threshold, the control system adjusts the humidity of the inkjet printing chamber according to the preset humidity data, and / or adjusts the inert gas flux according to the first preset flux increment data to reduce the impact of abnormal humidity changes as quickly as possible. Further, when the absolute value of the humidity difference is greater than the preset humidity difference threshold, if the humidity of the inkjet printing chamber in the second environmental parameter is less than the humidity of the inkjet printing chamber in the first environmental parameter, the humidity of the inkjet printing chamber is increased according to the preset humidity data, and / or the inert gas flux is decreased according to the first preset flux increment data; if the humidity of the inkjet printing chamber in the second environmental parameter is greater than the humidity of the inkjet printing chamber in the first environmental parameter, the humidity of the inkjet printing chamber is decreased according to the preset humidity data, and / or the inert gas flux is increased according to the first preset flux increment data.

[0105] If the environmental parameter difference data includes the absolute value of the droplet evaporation intensity difference, the control system compares this absolute value with a preset intensity difference threshold. When the absolute value of the droplet evaporation intensity difference is greater than the preset intensity difference threshold, the control system adjusts the inert gas flux according to the second preset flux increment data to reduce the impact of abnormal volatility changes as quickly as possible. Furthermore, when the absolute value of the droplet evaporation intensity difference is greater than the preset intensity difference threshold, if the droplet evaporation intensity in the second environmental parameter is less than the droplet evaporation intensity in the first environmental parameter, the inert gas flux is increased according to the second preset flux increment data; if the droplet evaporation intensity in the second environmental parameter is greater than the droplet evaporation intensity in the first environmental parameter, the inert gas flux is decreased according to the second preset flux increment data.

[0106] For example, the preset temperature data and preset humidity data can be pre-set standard temperature and humidity, and the first preset flux increment data and the second preset flux increment data can be pre-calibrated experimentally or determined by looking up tables or rules (such as fuzzy control rules).

[0107] Assuming a standard temperature of 25℃ and a standard humidity of 35%RH are set as the preset temperature data, and the preset temperature difference threshold is 1℃, the calculated absolute value of the temperature difference... When the temperature exceeds 1°C, the control system activates PID (Proportional-Integral-Derivative) temperature control to adjust the temperature of the inkjet printing chamber to 25°C.

[0108] Assuming a standard humidity of 35%RH is set as the preset humidity data, and a preset humidity difference threshold of 3%, the first preset flux increment data is determined to be 30% through experimental calibration or table lookup. When the calculated absolute value of the humidity difference... When the humidity is greater than 3%, the control system will adjust the humidity of the inkjet printing chamber to 35%RH, and / or trigger the nitrogen gas curtain to adjust the inert gas flow rate to 30%.

[0109] Assuming the absolute value of the droplet evaporation intensity difference is calculated using the rate of change (or by calculating the difference in evaporation intensity values), and the preset intensity difference threshold is set to 5%, the second preset flux increment data is determined to be 20% through experimental calibration or table lookup. When the calculated absolute value of the droplet evaporation intensity difference (i.e., the absolute value of the droplet evaporation intensity change rate) is greater than 5%, the control system adjusts the inert gas flux by 20%.

[0110] For example, the preset temperature data and preset humidity data can also be temperature increment data and humidity increment data, which can be determined in advance through experimental calibration, or through table lookup or rules (such as fuzzy control rules). The specific situation is similar to the description of the second preset flux increment data mentioned above, and will not be repeated here.

[0111] In some feasible implementations, the first environmental parameter is adjacent to the second environmental parameter, and the first preset flux increment data and the second preset flux increment data can be inert gas flow rate increment data.

[0112] Furthermore, the control system controls the inkjet printer to spray ink within preset temperature and humidity ranges, and controls the evaporation intensity of ink droplets within a preset intensity range. During the acquisition of environmental parameter sequences, if the acquired environmental parameters determine that the current printing chamber temperature exceeds the preset temperature range (e.g., 25±2℃), and / or that the current printing chamber humidity exceeds the preset humidity range (e.g., 35±5%RH), and / or that the current evaporation intensity exceeds the preset intensity range (e.g., 0.6±0.05μm / s), the control system can adjust the printing chamber temperature and / or printing chamber humidity and / or inert gas flux according to certain rules. The specific adjustment rules are similar to those described in the above example and will not be repeated here.

[0113] It should be noted that the inert gas can be nitrogen, argon, or other gases. The preset volume difference threshold and preset voltage amplitude increment data can be set according to the actual application scenario or requirements, and are not specifically limited in this manual.

[0114] In some implementations, the defect types include pinhole defects, crack defects, satellite droplet defects, edge warping defects, and uneven stacking defects.

[0115] Among them, the parameter adjustment strategies corresponding to the pinhole defect type include reducing the temperature of the perovskite solar cell substrate according to preset temperature increment data, and / or increasing the humidity of the inkjet printing chamber according to preset humidity increment data.

[0116] The parameter adjustment strategies corresponding to the crack defect type include increasing the flow rate gradient of the inert gas according to the preset flow rate increment data, and / or adjusting the temperature of the central or edge region according to the preset temperature difference between the central and edge regions of the perovskite solar cell substrate.

[0117] The parameter adjustment strategy corresponding to the satellite droplet defect type includes reducing the inkjet voltage of the inkjet printer according to preset voltage increment data, and / or increasing the pulse width of the inkjet printer according to a first preset time increment data. The parameter adjustment strategy corresponding to the edge warping defect type includes reducing the ink ejection volume of the edge nozzle channel of the inkjet printer according to the preset ink volume increment data, and / or delaying the ink ejection time of the edge nozzle channel according to the second preset time increment data.

[0118] The parameter adjustment strategy corresponding to the uneven stacking defect type includes grayscale printing calibration according to the preset ink droplet volume, and / or adjusting the inkjet printer's inkjet trigger time according to the preset timing compensation data; wherein, the preset timing compensation data is determined based on the deviation between the actual landing position and the expected landing position of the current ink droplet.

[0119] For the parameter adjustment strategy for each defect type, it can be carried out according to any one of the following two situations: (1) First, obtain the environmental parameter difference data based on the environmental parameters of two adjacent times, adjust the environmental parameters based on the environmental parameter difference data, and then adjust the printing process parameters according to the parameter adjustment strategy corresponding to the defect type. For the specific implementation process of obtaining the environmental parameter difference data and adjusting the environmental parameters based on the environmental parameter difference data, please refer to the above "The environmental parameter sequence includes the first environmental parameter at the third time and the second environmental parameter at the fourth time; In the process of obtaining the environmental parameter sequence, the inkjet printing control method may also include: calculating the environmental parameter difference based on the obtained first environmental parameter and second environmental parameter to obtain the environmental parameter difference data; wherein, the environmental parameter difference data includes at least one of the absolute value of the temperature difference of the inkjet printing chamber, the absolute value of the humidity difference of the inkjet printing chamber, and the absolute value of the ink droplet evaporation intensity difference; if the environmental parameter difference data includes the absolute value of the temperature difference, and the temperature If the absolute value of the difference is greater than the preset temperature difference threshold, the temperature of the inkjet printing chamber is adjusted according to the preset temperature data; if the environmental parameter difference data includes the absolute value of the humidity difference and the absolute value of the humidity difference is greater than the preset humidity difference threshold, the humidity of the inkjet printing chamber is adjusted according to the preset humidity data, and / or, the inert gas flux is adjusted according to the first preset flux increment data; if the environmental parameter difference data includes the absolute value of the droplet evaporation intensity difference and the absolute value of the droplet evaporation intensity difference is greater than the preset intensity difference threshold, the inert gas flux is adjusted according to the second preset flux increment data. The relevant part of the explanation will not be repeated here; (2) There is no need to consider the environmental parameter difference data between two adjacent environmental parameters, nor is it necessary to adjust the environmental parameters according to the environmental parameter difference data. That is, there is no need to perform the step of "obtaining the environmental parameter difference data according to the environmental parameters at two adjacent times and adjusting the environmental parameters according to the environmental parameter difference data", and the printing process parameters are adjusted directly according to the parameter adjustment strategy corresponding to the defect type. The first scenario described above involves adjusting the environmental parameters based on the differences between environmental parameters at two adjacent moments, and then adjusting the printing process parameters according to the defect type. This approach can stabilize the environmental parameters for inkjet printing, minimizing the impact of environmental parameters on defects. The second scenario describes adjusting the printing process parameters directly based on the defect type without pre-adjusting the environmental parameters, thus reducing the number of parameter adjustments and the complexity of parameter adjustment.

[0120] Accordingly, the printing process parameters can be adjusted based on the target parameter adjustment strategy in either of the following two cases: (1) Based on the environmental parameter difference data obtained from the first environmental parameter at the third time and the second environmental parameter at the fourth time, and the environmental parameters are adjusted according to the environmental parameter difference data, the printing process parameters are adjusted based on the target parameter adjustment strategy corresponding to the current defect type; (2) Without performing the process of “obtaining environmental parameter difference data from the first environmental parameter at the third time and the second environmental parameter at the fourth time, and adjusting the environmental parameters according to the environmental parameter difference data”, the printing process parameters are directly adjusted based on the target parameter adjustment strategy corresponding to the current defect type.

[0121] Among them, pinhole defects (hereinafter referred to as pinhole defects) refer to the defects in which tiny holes (usually with a diameter <5μm) appear after ink droplets form a film; crack defects (hereinafter referred to as crack defects) refer to the defects in which linear or branched cracks appear in the film layer; satellite droplet defects (hereinafter referred to as satellite droplets) refer to the defects in which tiny ink droplets separate during the main ink droplet ejection process, resulting in extra small spots on the substrate, which can easily cause local overthickness; edge warping defects (hereinafter referred to as edge warping) refer to the defects in which the edge area of ​​the ink droplet curls up or thickens after the ink droplet spreads on the substrate; and uneven stacking defects (hereinafter referred to as uneven stacking) refer to the defects in which the thickness of ink droplet stacking varies too much in different areas within the same film layer.

[0122] Specifically, pinhole defects are mostly caused by excessively rapid ink evaporation or insufficient substrate wettability. The corresponding parameter adjustment strategy is to reduce the temperature of the perovskite solar cell substrate according to preset temperature increment data (e.g., 3~5℃) to improve the wettability of the ink on the substrate surface, and / or to increase the humidity of the inkjet printing chamber according to preset humidity increment data (e.g., 5~10%RH) to slow down the evaporation rate of the ink solvent.

[0123] Crack defects are mainly caused by uneven drying stress (i.e., uneven ink evaporation rate, such as a large difference in evaporation rate between the edge and center regions). The corresponding parameter adjustment strategy is to increase the inert gas flow rate gradient according to the preset flow rate increment data (e.g., increase the gas flow rate in a certain region by 20%), so as to homogenize the evaporation rate of the ink droplet surface through the gas flow difference between different regions, and / or, according to the preset temperature difference (e.g., 2~5℃) between the center and edge regions of the perovskite solar cell substrate, adjust the temperature of the center or edge regions of the substrate to maintain a certain temperature difference between the center and edge regions to improve the problem of uneven drying stress.

[0124] Satellite droplet defects are often caused by excessive driving voltage or improper pulse parameters. The corresponding parameter adjustment strategy is to reduce the inkjet printer's inkjet voltage by a preset voltage increment, and / or to increase the duration of the voltage pulse by a first preset time increment (e.g., voltage rise time, fall time, and hold time are adjusted by ±1μs per pulse) to optimize the morphological stability of ink droplets when they are ejected from the printhead.

[0125] Edge warping defects are mainly caused by excessive ink volume at the edges and asynchronous drying speeds with the center. The corresponding parameter adjustment strategies are to reduce the ink volume of the inkjet printer's edge nozzle channels according to preset ink volume increment data to avoid excessive ink volume in the edge area, and / or to delay the ink jetting time of the edge nozzles according to a second preset time increment data (e.g., 10μs) to balance the spreading speed of the center and edge areas.

[0126] Uneven droplet stacking defects are often caused by abnormal droplet volume or droplet landing point positioning deviation. The corresponding parameter adjustment strategy involves calibrating grayscale printing parameters according to a preset droplet volume, achieving volume compensation by adjusting the inkjet volume (grayscale level), and / or adjusting the inkjet printer's inkjet trigger time according to preset timing compensation data to correct positioning deviations. The preset timing compensation data is determined based on the deviation between the actual and expected droplet landing positions. The actual landing position is obtained through trajectory analysis of the aforementioned image sequence, while the expected landing position is determined based on the target area where the current droplet should land on the perovskite solar cell substrate. Therefore, adjusting the inkjet printer's inkjet trigger time according to the preset timing compensation data is essentially adjusting the inkjet printer's inkjet trigger time based on the current droplet's positioning offset.

[0127] It should be noted that the preset temperature increment data, preset humidity increment data, preset flow rate increment data, preset temperature difference, preset voltage increment data, first preset time increment data, preset ink volume increment data, and second preset time increment data can be pre-calibrated experimentally and can be range data. The preset parameter adjustment rules can be a mapping table or rule base established based on these data. When matching the current defect data in the preset parameter adjustment rules, the target parameter adjustment strategy containing specific adjustment data can be determined by looking up the table or reasoning from the rule base.

[0128] In some implementations, after obtaining the current defect type, the method further includes: determining the defect cause corresponding to the current defect type based on the correspondence between defect types and defect causes; and matching the current defect type in a preset parameter adjustment rule, including: matching the current defect type and the corresponding defect cause in a preset parameter adjustment rule; the preset parameter adjustment rule contains the mapping relationship between the defect type, the corresponding defect cause, and the parameter adjustment strategy.

[0129] Specifically, relevant personnel can determine the defect causes (i.e., the specific reasons for the occurrence of defect types) through experiments and analysis, and establish a correspondence between defect types and defect causes. This design can help to move defect data from phenomenon description to cause analysis. For example, the cause of satellite droplet defects may be excessively high printhead voltage or abnormal ink viscosity; the cause of excessively large ink droplet spreading may be excessively low substrate temperature or excessively high ambient humidity, etc. For example, for typical failure modes of perovskite solar cell inkjet printing, five core defects are defined: ① Pinhole defects (holes with a diameter <5μm, caused by insufficient ink droplet volume or excessively rapid evaporation); ② Crack defects (linear or branched gaps, caused by uneven drying stress); ③ Satellite drops (tiny droplets around the main ink droplet, caused by excessively strong driving pulses leading to droplet splitting, easily resulting in local excessive thickness); ④ Edge warping (the edge of the film is higher than the center, caused by uneven spreading speed); ⑤ Uneven stacking (caused by ink droplet positioning deviation or ink droplet volume fluctuation).

[0130] Accordingly, the preset parameter adjustment strategy can include the mapping relationship between the defect type, the corresponding defect cause, and the parameter adjustment strategy. For example, for the satellite drop defect type and the cause being excessive printhead voltage, the corresponding parameter adjustment strategy could be to reduce the printhead voltage by 10%; for the satellite drop defect type and the cause being abnormal ink viscosity, the corresponding parameter adjustment strategy could be to increase the ink temperature to reduce the viscosity, and so on.

[0131] After identifying defects based on temporal characteristics, morphological characteristics, and environmental characteristics to determine the current defect type, the corresponding defect cause can be determined based on the correspondence between the defect type and the defect cause. Then, the current defect type and its corresponding defect cause can be matched against preset parameter adjustment rules to achieve accurate matching of the appropriate target parameter adjustment strategy.

[0132] In some feasible approaches, defect types can be defined based on specific defect types and their corresponding causes. For example, for the satellite drop defect type and the cause being excessive nozzle voltage, the corresponding defect type can be defined as satellite drop defect caused by excessive nozzle voltage.

[0133] For example, the defect causes and matching target parameter adjustment strategies for pinhole defects, crack defects, satellite droplet defects, edge warping defects, and uneven stacking defects can be as follows: Assuming the identified current defect type is a pinhole defect, and the defect is caused by excessively rapid ink droplet evaporation, the matching target parameter adjustment strategy could be to increase the humidity of the printing chamber by 5%RH (i.e., the preset humidity increment data) to slow down solvent evaporation, and to decrease the temperature of the battery substrate by 3℃ (i.e., the preset temperature increment data) to prolong the drying time and promote liquid film spreading.

[0134] Assuming the identified defect type is a crack, and the cause is uneven drying stress (specifically, the drying stress in the edge region is greater than that in the center region), the target parameter adjustment strategy could be to increase the flow rate of inert gas (such as nitrogen) in the edge region of the substrate by 20% to increase the flow rate gradient between the center and edge regions, thereby increasing the flow rate from the center to the edge and balancing the evaporation rate. Alternatively, zonal temperature control could be used (controlling the temperature in the edge region of the substrate to be 2°C higher than that in the center region) to alleviate edge shrinkage stress.

[0135] Assuming the identified defect type is uneven stacking, and the cause is abnormal droplet volume, the matching target parameter adjustment strategy can be to initiate grayscale printing calibration and droplet compensation based on a standard droplet volume (i.e., a preset droplet volume) of 20 pL. If the cause is droplet positioning deviation, the matching target parameter adjustment strategy can be to first calculate the droplet offset, then initiate automatic printhead calibration, and adjust the inkjet printer's inkjet trigger time based on the droplet offset.

[0136] Assuming the identified current defect type is a satellite droplet defect, and the defect cause is an excessively strong driving pulse, the matching target parameter adjustment strategy can be to reduce the driving voltage by 5% (which can be achieved by reducing the amplitude of the piezoelectric ceramic), or to adjust the voltage pulse width from 5μs to 6μs to smooth the injection force and avoid liquid column breakage.

[0137] Assuming the identified current defect type is edge warping, and the defect is caused by the ink droplet spreading speed being too fast at the edge, the matching target parameter adjustment strategy can be to reduce the ink ejection volume of the edge nozzle channel by 10%, or to delay the ink ejection time of the edge printhead channel by 2μs, in order to adjust the ejection sequence and balance the spreading speed.

[0138] It should be noted that, in this specification, the preset incremental data may be positive or negative. The defect causes corresponding to any defect type may also include other factors not mentioned in the above embodiments, in which case corresponding parameter adjustment strategies can be further set according to each defect type and its corresponding defect causes.

[0139] In some implementations, each defect type has its own priority. If multiple current defect types are obtained, before matching them according to the current defect types in the preset parameter adjustment rules to determine the target parameter adjustment strategy, the inkjet printing control method may further include: determining the target defect type according to the priority of each current defect type. Accordingly, determining the target parameter adjustment strategy by matching the current defect data according to the preset parameter adjustment rules may include: determining the target parameter adjustment strategy by matching the target defect type according to the preset parameter adjustment rules.

[0140] Specifically, during actual printing, ink droplets may exhibit multiple defects simultaneously. If parameter adjustment strategies for all defect types are directly applied together, it could lead to parameter conflicts and exacerbate the defects. Therefore, a priority attribute can be assigned to each defect type. Priority is typically set based on factors such as the defect's impact on battery performance, the risk of defect propagation, and the urgency of the adjustment.

[0141] When multiple current defect types are identified, the control system can select the current defect type with the highest priority as the target defect type based on the priority of each current defect type. Then, the control system determines and executes the corresponding target parameter adjustment strategy by matching the target defect type with preset parameter adjustment rules.

[0142] For example, taking the defect type classification network MM-DTCN mentioned earlier as an example, assuming that the classification network outputs multiple current defect types, including pinhole defects, crack defects and satellite drop defects, and that crack defects have the highest priority, pinhole defects have the second highest priority, and satellite drop defects have the lowest priority, then the control system can identify crack defects as the target defect type and match it to obtain the target parameter adjustment strategy.

[0143] In some feasible approaches, each defect type can be assigned a corresponding weight, which then serves as a priority. For example, the weight can be assigned to each defect type based on factors such as the degree of impact of the defect on battery performance, the risk of defect propagation, and the urgency of adjustment.

[0144] Furthermore, after a period of time, the control system can count the total number of defects during this period and calculate the defect probability (or defect frequency). If the defect probability is greater than a preset probability threshold (e.g., 3% or 5%), the control system can activate printhead pulse correction and nitrogen gas protection enhancement before the next printing cycle.

[0145] This specification also provides an inkjet printing control device for perovskite solar cells, see reference. Figure 3 As shown, the inkjet printing control device 300 includes: a feature acquisition module 310, a defect identification module 320, an adjustment strategy determination module 330, and a parameter adjustment module 340.

[0146] The feature acquisition module 310 is used to acquire the temporal features of the current ink droplet falling onto the perovskite solar cell substrate, the spreading morphology features of the ink droplet on the current perovskite solar cell substrate, and the environmental features from the time the current ink droplet is ejected by the inkjet printer to the time it forms a film on the perovskite solar cell substrate.

[0147] The defect identification module 320 is used to identify defects based on temporal characteristics, morphological characteristics, and environmental characteristics to obtain the current defect type.

[0148] The adjustment strategy determination module 330 is used to match the current defect type with the preset parameter adjustment rules to determine the target parameter adjustment strategy; wherein, the preset parameter adjustment rules contain the mapping relationship between defect type and parameter adjustment strategy.

[0149] The parameter adjustment module 340 is used to adjust the printing process parameters based on the target parameter adjustment strategy, so as to perform subsequent inkjet printing based on the adjusted printing process parameters.

[0150] Specific limitations regarding the inkjet printing control device for perovskite solar cells can be found in the limitations of the inkjet printing control method for perovskite solar cells described above, and will not be repeated here. Each module in the aforementioned inkjet printing control device for perovskite solar cells can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0151] This specification also provides a computer device, see embodiments thereof. Figure 4 As shown, the computer device 400 includes a memory 410, a processor 420, and a computer program 430 stored in the memory 410 and executable on the processor 420. When the processor 420 executes the computer program 430, it implements the inkjet printing control method for perovskite solar cells according to any of the aforementioned embodiments.

[0152] This specification also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the inkjet printing control method for perovskite solar cells according to any of the foregoing embodiments.

[0153] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0154] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0155] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0156] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0157] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0158] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for controlling inkjet printing of perovskite solar cells, characterized in that, The method includes: The timing characteristics of the current ink droplet falling onto the perovskite solar cell substrate, the spreading morphology characteristics of the ink droplet on the current perovskite solar cell substrate, and the environmental characteristics from the current ink droplet being ejected by the inkjet printer to the formation of a film on the perovskite solar cell substrate are obtained. Defects are identified based on the temporal characteristics, the spreading morphology characteristics, and the environmental characteristics to obtain the current defect type. The target parameter adjustment strategy is determined by matching the current defect type with preset parameter adjustment rules; wherein, the preset parameter adjustment rules include a mapping relationship between defect types and parameter adjustment strategies. The printing process parameters are adjusted based on the target parameter adjustment strategy, so that subsequent inkjet printing can be performed based on the adjusted printing process parameters.

2. The method according to claim 1, characterized in that, Obtaining the temporal features, the spreading morphology features, and the environmental features includes: The image sequence of the current ink droplet falling onto the perovskite solar cell substrate, the image of the spread pattern of the ink droplet on the perovskite solar cell substrate, and the environmental parameter sequence from the current ink droplet being ejected by the inkjet printer to the formation of a film on the perovskite solar cell substrate are obtained. The image sequence and the spreading pattern image are subjected to feature extraction using an ink droplet feature extraction model to obtain the temporal features corresponding to the image sequence and the spreading pattern features corresponding to the spreading pattern image; the environmental parameter sequence is subjected to feature extraction using an environmental feature extraction model to obtain the environmental features.

3. The method according to claim 2, characterized in that, The step of identifying defects based on the temporal characteristics, the spreading morphology characteristics, and the environmental characteristics to obtain the current defect type includes: The temporal features, the spreading morphology features, and the environmental features are input into a pre-trained defect type classification model to perform defect classification and output the current defect type.

4. The method according to claim 2, characterized in that, The image sequence includes a first image at a first time point and a second image at a second time point; the method further includes, during the acquisition of the image sequence: The volume difference of the current ink droplet is calculated based on the acquired first image and second image to obtain the absolute value of the volume difference of the current ink droplet; If the absolute value of the volume difference is greater than the preset volume difference threshold, the pulse voltage amplitude of the inkjet printer is adjusted according to the preset voltage amplitude increment data.

5. The method according to claim 2, characterized in that, The environmental parameter sequence includes a first environmental parameter at a third time point and a second environmental parameter at a fourth time point; the method further includes, in the process of obtaining the environmental parameter sequence: Environmental parameter difference calculations are performed based on the first environmental parameter and the second environmental parameter obtained to obtain environmental parameter difference data; wherein, the environmental parameter difference data includes at least one of the absolute value of the temperature difference of the inkjet printing chamber, the absolute value of the humidity difference of the inkjet printing chamber, and the absolute value of the difference in ink droplet evaporation intensity; If the environmental parameter difference data includes the absolute value of the temperature difference, and the absolute value of the temperature difference is greater than the preset temperature difference threshold, the temperature of the inkjet printing chamber is adjusted according to the preset temperature data. If the environmental parameter difference data includes the absolute value of the humidity difference, and the absolute value of the humidity difference is greater than the preset humidity difference threshold, the humidity of the inkjet printing chamber is adjusted according to the preset humidity data, and / or the inert gas flux is adjusted according to the first preset flux increment data; If the environmental parameter difference data includes the absolute value of the ink droplet evaporation intensity difference, and the absolute value of the ink droplet evaporation intensity difference is greater than the preset intensity difference threshold, the inert gas flux is adjusted according to the second preset flux increment data.

6. The method according to claim 1 or 5, characterized in that, Defect types include pinhole defects, crack defects, satellite drop defects, edge warping defects, and uneven stacking defects; The parameter adjustment strategy corresponding to the pinhole defect type includes reducing the temperature of the perovskite solar cell substrate according to preset temperature increment data, and / or increasing the humidity of the inkjet printing chamber according to preset humidity increment data. The parameter adjustment strategy corresponding to the crack defect type includes increasing the flow rate gradient of the inert gas according to the preset flow rate increment data, and / or adjusting the temperature of the central region or the edge region according to the preset temperature difference between the central region and the edge region of the perovskite solar cell substrate. The parameter adjustment strategy corresponding to the satellite droplet defect type includes reducing the inkjet voltage of the inkjet printer according to preset voltage increment data, and / or increasing the pulse width of the inkjet printer according to a first preset time increment data. The parameter adjustment strategy corresponding to the edge warping defect type includes reducing the ink ejection volume of the edge nozzle channel of the inkjet printer according to the preset ink volume increment data, and / or delaying the ink ejection time of the edge nozzle channel according to the second preset time increment data. The parameter adjustment strategy corresponding to the uneven stacking defect type includes grayscale printing calibration according to a preset ink droplet volume, and / or adjusting the inkjet printer's inkjet trigger time according to preset timing compensation data; wherein, the preset timing compensation data is determined based on the deviation between the actual landing position and the expected landing position of the current ink droplet.

7. The method according to claim 1, characterized in that, After obtaining the current defect type, it also includes: Based on the correspondence between defect types and defect causes, determine the defect cause corresponding to the current defect type; Matching is performed according to the current defect type within the preset parameter adjustment rules, including: The current defect type and its corresponding cause are matched in the preset parameter adjustment rules; the preset parameter adjustment rules include the mapping relationship between the defect type, its corresponding cause, and the parameter adjustment strategy.

8. The method according to claim 1, characterized in that, Each defect type has its own priority; if multiple current defect types are obtained, before matching them according to the current defect types in the preset parameter adjustment rules to determine the target parameter adjustment strategy, the method further includes: The target defect type is determined based on the priority of each current defect type; The step of matching the current defect data with preset parameter adjustment rules to determine the target parameter adjustment strategy includes: The target parameter adjustment strategy is determined by matching the target defect type with the preset parameter adjustment rules.

9. An inkjet printing control device for perovskite solar cells, characterized in that, The device includes: The feature acquisition module is used to acquire the temporal features of the current ink droplet falling onto the perovskite solar cell substrate, the spreading morphology features of the ink droplet on the current perovskite solar cell substrate, and the environmental features from the time the current ink droplet is ejected by the inkjet printer to the time it forms a film on the perovskite solar cell substrate. The defect identification module is used to identify defects based on the temporal characteristics, the spreading morphology characteristics, and the environmental characteristics to obtain the current defect type; The adjustment strategy determination module is used to match the current defect type with preset parameter adjustment rules to determine the target parameter adjustment strategy; wherein, the preset parameter adjustment rules include a mapping relationship between defect types and parameter adjustment strategies; The parameter adjustment module is used to adjust the printing process parameters based on the target parameter adjustment strategy, so as to perform subsequent inkjet printing based on the adjusted printing process parameters.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

Images