Dipping-ponding control method and system based on biomimetic vision system

Abstract

This invention relates to the field of electrode fabrication technology and discloses an immersion-lifting-space slope flow control method based on a biomimetic vision system. The method includes: generating corresponding control commands based on the initial slope flow angle and the current state of the robotic arm to ensure that the electrode plate held by the robotic arm is in the initial slope flow state, thereby causing the precursor solution to flow on the electrode plate under gravity; acquiring slope flow image data of the electrode plate and the precursor solution on its surface in real time using a camera, and analyzing the slope flow image data to determine the motion characteristic parameters of the precursor solution; and generating adjustment commands for the robotic arm based on the comparison results to dynamically adjust the angle of the electrode plate. In this invention, the immersion-lifting-space slope flow control method based on a biomimetic vision system adjusts the operation of the robotic arm to cause the precursor solution to flow under gravity to all corners of the electrode plate surface, optimizing the coating effect and ensuring uniform coverage of the electrode plate.

Description

Technical Field

[0001] This invention relates to the field of electrode fabrication technology, specifically to an impregnation-lifting-spatial slope flow control method and system based on a biomimetic vision system. Background Technology

[0002] Currently, climate change, water quality deterioration, and resource scarcity have become three increasingly serious global problems. Hybrid metal oxide electrocatalytic oxidation technology, with its outstanding advantages such as high processing efficiency, simple operation, simple equipment manufacturing, no need for chemical additives, no secondary pollution, and more thorough degradation, has attracted widespread attention and has broad market prospects in environmental governance, resource and energy conversion, storage and recycling (such as electrocatalytic oxidation of small organic molecules), analytical sensing, and biomedicine. It is a highly influential cutting-edge technology whose core lies in using electrical energy to drive catalytic oxidation reactions occurring on the electrode surface, providing green, efficient, and controllable solutions to a series of global challenges, including promoting green electricity conversion and storage, providing green purification methods, and transforming traditional synthesis processes.

[0003] Electrodes are the core components of electrocatalytic oxidation reactors. In electrocatalytic oxidation reactions, the catalyst loading and preparation efficiency directly affect the reaction efficiency, stability, preparation cost, and large-scale industrial applications. The dip-coating method is one of the most commonly used and cost-effective methods for electrode fabrication, widely applied in food, pharmaceutical, chemical, and electronics industries. The properties of the substrate and the extraction solution, their dispersibility, contact area and uniformity, dip-coating time, and post-dipping treatment are all important factors affecting the extraction effect. Currently, dip-coating equipment has limited functionality and low integration. Traditional dip-coating processes, due to gravity flow during film formation, result in low single-load loading of the precursor solution on the substrate, requiring repeated dip-coating and firing. This prevents the precursor solution from fully circulating on the substrate surface, leading to large fluctuations in film thickness uniformity and long electrode fabrication cycles. This degrades the electrochemical performance and fabrication efficiency of the membrane electrode, resulting in high fabrication costs and making it difficult to meet the demands of current large-scale industrial applications. Summary of the Invention

[0004] To address the aforementioned shortcomings, this invention discloses an immersion-lifting-spatial slope flow control method based on a biomimetic vision system, which can improve electrode coating efficiency and electrode load.

[0005] The first aspect of this invention discloses an impregnation-lifting-spatial slope flow control method based on a biomimetic vision system, comprising:

[0006] Obtain the precursor solution parameters and determine the corresponding initial slope angle based on the precursor solution parameters;

[0007] Based on the initial slope angle and the current state of the robotic arm, corresponding control commands are generated to make the electrode plate held by the robotic arm in the initial slope state, thereby causing the precursor solution to flow on the electrode plate under gravity.

[0008] During the flow of the precursor solution in the electrode plate under gravity, the slope flow image data of the electrode plate and the precursor solution on the surface of the electrode plate are collected in real time by a camera, and the slope flow image data is analyzed to determine the motion characteristic parameters of the precursor solution.

[0009] The motion characteristic parameters are compared with the set slope flow conditions, and the adjustment command of the robotic arm is generated based on the comparison result to dynamically adjust the angle of the electrode plate, so that the motion characteristic parameters of the precursor solution match the set control conditions.

[0010] As an optional implementation, in the first aspect of the present invention, the control method further includes:

[0011] The corresponding lifting speed of the electrode plate in the immersion tank is determined based on the parameters of the precursor solution.

[0012] According to the lifting speed, the robotic arm is controlled to hold the electrode plate and lift it from the immersion tank at a set angle, parallel to the top and bottom surfaces of the immersion tank directly opposite the electrode plate, and remove it from the liquid surface.

[0013] As an optional implementation, in a first aspect of the present invention, the camera includes a first camera and a second camera. The first camera is used to image the electrode plate and determine the shape and boundary coordinates of the electrode plate. The second camera is used to acquire images during the slope flow process. During the spatial slope flow process, the second camera shares the boundary coordinates provided by the first camera and tracks the movement of liquid traces within this coordinate range.

[0014] As an optional implementation, in the first aspect of the present invention, the control method further includes:

[0015] The slope flow image data is identified to determine the current slope flow velocity, slope flow direction, and historical coating information;

[0016] The area to be coated is determined based on the historical coating information, electrode plate shape, and boundary coordinates.

[0017] Based on the uncoated area, a corresponding tilt angle is determined, and the robotic arm is controlled to tilt the electrode plate so that the precursor solution flows to the area to be coated. The tilt angle includes a first tilt angle and a second tilt angle.

[0018] As an optional implementation, in a first aspect of the present invention, analyzing the slope flow image data to determine the motion characteristic parameters of the precursor solution includes:

[0019] The slope flow image data is analyzed using a semantic segmentation model to generate a liquid trace pixel mask associated with the liquid film;

[0020] The vector motion direction and speed of the liquid trace pixel mask are calculated by determining the pixel displacement of the liquid trace pixel mask between frames of the image data sequence, and the vector motion direction and speed of the liquid trace pixel mask are used as motion characteristic parameters of the precursor solution.

[0021] As an optional implementation, in a first aspect of the present invention, comparing the motion characteristic parameters with set slope flow conditions and generating adjustment commands for the robotic arm based on the comparison results to dynamically adjust the angle of the electrode plate includes:

[0022] The vector motion direction and velocity of the liquid trace are compared with the set slope flow conditions;

[0023] When the vector motion rate of the detected liquid trace is lower than the preset rate threshold, a control command is generated to increase the tilt angle of the electrode plate.

[0024] When the vector motion rate of the liquid trace is detected to be higher than the preset rate threshold, a control command is generated to reduce the tilt angle of the electrode plate.

[0025] When the liquid trace is detected to be approaching the substrate boundary coordinates, a control command is generated to change the tilt direction of the electrode plate to adjust the flow of the liquid trace in the current direction.

[0026] The motion characteristic parameters also include liquid mark height information;

[0027] The step of comparing the motion characteristic parameters with the set slope flow conditions and generating adjustment commands for the robotic arm based on the comparison results to dynamically adjust the angle of the electrode plate further includes:

[0028] When the liquid level is detected to be below a set height threshold, the slope flow is stopped.

[0029] As an optional implementation, in the first aspect of the present invention, the control method further includes:

[0030] After the coating is applied, an image of the coated electrode plate is captured using a camera;

[0031] The electrode plate image is subjected to quality inspection to determine whether there are coating defects. If coating defects are detected in the electrode plate image, the system issues an alarm.

[0032] Record all coating operation data, including solution flow rate, lifting speed, slope angle, coating uniformity, and precursor solution parameters.

[0033] A second aspect of this invention discloses an impregnation-lifting-spatial slope flow control system based on a biomimetic vision system, comprising:

[0034] Data acquisition module: used to acquire precursor solution parameters and determine the corresponding initial slope angle based on the precursor solution parameters;

[0035] Tilting module: used to generate corresponding control commands based on the initial slope angle and the current state of the robotic arm so that the electrode plate held by the robotic arm is in the initial slope state, thereby causing the precursor solution to flow on the electrode plate under gravity.

[0036] Image acquisition module: used to acquire in real time the slope flow image data of the precursor solution on the electrode plate and the surface of the electrode plate through a camera during the gravity-driven flow of the precursor solution on the electrode plate, and to analyze the slope flow image data to determine the motion characteristic parameters of the precursor solution.

[0037] The comparison and adjustment learning module is used to compare the motion characteristic parameters with the set slope flow conditions, and generate adjustment commands for the robotic arm based on the comparison results to dynamically adjust the angle of the electrode plate, thereby making the motion characteristic parameters of the precursor solution match the set control conditions.

[0038] A third aspect of the present invention discloses an electronic device, comprising: a memory storing executable program code; a processor coupled to the memory; the processor calling the executable program code stored in the memory to execute the immersion lift-space slope flow control method based on a bionic vision system disclosed in the first aspect of the present invention.

[0039] The fourth aspect of this invention discloses a computer-readable storage medium storing a computer program, wherein the computer program causes a computer to execute the immersion lift-space slope flow control method based on a biomimetic vision system disclosed in the first aspect of this invention.

[0040] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0041] In this embodiment of the invention, the immersion-lifting-spatial slope flow control method based on a biomimetic vision system adjusts the operation of the robotic arm to allow the precursor solution to flow towards all corners of the electrode plate surface under gravity, optimizing the coating effect and ensuring that the solution uniformly covers the electrode plate without overflowing. During the coating process, the vision recognition system continuously acquires real-time image data and compares it with the set coating standards to correct the uniformity and flow state of the precursor solution, thereby improving the overall control efficiency. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a schematic flowchart of the immersion lifting-space slope flow control method based on a biomimetic vision system disclosed in an embodiment of the present invention;

[0044] Figure 2 This is a structural diagram of the control system disclosed in an embodiment of the present invention;

[0045] Figure 3 This is a schematic diagram of the slope flow control adjustment process disclosed in an embodiment of the present invention;

[0046] Figure 4 This is a schematic diagram of the load comparison results disclosed in an embodiment of the present invention;

[0047] Figure 5 This is a schematic diagram showing the loading of Ti / IrO2-RuO2 electrodes prepared by different impregnation and pulling methods disclosed in the embodiments of the present invention;

[0048] Figure 6 This is another schematic diagram showing the loading of Ti / IrO2-RuO2 electrodes prepared by different impregnation and pulling methods disclosed in the embodiments of the present invention;

[0049] Figure 7 This is an illustration of the effect of treating dyeing and printing wastewater using a self-made electrode and a commercial electrode disclosed in an embodiment of the present invention.

[0050] Figure 8 This is another effect diagram of the treatment of dyeing and printing wastewater using the self-made electrode and commercial electrode disclosed in the embodiments of the present invention;

[0051] Figure 9 This is a schematic diagram of the structure of an immersion lifting-space slope flow control system based on a biomimetic vision system provided in an embodiment of the present invention;

[0052] Figure 10This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] It should be noted that the terms first, second, third, fourth, etc., in the specification and claims of this invention are used to distinguish different objects, not to describe a specific order. The terms used in the embodiments of this invention include and have, and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products, or devices.

[0055] Example 1

[0056] Please see Figure 1 , Figure 1 This is a flowchart illustrating the immersion-lifting-spatial slope flow control method based on a biomimetic vision system disclosed in this invention. The execution entity of the method described in this embodiment is an execution entity composed of software and / or hardware. This execution entity can receive relevant information via wired or / or wireless means and can send certain instructions. It may also have certain processing and storage functions. This execution entity can control multiple devices, such as remote physical servers or cloud servers and related software, or local hosts or servers and related software that perform related operations on devices located in a certain place. In some scenarios, multiple storage devices can also be controlled; these storage devices can be placed in the same location as the devices or in different locations. Figure 1 As shown, the impregnation-lifting-space slope flow control method based on a biomimetic vision system includes the following steps:

[0057] S101: Obtain the precursor solution parameters and determine the corresponding initial slope angle based on the precursor solution parameters;

[0058] S102: Generate corresponding control commands based on the initial slope angle and the current state of the robotic arm so that the electrode plate held by the robotic arm is in the initial slope state, thereby causing the precursor solution to flow on the electrode plate under gravity.

[0059] S103: During the gravity-driven flow of the precursor solution on the electrode plate, the slope flow image data of the electrode plate and the precursor solution on the surface of the electrode plate are collected in real time by a camera, and the slope flow image data is analyzed to determine the motion characteristic parameters of the precursor solution.

[0060] S104: The motion characteristic parameters are compared with the set slope flow conditions, and an adjustment command for the robotic arm is generated based on the comparison result to dynamically adjust the angle of the electrode plate, thereby making the motion characteristic parameters of the precursor solution match the set control conditions.

[0061] Specifically, spatial slope flow achieves uniform liquid film distribution and thickness control by adjusting the spatial orientation of the electrode plate (such as tilting or rotating) and utilizing the component of gravity to drive the precursor solution to flow in a directional manner. This is combined with the volatilization kinetics of the organic solvent. When the electrode plate is tilted at an angle θ, the solution flows along the tilt direction under the influence of gravity, overcoming surface tension and viscous resistance to form a stable laminar flow. The component of gravity drives the solution to flow from high-thickness areas (accumulation areas) to low-thickness areas (uncovered areas), while shear force breaks up local agglomerations, achieving thickness uniformity. The core problem that the spatial slope flow of this invention aims to solve is increasing the load capacity and electrode surface uniformity. It has significant advantages over existing solutions. Specifically, the initial slope angle is related to the properties of the precursor. It aims to make the precursor solution remaining on the electrode plate flow as evenly as possible on the electrode plate. The clamp holds the electrode plate with a perforated handle, so the angle between this angle and the horizontal plane is an obtuse angle greater than 90 degrees. The initial slope angle can be set to any value between 120 degrees and 150 degrees. The above angle is the angle between the electrode plate and the horizontal plane.

[0062] The solution of this invention determines the initial slope angle by acquiring precursor solution parameters and generates control commands accordingly to position the electrode plate held by the robotic arm in the initial slope state. This ensures that the precursor solution begins to flow on the electrode plate in the expected manner, laying the foundation for subsequent precise control. This precise initial state setting improves the accuracy of slope control. Because different solutions will exhibit different flow velocities under the same electrode plate and the same tilt angle, it is necessary to determine the initial slope state by considering the specific conditions of the solution and the required velocity. Here, the initial slope state refers to the tilt angle.

[0063] During the flow of the precursor solution, a camera is used to collect real-time slope flow image data, which is then analyzed to determine motion characteristic parameters. These parameters are compared with the set slope flow conditions, and adjustment commands for the robotic arm are generated in a timely manner to dynamically adjust the angle of the electrode plate. This real-time monitoring and dynamic adjustment mechanism allows the system to react promptly to the actual flow of the precursor solution, adapting to different working conditions and changes. This ensures that the motion characteristic parameters of the precursor solution match the set control conditions, improving the accuracy and stability of slope flow control and ultimately achieving uniform coating on the electrode surface. Subsequent dynamic tilt adjustments typically involve fine-tuning the tilt angle left or right along the long side, with this adjustment angle generally not exceeding 15 degrees.

[0064] The method of this invention, from acquiring solution parameters, determining the initial angle, generating control commands to real-time monitoring and dynamic adjustment, achieves automated control without human intervention. This not only improves work efficiency and reduces errors and uncertainties caused by human factors, but also enables the system to operate stably for extended periods without human intervention. It is suitable for various applications requiring precise slope flow control, such as material preparation and surface treatment.

[0065] In specific processing, slope control can be achieved by detecting the height of the liquid mark. When the liquid mark height is below 0.1 micrometers, the electrode sheet can be kept horizontal, ultimately achieving uniform coating on the surface. Specific height detection methods include laser triangulation, structured light projection to analyze grating deformation for ranging, or a combination of monocular vision and a priori models for height recognition. Even binocular stereo vision can be used to analyze the height information at the corresponding location. In this embodiment, a single camera and a priori model are used for height recognition. The system can establish a mathematical relationship between the apparent width of the liquid mark in the image and its actual height through calibration. The camera captures the liquid mark. The liquid mark appears as a bright band in the image. By measuring the pixel width of this bright band and combining it with the known liquid mark shape model and camera parameters, the approximate height of its center can be calculated.

[0066] More preferably, the control method further includes:

[0067] The corresponding lifting speed of the electrode plate in the immersion tank is determined based on the parameters of the precursor solution.

[0068] According to the lifting speed, the robotic arm is controlled to hold the electrode plate and lift it from the immersion tank at a set angle, parallel to the top and bottom surfaces of the immersion tank directly opposite the electrode plate, and remove it from the liquid surface.

[0069] In this embodiment of the invention, the set angle is any value between 30 and 60 degrees. This set angle refers to the angle between the electrode plate and the horizontal plane. Pulling the electrode plate parallel to its plane at this set angle, i.e., pulling it obliquely upwards, can significantly increase the load on the final electrode surface. Specifically, the impregnation tank can be a flat rectangular impregnation tank or other shapes, its main purpose being to hold the precursor solution. This flat rectangular impregnation tank also has an angle with the horizontal plane, meaning it is tilted at a certain angle, such as 30 to 60 degrees. The electrode plate is pulled parallel to its plane, and during the pull, the angle is consistent with the tilt angle of the impregnation tank, meaning the electrode plate is parallel to both the front and back surfaces of the flat rectangular impregnation tank, facilitating stable pulling. In specific implementation, a camera can be used to identify the angle of the corresponding impregnation tank, and then the robotic arm can be adjusted to grip the electrode plate and insert it deeper into the impregnation tank, ensuring it is parallel to the normal direction of the impregnation tank.

[0070] The angle adjustment during the impregnation process is related to the precursor solution level, ideally ensuring complete immersion of the electrode plate. If complete immersion is not achieved when tilted at 30 degrees, an alarm must be triggered to add precursor solution. The reason for designing the impregnation tank as a flat, rectangular shape is that precursor solutions are generally expensive. A flat tank with a higher precursor solution level easily immerses the entire electrode plate. If additional solution is needed, the amount added is also smaller. This improves the utilization rate of the precursor solution and reduces waste when electrode preparation is complete.

[0071] The parameters of the precursor solution in this embodiment of the invention (such as viscosity, concentration, surface tension, etc.) directly determine its adhesion ability and initial distribution state on the electrode plate surface.

[0072] By linking the lifting speed to the precursor solution parameters, it can be ensured that the amount and uniformity of the precursor solution adhering to the surface of the electrode plate after it is lifted from the impregnation tank are perfectly suited to the subsequent slope flow requirements. This avoids problems such as insufficient solution volume during slope flow due to inadequate impregnation or uneven solution distribution due to excessively rapid lifting, thus laying a better initial foundation for stable start-up and precise control of the subsequent slope flow. Under certain conditions, the impregnation time can also affect the amount of surface adhesion.

[0073] If the lifting speed is not matched with the solution parameters, it may cause initial disturbances such as local accumulation of solution droplets or surface bubbles when the electrode plate is lifted. For example, after a short period of immersion, rapid lifting can easily cause solution droplets to form on the edge of the electrode plate. During subsequent sloping flow, these droplets will disrupt the continuity of the flow and cause sudden changes in local flow velocity.

[0074] The solution of this invention can minimize such initial disturbances by precisely controlling the impregnation and lifting processes, so that the precursor solution forms a uniform and continuous initial liquid film on the surface of the electrode plate. The subsequent gravity-driven slope flow process is more stable, and the fluctuation of motion characteristic parameters (such as flow rate and liquid film thickness) is smaller, further reducing the difficulty of dynamic adjustment.

[0075] In material preparation scenarios (such as sol-gel method for thin film preparation), traditional dip-lift operations often rely on human experience to judge time and speed. Differences in operation between different operators can easily lead to batch-to-batch product quality fluctuations. For example, the initial adhesion amount of the same solution can vary significantly due to different lifting speeds.

[0076] The solution of this invention incorporates the determination and control of immersion time and lifting speed into an automated process. It can generate precise instructions based on solution parameters without manual intervention, and achieve standardization of the entire process from solution parameter input, automatic determination of immersion / lifting parameters, automatic execution, and automatic entry into slope flow control. This completely avoids human error, ensures consistency of results for different batches and different operators, and improves the stability of product quality.

[0077] More preferably, after the robotic arm controls the lifting speed to lift the electrode plate from the immersion tank parallel to the upper and lower surfaces of the immersion tank directly opposite the electrode plate, the method further includes:

[0078] The lifting image data during the lifting process is acquired in real time by a camera, and the lifting image data is identified by an edge recognition algorithm to determine the corresponding liquid film image;

[0079] Feature extraction is performed on the liquid film image to determine the corresponding liquid film edge features, and uniformity analysis is performed on the liquid film edge features to determine the distribution state information of the corresponding liquid film;

[0080] Analyze consecutive frames of lifting image data to determine the velocity information of the liquid film leading edge;

[0081] The lifting speed is adjusted in real time by comparing the distribution status information and the velocity information of the liquid film front with the set lifting conditions.

[0082] Specifically, by capturing real-time images of the liquid film being pulled up using a camera and identifying the edges of the liquid film, the system can instantly detect uneven distribution of the liquid film (such as localized excessive thickness or thickness differences). For example, when the edge recognition algorithm detects that the thickness on the left side of the liquid film exceeds a set threshold, the system can reduce the pulling speed or fine-tune the angle of the electrode plate to allow the solution to redistribute under gravity and eliminate the deviation. This closed-loop control mechanism overcomes the limitations of simply pulling up according to preset parameters, and can dynamically correct liquid film defects caused by fluctuations in solution parameters (such as localized viscosity changes) and minor vibrations of the equipment, ensuring that the initial liquid film state formed during the pulling stage is closer to the ideal value.

[0083] The velocity of the liquid film front directly affects the uniformity of the final film layer. Excessive velocity can lead to overstretching, resulting in film breakage or pinholes, while insufficient velocity may cause excessive solution buildup, forming flow marks. Real-time acquisition of the front velocity through continuous frame image analysis allows comparison with a set threshold (e.g., an optimal velocity range calculated based on solution viscosity). If the velocity exceeds the upper limit, the pulling speed is immediately reduced to minimize solution stretching; if it falls below the lower limit, the speed is appropriately increased to prevent localized buildup. This dynamic adjustment keeps fluctuations in the liquid film front velocity within a minimal range, reducing the correction pressure in subsequent slope flow stages from the outset.

[0084] In practice, the precursor solution may deviate from its initial parameters due to changes in ambient temperature, such as viscosity changes caused by room temperature fluctuations, or batch differences. For example, if the actual viscosity of a batch of solution is higher than the preset value, maintaining the original lifting speed will result in an excessively thick liquid film. This invention, by monitoring the liquid film distribution and leading edge speed in real time, can automatically detect this deviation and adjust the speed, such as reducing the lifting speed to decrease the amount of adhesion, thus maintaining liquid film quality without recalibrating solution parameters.

[0085] The liquid film formed during the lifting stage is the starting point for slope flow control, and its uniformity directly determines the complexity of slope flow adjustment. By optimizing the lifting speed in real time, it can be ensured that the liquid film has a highly uniform thickness and continuous edge morphology before entering the slope flow stage, reducing local velocity abrupt changes caused by differences in the initial state during slope flow. For example, if the excessively thick area on the left side of the liquid film has been eliminated through adjustments during the lifting stage, the system only needs to fine-tune the body angle to control the flow velocity during slope flow, without the need for complex adjustments to address local defects, significantly improving the efficiency and stability of the entire process control.

[0086] The real-time acquired liquid film image data, edge features, and velocity information in this invention can be recorded and archived, providing data support for subsequent process optimization. For example, by analyzing the correlation between liquid film uniformity and the adjustment range of lifting speed in multiple batches of data, the speed adjustment algorithm can be further optimized; or a more accurate solution parameter-lifting speed mapping model can be established based on the liquid film performance of different solutions under the same parameters. This data-driven optimization capability enables the control method to continuously adapt to new material systems and process requirements, extending the technology's life cycle.

[0087] More preferably, the camera includes a first camera and a second camera. The first camera is used to capture images during the lifting process and is also used to image the electrode plate and determine the shape and boundary coordinates of the electrode plate. The second camera is used to capture images during the slope flow process. During the spatial slope flow process, the second camera shares the boundary coordinates provided by the first camera and tracks the movement of the liquid trace within this coordinate range.

[0088] Specifically, such as Figure 2 As shown, the rapid loading system for impregnated lifting-space slope flow films based on a biomimetic vision system consists of: a biomimetic vision system (high-definition video stream data acquisition camera group, image information extraction, recognition, processing and feedback unit), a motion execution system (multi-axis robotic arm, motion controller), a precursor solution circulation system (temperature control tank, metering pump, control system panel, liquid storage tank), a fixed platform system and a control system.

[0089] Bionic Vision System: The first vision camera photographs the substrate to determine its shape and boundary coordinates. After impregnation and lifting are completed, it extracts, identifies, and confirms the precursor solution traces and vector movement direction and speed on the substrate surface. Then, it processes the image information according to a set threshold and feeds the information back to the terminal system to direct the robotic arm. The second vision camera shares its substrate edge coordinates, trace coordinates, vector movement direction, and speed data with the first vision camera. It tracks and directs the robotic arm to move the traces within the substrate edge coordinate range, ultimately achieving complete and uniform coating of the precursor solution on the substrate surface. The first vision camera is the first camera, and the second vision camera is the second camera.

[0090] High-definition video stream data acquisition camera group: Real-time acquisition of high-definition video stream data of the electrode plate and the precursor solution on the electrode plate surface. Image processing unit: Identifies liquid mark edges, flow velocity, and thickness distribution based on OpenCV or deep learning algorithms.

[0091] Motion execution system: Grabs the substrate and controls its movement.

[0092] Multi-axis robotic arm: Holds the electrode substrate and controls its movement to achieve impregnation, lifting, and tilting actions. Motion controller: Dynamically adjusts the lifting speed (0.1–10 mm / s) and angle (±5° deflection) according to feedback from the vision system and instructions from the central control system.

[0093] Precursor solution circulation system (impregnation tank, metering pump, storage tank):

[0094] Impregnation tank: Maintains the stability of precursor solution concentration and viscosity during delivery and recovery;

[0095] Metering pump: Replenishes precursor solution and maintains liquid level;

[0096] Storage tank: for storing precursor solutions.

[0097] Fixed platform system: physical support platform for each subsystem component; Control system: central control unit for the entire process load, equipped with a terminal panel, which can receive information, issue instructions, control execution, and perform dynamic monitoring, and can input the precursor solution type, electrode substrate shape and size, initial impregnation and lifting speed, upper and lower limits of spatial slope flow time, and maximum movement distance.

[0098] Specifically, such as Figure 2 As shown, A is the precursor solution storage tank; B is the multi-way valve B1 and the micro-pump metering pump B2; B3 is the cleaning waste liquid discharge; C is the fixed platform; C1 is the platform baffle; D is the immersion tank; S1 is the high-definition camera 1; S2 is the high-definition camera 2; S3 is the cold light source; X is the multi-axis robotic arm; E is the motion controller; J is the fixture; DP is the control system (bionic vision system); the dashed line indicates the connection between the sensor and the terminal display control system.

[0099] The immersion-lift-space slope flow method involves filling a specially designed square glass container with a precursor mixture solution. During immersion, the normal direction of the square container is at an angle of 30-60° to the horizontal plane, and the titanium substrate is completely immersed in the precursor solution. After a certain period of time, the substrate is lifted at a speed U along the normal direction, utilizing the interaction between the viscosity, surface tension, and gravity of the precursor solution. Once the edge of the wedge-shaped liquid mark at the bottom leaves the liquid surface, the substrate is quickly tilted towards the handle to prevent backflow of the precursor solution. The wedge-shaped liquid mark will form a slope flow (primary slope flow) towards the handle. By tracking the flow of the precursor solution liquid mark on the surface of the substrate, the substrate is rotated around its center point, causing the liquid mark to undergo multiple slope flow loads on the substrate under the action of gravity. Simultaneously, due to the slope flow, the surface area of ​​the solvent in the precursor solution increases, and the solvent vapor diffuses outward. During the drying kinetics, the load is accelerated to form a gel-like film. The gel-like film coating material is then dried by forced drying or baking.

[0100] The present invention employs a dip-coating process to increase the precursor solution loading during electrode coating; panel-controlled robotic arm operation avoids inconsistencies in each coating operation compared to manual operation; based on the original vertical dip-lifting method, universal joints and additional control programs are used to adjust control parameters according to relevant theoretical calculations, incorporating gravity and precursor solution surface tension to achieve uniform distribution of the precursor solution on the electrode surface and recovery of excess precursor solution. Based on experimental data or by using AI vision to identify and control advection, integrated technology enhances the system's intelligence and further optimizes process efficiency.

[0101] In this embodiment of the invention, the functional separation of the first camera and the second camera allows for specialized optimization to meet the acquisition needs at different stages, avoiding compromises in acquisition accuracy caused by a single camera handling multiple scenarios.

[0102] More accurate acquisition during the lifting stage: The first camera focuses on image acquisition and electrode plate shape / boundary coordinate recognition during the lifting process. There is no need to switch to parameters for the slope flow scene (such as focal length and exposure). By fixing the focus on the electrode plate area and optimizing the edge recognition algorithm parameters (such as adjusting the grayscale threshold and edge detection operator), the recognition error of the electrode plate boundary coordinates is kept within a very small range. At the same time, it clearly captures the subtle morphological changes of the liquid film during lifting (such as local thickness difference and edge curling).

[0103] More focused slope flow tracking: The second camera does not need to undertake the task of electrode plate boundary recognition. It can lock the monitoring area in advance based on the coordinate data of the first camera (tracking only the movement of liquid marks on the surface of the electrode plate, rather than full-screen scanning), reducing the impact of background interference (such as the edge of the immersion tank and changes in ambient light) on the extraction of liquid mark features. This makes the analysis of parameters such as the contour and flow velocity of liquid marks in the slope flow image more accurate, especially suitable for slope flow control scenarios of high viscosity solutions or narrow electrode plates.

[0104] Specifically, after the lifting phase, when the electrode plate enters the initial state of the slope flow, the second camera can immediately lock the surface area of ​​the electrode plate based on shared coordinates, avoiding tracking delays caused by re-identifying boundaries. During the slope flow process, the trajectory and position parameters of the liquid traces are calculated based on unified electrode plate coordinates, ensuring consistent data throughout the entire process from the initial state of the liquid film after lifting to the movement of the liquid traces in the slope flow, reducing errors in flow velocity calculation and deviations in judging the uniformity of the liquid film caused by inconsistent coordinates.

[0105] If a single-camera solution requires frequent switching between two scenarios—identifying the electrode plate boundary during the lifting stage and tracking liquid marks during the slope flow stage—the algorithm must first perform edge detection and coordinate calibration, and then perform liquid mark feature extraction. Furthermore, each switch requires readjusting image preprocessing parameters (such as filtering methods and ROI region selection), which not only consumes more computing power but also leads to data processing delays and affects the timeliness of dynamic adjustments.

[0106] More preferably, the control method further includes:

[0107] The slope flow image data is identified to determine the current slope flow velocity, slope flow direction, and historical coating information;

[0108] The area to be coated is determined based on the historical coating information, electrode plate shape, and boundary coordinates.

[0109] Based on the uncoated area, a corresponding tilt angle is determined, and the robotic arm is controlled to tilt the electrode plate so that the precursor solution flows to the area to be coated. The tilt angle includes a first tilt angle and a second tilt angle.

[0110] In practical implementation, by recognizing the slope flow image data, the system can grasp two key pieces of information in real time: first, the current slope flow state (velocity, direction) to determine the current flow trend of the solution; second, historical coating information, which, through continuous frame image overlay analysis, clearly marks the areas on the electrode plate surface that have been covered by the solution (such as the middle and right sides) and the uncovered areas to be coated.

[0111] By combining the previously accurately acquired electrode plate shape and boundary coordinates, the system can precisely locate the specific position of the area to be coated, avoiding deviations caused by manual prediction. The visual recognition system continuously acquires real-time image data and compares it with the set coating standards to correct the uniformity and flow state of the precursor solution. If uneven liquid line thickness or solution overflow at the edge of the electrode plate is detected, the system will automatically correct it, stopping the spatial slope flow in that direction and immediately adjusting the direction to make the precursor solution flow uniformly again, until the visual recognition shows that the liquid line thickness reaches the set standard.

[0112] Even if a single flow doesn't completely reach the edge, as long as the path of the next flow overlaps sufficiently with the previous one, multiple cross-flows can fill in any thin areas, ultimately achieving overall uniformity. The vision system records the area covered by the liquid trails and intelligently plans the direction of the next flow to ensure no blind spots.

[0113] The solution in this invention achieves multi-dimensional attitude control of the electrode plate in space by combining and adjusting a first tilt angle (e.g., tilting along the width of the electrode plate to control the left-right flow of the solution) and a second tilt angle (e.g., tilting along the length of the electrode plate to control the forward-backward flow of the solution). When the area to be coated is the upper right corner of the electrode plate, the system can simultaneously increase the first tilt angle (tilting to the right) and the second tilt angle (tilting backward) to guide the solution to flow precisely to that area. If the area to be coated is a recessed area in the middle of the electrode plate, a local gravity gradient can be formed by finely adjusting the difference between the two angles, pushing the solution to fill the recess. This dual-dimensional control overcomes the limitations of a single angle, ensuring that any area on the electrode plate surface to be coated can be covered by the solution.

[0114] Since the coating thickness of the precursor solution directly affects the performance of the final product, repeated coverage of the same area by the solution can easily lead to localized over-thickness. The solution in this invention, by recording historical coating information in real time (such as the coating time and solution adhesion amount in a certain area), can automatically exclude areas that are already sufficiently covered when determining the area to be coated. For example, when the left side of the electrode plate has undergone two effective coatings and the thickness reaches a set threshold, the system will not guide the solution flow to that area, but will instead prioritize directing the solution to uncoated or insufficiently coated areas.

[0115] Meanwhile, by combining slope velocity data (if the solution flows too fast in a certain area, it is easy to cause the coating to be too thin), the system can adjust the rate of change of the tilt angle (such as slowly increasing the angle and reducing the solution flow rate to increase the local adhesion) to achieve precise recoating of the area to be coated and dynamic planning to avoid repeated coverage of the already coated area. This makes the final film thickness difference controlled within a very small range, which is significantly better than the uniformity performance of traditional single-angle slope flow.

[0116] More preferably, the analysis of the slope flow image data to determine the kinematic characteristic parameters of the precursor solution includes:

[0117] The slope flow image data is analyzed using a semantic segmentation model to generate a liquid trace pixel mask associated with the liquid film;

[0118] The vector motion direction and speed of the liquid trace pixel mask are calculated by determining the pixel displacement of the liquid trace pixel mask between frames of the image data sequence, and the vector motion direction and speed of the liquid trace pixel mask are used as motion characteristic parameters of the precursor solution.

[0119] Image analysis methods (such as threshold segmentation and edge detection) are easily affected by environmental interference (such as electrode plate reflection, uneven lighting, and background noise), which leads to blurred distinction between liquid film and non-liquid film areas (for example, misjudging scratches on the electrode plate surface as liquid marks edges).

[0120] In this embodiment of the invention, the slope flow image is analyzed by a semantic segmentation model, which can accurately generate liquid trace pixel masks based on pixel-level classification capabilities. By learning a large amount of labeled data (including liquid film images with different lighting, different solution colors, and different electrode plate materials), the model can automatically identify the unique features of the liquid film, such as texture, color gradient, and reflective properties. Even under complex backgrounds (such as patterned electrode plates or semi-transparent solutions), the liquid film area can be completely separated from the image, providing a reliable liquid trace benchmark for subsequent motion parameter calculations.

[0121] Different precursor solutions (such as transparent sols, dark suspensions, and high-viscosity pastes) exhibit vastly different visual characteristics in images: transparent solutions are easily confused with the surface of the electrode plate, dark solutions are prone to forming false edges due to reflection, and the boundaries of liquid traces in high-viscosity solutions are blurred.

[0122] The semantic segmentation model exhibits strong feature generalization capabilities. By incorporating diverse solution samples during the training phase, it can automatically adapt to the visual characteristics of different solutions: for example, for transparent solutions, the model focuses on capturing the difference in refractive index between the solution and the electrode plate; for high-viscosity solutions, the model focuses on the gradient changes at the edge of the liquid mark. This adaptability allows the same analysis algorithm to cover a variety of solution types, from low to high viscosity and from transparent to opaque, without the need to redevelop recognition logic for specific solutions, significantly improving the process compatibility of the control method.

[0123] The adjustment of the electrode plate angle by the robotic arm depends on the accuracy of the liquid film motion characteristic parameters. If the flow rate calculation is too high, it may lead to over-adjustment of the angle (the solution flows to the edge too quickly); if the flow direction judgment is inaccurate, the solution may not flow accurately to the area to be coated.

[0124] The motion parameters (vector direction, velocity) obtained through semantic segmentation and sequence frame displacement analysis exhibit high fidelity: on the one hand, the liquid mark pixel mask ensures the accuracy of the analysis object (only for the liquid film area); on the other hand, the overall displacement calculation avoids local noise interference. Adjustment commands based on these parameters can more accurately match the actual state of the liquid film, improving the matching degree between the liquid film motion characteristic parameters and the set control conditions, ultimately significantly improving the stability of slope flow control and coating quality.

[0125] Specifically, semantic segmentation is performed on the acquired image sequence to accurately separate the liquid trace region from the electrode plate background; optical flow analysis is performed on the segmented liquid trace region to calculate the pixel-level motion vector field; and the motion vector field is converted into the motion direction and speed in the physical world.

[0126] The semantic segmentation model in this invention automatically learns features through a data-driven approach, adapting to changing operating conditions without manual intervention: when the solution type changes, the model can directly generate an accurate liquid mark mask based on the visual features of the new solution; when lighting conditions change, the model's robustness to light fluctuations ensures stable recognition results. This end-to-end analysis process significantly reduces manual calibration costs, enabling the system to achieve fully automated motion feature parameter recognition without human intervention, further enhancing the overall intelligence level of the control method.

[0127] Specifically, the movement of the liquid trail between consecutive frames is analyzed using optical flow or feature point tracking algorithms to calculate its velocity vector (including magnitude and direction). Then, the velocity and direction distribution of each pixel are determined, and tilt control is performed in conjunction with the state of the specific electrode plate.

[0128] More preferably, such as Figure 3 As shown, the step of comparing the motion characteristic parameters with the set slope flow conditions and generating adjustment commands for the robotic arm based on the comparison results to dynamically adjust the angle of the electrode plate includes:

[0129] S1041: Compare the vector motion direction and speed of the liquid trace with the set slope flow conditions;

[0130] S1042: When the vector motion rate of the detected liquid trace is lower than the preset rate threshold, a control command is generated to increase the tilt angle of the electrode plate.

[0131] S1043: When the vector motion rate of the liquid trace is detected to be higher than the preset rate threshold, a control command is generated to reduce the tilt angle of the electrode plate.

[0132] S1044: When a liquid mark is detected approaching the substrate boundary coordinates, a control command is generated to change the tilt direction of the electrode plate to adjust the flow of the liquid mark in the current direction.

[0133] The solution of this invention directly correlates the velocity of the liquid trace vector with a preset threshold, establishing a clear correspondence rule between velocity deviation and angle adjustment, thus completely solving the flow velocity fluctuation problem caused by traditional fuzzy adjustment.

[0134] Low-speed compensation avoids liquid film stagnation. When the flow rate is lower than the threshold (such as when the flow rate of a high-viscosity solution is too slow due to insufficient gravity, or even local stagnation), the system immediately generates an instruction to increase the tilt angle (such as from 5° to 8°). By increasing the component of gravity, the liquid film flow is accelerated, ensuring that the solution can cover the area to be coated at the expected speed, and avoiding problems such as local accumulation and excessive film thickness caused by slow flow rate.

[0135] High-speed suppression prevents liquid film loss. When the rate exceeds the threshold (e.g., low-viscosity solutions flow too fast due to excessive tilt angle), the system quickly generates an instruction to reduce the tilt angle (e.g., from 10° to 6°), reducing the gravity driving effect to slow down the flow rate and prevent the solution from flowing too fast to the edge of the electrode plate due to inertia and dripping, thus reducing material waste and coating defects (e.g., no film at the edge, local thin leakage).

[0136] By combining the precisely obtained electrode plate boundary coordinates from the previous stage, the tilt direction is actively corrected when the liquid mark approaches the boundary, overcoming the limitations of traditional single-direction tilting and simultaneously solving the two core pain points of edge overflow and localized missed coating:

[0137] To prevent solution loss and equipment contamination, when the liquid trace flows to near the electrode plate boundary (e.g., 2-3 mm from the edge), the system no longer continues to adjust the angle along the original direction. Instead, it immediately generates a command to change the tilt direction (e.g., instead of tilting to the right along the X-axis, it tilts forward along the Y-axis). The reverse gravity component intercepts the liquid trace flowing towards the boundary, preventing the solution from overflowing the electrode plate and contaminating the robotic arm or immersion tank, thus reducing cleaning costs and material loss.

[0138] Full coverage, guiding the liquid trail to uncoated areas: Changing the tilt direction does not simply stop the flow, but rather, in conjunction with the previously located area to be coated (such as the uncovered area on the other side of the electrode plate), the liquid trail is guided to flow towards the uncoated area by adjusting the tilt direction (e.g., from tilting to the right to tilting to the left), ensuring no dead corners are covered on the electrode plate surface. For example, when the liquid trail approaches the right boundary, adjusting to tilting to the left can push the solution to the uncovered area on the left, achieving full-area coating without additional adjustment steps.

[0139] The above method does not require complex algorithms for multi-parameter coupled analysis (such as simultaneously calculating the combined effects of rate, direction, and coverage). It only needs to execute simple logic in the order of rate comparison and boundary judgment, which shortens the data processing time, ensures the immediacy of the robotic arm's adjustment instructions, and avoids the liquid trace from crossing the boundary or the flow rate deviation from expanding due to decision delay.

[0140] Improved debugging and maintenance efficiency; standardized rules make it easier for engineers to quickly locate problems (e.g., if the flow rate is consistently low, the angle increase command can be checked directly to see if it is being sent correctly), without the need to disassemble complex algorithm models, reducing the difficulty of system maintenance; at the same time, preset parameters can be flexibly adjusted according to different solution characteristics (e.g., high viscosity solutions require higher rate thresholds), making it more adaptable.

[0141] More preferably, the motion characteristic parameters also include liquid mark height information;

[0142] The step of comparing the motion characteristic parameters with the set slope flow conditions and generating adjustment commands for the robotic arm based on the comparison results to dynamically adjust the angle of the electrode plate further includes:

[0143] When the liquid level is detected to be below a set height threshold, the slope flow is stopped.

[0144] More preferably, the control method further includes:

[0145] After the coating is applied, an image of the coated electrode plate is captured using a camera;

[0146] The electrode plate image is subjected to quality inspection to determine whether there are coating defects. If coating defects are detected in the electrode plate image, the system issues an alarm.

[0147] Record all coating operation data, including solution flow rate, lifting speed, slope angle, coating uniformity, and precursor solution parameters.

[0148] Traditional coating processes require manual inspection of electrode plates after coating (such as visual inspection or offline testing equipment to determine defects). This results in problems such as delayed detection, high rate of missed detection, and defective products flowing into the next process. For example, defects such as local missed coating or film cracking may only be exposed in subsequent drying and sintering stages, leading to waste of raw materials and time.

[0149] The solution of this invention uses a camera to acquire images of the coating in real time and performs automated quality inspection, and has the following advantages:

[0150] Image analysis algorithms (such as comparing pixel differences between ideal coating images and actual images, and detecting deviations in film thickness uniformity) can automatically determine whether coating defects exist (such as missed areas, local over-thickness / under-thinness, and liquid residue). The recognition accuracy is higher than that of manual methods, avoiding missed detections caused by subjective human judgment.

[0151] It can provide real-time alerts and block the flow of defective products. Once a defect is detected, the system can immediately trigger an alarm (such as an audible and visual alarm or a system pop-up notification). At the same time, it can link with subsequent processes (such as pausing the robotic arm transfer and marking defective product numbers) to prevent defective products from entering subsequent stages such as drying and cutting, thereby significantly reducing subsequent rework costs and raw material losses.

[0152] In industrial production, data traceability is a core prerequisite for solving quality problems and optimizing process parameters. Traditional processes often rely on manual recording of key parameters (such as lifting speed and slope angle), which is prone to problems such as missing records, data errors, and inability to link batches, making it difficult to pinpoint the cause when subsequent quality fluctuations occur.

[0153] The solution of this invention automatically records comprehensive data such as solution flow rate, lifting speed, slope angle, coating uniformity, and precursor solution parameters, and archives this data in association with electrode plate batch number, coating time, and equipment number, thereby achieving the following:

[0154] Precise traceability allows for the rapid retrieval of complete operational data for a batch when coating defects are found (e.g., checking for anomalies such as delayed slope angle adjustment or excessive lifting speed). This helps pinpoint the root cause of the problem (e.g., equipment parameter drift or solution batch anomalies) and prevent similar issues from recurring.

[0155] The continuous accumulation of coating operation data can form a parameter-quality correlation database, providing a data-driven basis for subsequent process optimization and breaking through the traditional optimization model that relies on trial and error based on engineer experience.

[0156] By optimizing the parameter range and analyzing a large amount of data (such as the relationship between the lifting speed and coating uniformity under different precursor viscosities, the slope angle adjustment range and liquid mark coverage efficiency), the optimal parameter range for each type of solution can be determined, thus shortening the process debugging cycle for new solutions.

[0157] In electrode fabrication, the goal is never simply to pursue the maximum loading, but to obtain the highest possible effective loading while ensuring the film's high uniformity and good structure.

[0158] The present invention creatively treats excess solution, which is considered waste in traditional processes, as a deployable resource and redistributes it through spatial gradient flow, thereby significantly increasing the load without increasing raw material consumption.

[0159] It utilizes gravity, the most natural and uniform force, as the driving force, causing the liquid to spontaneously flow towards lower areas (thinner regions). Its homogenization effect is physical, gentler and more thorough than mechanical external forces (such as scrapers or air pressure). The advantages of this spatial slope flow technology are: it intelligently solves the impossible triangle of high load, high uniformity, and high material utilization by utilizing natural physical forces, providing a completely new solution for the large-scale, reproducible fabrication of high-performance electrodes.

[0160] like Figure 5 and Figure 6 As shown, based on the loading experiments of RuCl3 and IrCl3 in isopropanol / concentrated hydrochloric acid solutions on titanium plates, the average loading per unit area for every three dips using the traditional dip-coating method is 1.0760 mg / cm². 2 Load variance S 2 =0.0011<0.05, the average load per unit area for the impregnation reverse lifting method, the impregnation lifting-1st spatial slope flow method (bottom to handle slope flow once), and the impregnation lifting-multiple spatial slope flow method are 1.3185 mg / cm³. 2 1.7958 mg / cm 2 2.4427 mg / cm 2 Load variance S 2 =0.0011, 0.0014, 0.0075 < 0.05; the dip-coating-multiple-space slope flow method improved the average load per unit area by 127.01%, 85.26%, and 36.02% respectively compared with other methods. The production efficiency of 15 dip coatings was 50%~66.67% higher than that of the traditional dip-coating-pulling method (18 dip coatings), and the maximum load of 15 dip coatings reached 2.5638 mg / cm³. 2 It significantly improves film loading, uniformity, and consistency.

[0161] Depend on Figure 7 and Figure 8 It can be seen that, compared with the degradation effect of the self-made electrode and the commercial electrode on the water collection tank and MBR effluent of the dyeing and printing wastewater treatment plant, the self-made Ti / RuO2-IrO2 electrode has a significantly better treatment effect than the commercial electrode.

[0162] Key advantages include:

[0163] Uniformity: As can be seen from the load variance, the dip-coating process can make the surface coating more uniform, reduce the number of coatings for the same load, reduce the impact of interlayer differences on electrode performance, and improve the stability of the electrode when it is used after coating.

[0164] Production efficiency: Increased single coating amount reduces the time required to coat the same target amount per unit area, resulting in higher production efficiency; the simulation vision system recognizes and executes spatial slope flow coating, which increases the fluidity of the precursor solution compared to traditional or reverse dip coating processes, making it more conducive to solvent evaporation and resulting in shorter drying time;

[0165] Production costs: Fast and efficient loading reduces the production time of a single electrode; the simulation vision system recognizes and executes spatial slope flow coating, improving the yield and consistency of high-performance electrode preparation, saving reagents and substrates, and effectively reducing production costs.

[0166] The specific implementation principle of this invention embodiment:

[0167] The visual recognition system is initialized before the immersion and lifting process begins. A high-definition camera is deployed and positioned on the surface of the electrode plate to complete the initialization. This system uses a CCD or CMOS image sensor and a high-speed image acquisition module to monitor the solution flow on the electrode plate in real time.

[0168] After the visual recognition system is started, the image processing software uses image analysis algorithms (OpenCV, deep learning algorithms, edge detection methods, region growing methods, etc.) to process the images acquired by the camera in real time and analyze information such as the distribution, morphology, and flow direction of the liquid on the surface of the electrode plate.

[0169] Before the impregnation and lifting process begins, the parameters of the precursor solution to be used are initially entered (including solution type, solute type, solute concentration, and solution viscosity). If the precursor solution parameters match the precursor solution type used by the visual recognition system through deep learning, the visual recognition system can directly call the relevant impregnation and lifting and spatial slope flow stage parameters. If they do not match, the relevant parameters are manually entered, and the visual recognition system provides feedback and automatically adjusts them during the impregnation and lifting process, records the results, and adds them to the deep learning database.

[0170] The control system starts the metering pump to pump the precursor solution from the storage tank into the impregnation tank and maintain it at a constant height.

[0171] The robotic arm precisely controls the immersion motion of the electrode plate, dipping it into the precursor solution. The electrode plate descends slowly vertically, ensuring the solution evenly covers its surface. The immersion time and depth are determined by feedback from a vision system. If the vision system detects uneven solution distribution, the robotic arm adjusts the immersion time according to a pre-set algorithm to increase the uniformity of solution coverage. The robotic arm also adjusts the lifting speed based on feedback from the vision system. This adjustment is based on the spread of the solution on the electrode plate surface. A high lifting speed may result in insufficient solution distribution, while a low speed may cause excessive solution retention on the electrode plate surface. The vision system assesses the solution distribution in real time during each lifting operation. This speed must be neither too high nor too low; if too low a speed results in excessive solution retention on the electrode plate surface, it will ultimately affect the coating effect.

[0172] Spatial slope flow achieves uniform liquid film distribution and thickness control by adjusting the spatial orientation of the electrode plates (such as tilting or rotating) and utilizing the component of gravity to drive the directional flow of the precursor solution. This is combined with the volatilization kinetics of the organic solvent. When the electrode plates are tilted at an angle θ, the solution flows along the tilt direction under the influence of gravity, overcoming surface tension and viscous resistance to form a stable laminar flow. Gravity drives the solution to flow from high-thickness areas (accumulation points) to low-thickness areas (uncovered areas), while shear force breaks up local agglomerations, achieving thickness uniformity.

[0173] In the following situations, the robotic arm rotates the portion holding the electrode plate, using gravity to drive the precursor solution to flow on the electrode plate to complete the coating:

[0174] 1. After completing the impregnation and lifting process

[0175] 2. Localized accumulation occurs at the edges of the liquid trace, with a thickness deviation exceeding the set threshold;

[0176] 3. Liquid flow stagnates, and the flow rate is less than the set threshold;

[0177] 4. The precursor solution needs to be actively guided to cover the dead corner area at the edge of the electrode plate.

[0178] The robotic arm can perform two rotation modes:

[0179] 1. Single-axis tilt: The robotic arm rotates around the horizontal axis at 5~10° / s, and the speed is controlled to avoid inertia causing liquid film splashing, so that the electrode plate forms a set tilt angle θ (5~60°) with the horizontal plane. The liquid flow rate is controlled by controlling the tilt angle θ, and the component of gravity drives the solution to flow to the lower side, which is maintained for 3~30s.

[0180] 2. Multi-axis coordinated rotation: The robotic arm rotates around the x / y axis at 5~10° / s simultaneously. The centrifugal acceleration is less than or equal to the set threshold. Combined with the combined effect of centrifugal force and gravity, it guides the solution to spread evenly along the curved surface.

[0181] After the spatial gradient flow is completed, the robotic arm returns to a vertical position, allowing the electrode plate to be transferred to the next process. Based on the results of 18 loading experiments with isopropanol / concentrated hydrochloric acid solutions of SnCl2 and SbCl3 on titanium plates, the average loading capacity of the spatial gradient flow method is 1.31 mg / cm³. 2 The average loading of the traditional dip-coating method is 0.90 mg / cm³. 2 In comparison, the load efficiency of the spatial slope flow method is improved by 45%, such as Figure 4 As shown.

[0182] When performing specific visual recognition, a pre-built recognition model can be used for image import and data augmentation: collect images of liquid marks on electrode plates under different scenarios, including normal coating, edge overflow, and local accumulation, and build an labeled database (sample size ≥ 10,000 images). Image preprocessing includes normalization (resolution unified to 1920×1080), denoising, and enhancement (rotation and affine transformation to simulate the robotic arm's motion perspective).

[0183] The visual recognition system monitors the flow state of the solution on the electrode plate surface in real time through image acquisition and analysis. At this time, the algorithm automatically calculates the thickness of the precursor solution to infer whether local accumulation or overflow has occurred. The visual recognition system includes the following three modules: a parallel sampling processing module, used to perform dual-channel parallel sampling processing on the acquired high-definition video stream data to obtain a set of keyframe detail images and a set of motion capture images; a fast feature recognition module, used to quickly identify the liquid mark location and velocity features based on the keyframe detail image set to obtain the motion characteristics of the precursor solution on the electrode plate surface; and a feedback control module, used to feed back the data from the fast feature recognition module to the robotic arm control system, issuing control commands to the robotic arm controlling the electrode to adjust the flow state of the precursor solution on the electrode surface.

[0184] Based on high-definition video stream data, electrode plate images are extracted and semantically segmented to remove interfering backgrounds and generate an image set. Based on the image set results, a visual recognition system acquires the location and velocity data of the precursor solution traces and performs multimodal threshold fusion: First, the pixel distance from the trace edge to the electrode plate boundary is defined as the geometric boundary; second, the maximum allowable flow velocity (e.g., 3 mm / s) is defined as the flow velocity threshold; third, the maximum allowable deviation (±5%) of the region thickness from the mean is defined as the thickness deviation threshold. The data is then transmitted to the control system, which adjusts the robotic arm's operation (tilting the electrode plate speed, angle, etc.) to drive the precursor towards various corners of the electrode plate surface using multi-directional gravity, optimizing the coating effect and ensuring the solution uniformly covers the electrode plate without overflowing.

[0185] During the coating process, the vision recognition system continuously acquires real-time image data and compares it with the set coating standards to correct the uniformity and flow state of the precursor solution. If uneven liquid trace thickness or solution leakage at the edge of the electrode plate is detected, the system will automatically correct it, stop the spatial slope flow in that direction, and immediately adjust the direction to make the precursor solution flow uniformly again until the vision recognizes that the liquid trace thickness has reached the set standard.

[0186] After coating is completed, the visual recognition system continues to inspect the electrode plate for quality. This includes checking the uniformity of the solution coating, whether the coating thickness meets requirements, and whether there are any defects such as overflow. If defects are detected, such as uneven coating or overflow, the system will issue an alarm, suggesting that automatic adjustments may be needed to the expected immersion-lift or spatial slope flow time, or parameter corrections may be performed through new deep learning targeting the new precursor solution or electrode surface. The system also records coating data for each batch. Data from each coating operation (such as solution flow rate, lift speed, angle, and uniformity of precursor solution coating) is recorded and transmitted to the data management system for subsequent analysis and optimization. Through big data analysis, the system can continuously optimize operating parameters during the coating process, improving overall coating quality.

[0187] In this embodiment of the invention, the immersion-lifting-spatial slope flow control method based on a biomimetic vision system adjusts the operation of the robotic arm to allow the precursor solution to flow towards all corners of the electrode plate surface under gravity, optimizing the coating effect and ensuring that the solution uniformly covers the electrode plate without overflowing. During the coating process, the vision recognition system continuously acquires real-time image data and compares it with the set coating standards to correct the uniformity and flow state of the precursor solution, thereby improving the overall control efficiency.

[0188] Example 2

[0189] Please see Figure 9 , Figure 9This is a schematic diagram of the structure of the immersion-lifting-space slope flow control system based on a biomimetic vision system disclosed in an embodiment of the present invention. Figure 9 As shown, the biomimetic vision system-based immersion lifting-space slope flow control system may include:

[0190] Data acquisition module 21: used to acquire precursor solution parameters and determine the corresponding initial slope angle based on the precursor solution parameters;

[0191] Tilting module 22: Used to generate corresponding control commands based on the initial slope angle and the current state of the robotic arm so that the electrode plate held by the robotic arm is in the initial slope state, thereby causing the precursor solution to flow on the electrode plate under gravity.

[0192] Image acquisition module 23: used to acquire in real time the slope flow image data of the electrode plate and the precursor solution on the surface of the electrode plate through a camera during the gravity-driven flow of the precursor solution on the electrode plate, and to analyze the slope flow image data to determine the motion characteristic parameters of the precursor solution.

[0193] Comparison and adjustment learning module 24: used to compare the motion characteristic parameters with the set slope flow conditions, and generate adjustment instructions for the robotic arm based on the comparison results to dynamically adjust the angle of the electrode plate, thereby making the motion characteristic parameters of the precursor solution match the set control conditions.

[0194] In this embodiment of the invention, the immersion-lifting-spatial slope flow control method based on a biomimetic vision system adjusts the operation of the robotic arm to allow the precursor solution to flow towards all corners of the electrode plate surface under gravity, optimizing the coating effect and ensuring that the solution uniformly covers the electrode plate without overflowing. During the coating process, the vision recognition system continuously acquires real-time image data and compares it with the set coating standards to correct the uniformity and flow state of the precursor solution, thereby improving the overall control efficiency.

[0195] Example 3

[0196] Please see Figure 10 , Figure 10 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of the present invention. The electronic device can be a computer, a server, etc. Of course, in certain cases, it can also be a mobile phone, tablet computer, monitoring terminal, or other smart device, as well as an image acquisition device with processing capabilities. Figure 10 As shown, the electronic device may include:

[0197] Memory 510 storing executable program code;

[0198] Processor 520 coupled to memory 510;

[0199] The processor 520 calls the executable program code stored in the memory 510 to execute some or all of the steps in the immersion lifting-space slope flow control method based on the bionic vision system in Embodiment 1.

[0200] This invention discloses a computer-readable storage medium storing a computer program that causes a computer to perform some or all of the steps in the immersion lift-space slope flow control method based on a biomimetic vision system in Embodiment 1.

[0201] This invention also discloses a computer program product, wherein when the computer program product is run on a computer, the computer performs some or all of the steps in the immersion lift-space slope flow control method based on a bionic vision system in Embodiment 1.

[0202] This invention also discloses an application publishing platform, which is used to publish computer program products. When the computer program products are run on a computer, the computer executes some or all of the steps in the immersion lift-space slope flow control method based on a bionic vision system in Embodiment 1.

[0203] In various embodiments of the present invention, it should be understood that the sequence number of each process does not necessarily imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0204] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they can be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0205] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0206] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-accessible memory. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several requests to cause a computer device (which can be a personal computer, server, or network device, specifically a processor in the computer device) to execute some or all of the steps of the methods described in the various embodiments of the present invention.

[0207] In the embodiments provided by this invention, it should be understood that B corresponding to A means that B is associated with A, and B can be determined based on A. However, it should also be understood that determining B based on A does not mean determining B solely based on A; B can also be determined based on A and / or other information.

[0208] Those skilled in the art will understand that some or all of the steps in the various methods of the embodiments described can be implemented by a program instructing related hardware. This program can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.

[0209] The above provides a detailed description of the immersion-lift-space slope flow control method, system, electronic device, and storage medium based on a biomimetic vision system disclosed in the embodiments of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A dip-and-pull-space-slope-flow control method based on a biomimetic vision system, characterized by, include: Obtain the precursor solution parameters and determine the corresponding initial slope angle based on the precursor solution parameters; Based on the initial slope angle and the current state of the robotic arm, corresponding control commands are generated to make the electrode plate held by the robotic arm in the initial slope state, thereby causing the precursor solution to flow on the electrode plate under gravity. During the flow of the precursor solution in the electrode plate under gravity, the slope flow image data of the electrode plate and the precursor solution on the surface of the electrode plate are collected in real time by a camera, and the slope flow image data is analyzed to determine the motion characteristic parameters of the precursor solution. The motion characteristic parameters are compared with the set slope flow conditions, and the adjustment command of the robotic arm is generated based on the comparison result to dynamically adjust the angle of the electrode plate, so that the motion characteristic parameters of the precursor solution match the set control conditions.

2. The biomimetic vision system based dip and draw - spatial flow control method as claimed in claim 1, wherein, The control method further includes: The corresponding lifting speed of the electrode plate in the immersion tank is determined based on the parameters of the precursor solution. According to the lifting speed, the robotic arm is controlled to hold the electrode plate and lift it from the immersion tank at a set angle, parallel to the top and bottom surfaces of the immersion tank directly opposite the electrode plate, and remove it from the liquid surface.

3. The biomimetic vision system based dip and draw - spatial flow control method as claimed in claim 2, wherein, The camera includes a first camera and a second camera. The first camera is used to image the electrode plate and determine the shape and boundary coordinates of the electrode plate. The second camera is used to acquire images during the slope flow process. During the spatial slope flow process, the second camera shares the boundary coordinates provided by the first camera and tracks the movement of liquid traces within this coordinate range.

4. The biomimetic vision system based dip and draw - spatial flow control method according to claim 3, wherein, The control method further includes: The slope flow image data is identified to determine the current slope flow velocity, slope flow direction, and historical coating information; The area to be coated is determined based on the historical coating information, electrode plate shape, and boundary coordinates. Based on the area to be coated, a corresponding tilt angle is determined, and the robotic arm is controlled to tilt the electrode plate so that the precursor solution flows to the area to be coated. The tilt angle includes a first tilt angle and a second tilt angle.

5. The biomimetic vision system based dip and draw - spatial flow control method as claimed in claim 1, wherein, The analysis of the slope flow image data to determine the kinematic characteristic parameters of the precursor solution includes: The slope flow image data is analyzed using a semantic segmentation model to generate a liquid trace pixel mask associated with the liquid film; The vector motion direction and speed of the liquid trace pixel mask are calculated by determining the pixel displacement of the liquid trace pixel mask between frames of the image data sequence, and the vector motion direction and speed of the liquid trace pixel mask are used as motion characteristic parameters of the precursor solution.

6. The biomimetic vision based dip and draw - spatial flow control method as claimed in claim 5, wherein, The step of comparing the motion characteristic parameters with the set slope flow conditions and generating adjustment commands for the robotic arm based on the comparison results to dynamically adjust the angle of the electrode plate includes: The vector motion direction and velocity of the liquid trace are compared with the set slope flow conditions; When the vector motion rate of the detected liquid trace is lower than the preset rate threshold, a control command is generated to increase the tilt angle of the electrode plate. When the vector motion rate of the liquid trace is detected to be higher than the preset rate threshold, a control command is generated to reduce the tilt angle of the electrode plate. When the liquid trace is detected to be approaching the substrate boundary coordinates, a control command is generated to change the tilt direction of the electrode plate to adjust the flow of the liquid trace in the current direction.

7. The biomimetic vision based dip and draw - spatial flow control method as claimed in claim 5, wherein, The motion characteristic parameters also include liquid mark height information; The step of comparing the motion characteristic parameters with the set slope flow conditions and generating adjustment commands for the robotic arm based on the comparison results to dynamically adjust the angle of the electrode plate further includes: When the liquid level is detected to be below a set height threshold, the slope flow is stopped.

8. The immersion-lifting-spatial slope flow control method based on a biomimetic vision system as described in claim 1, characterized in that, The control method further includes: After the coating is applied, an image of the coated electrode plate is captured using a camera; The electrode plate image is subjected to quality inspection to determine whether there are coating defects. If coating defects are detected in the electrode plate image, the system issues an alarm. Record all coating operation data, including solution flow rate, lifting speed, slope angle, coating uniformity, and precursor solution parameters.

9. An immersion-lifting-spatial slope flow control system based on a biomimetic vision system, characterized in that, include: Data acquisition module: used to acquire precursor solution parameters and determine the corresponding initial slope angle based on the precursor solution parameters; Tilting module: used to generate corresponding control commands based on the initial slope angle and the current state of the robotic arm so that the electrode plate held by the robotic arm is in the initial slope state, thereby causing the precursor solution to flow on the electrode plate under gravity. Image acquisition module: used to acquire in real time the slope flow image data of the precursor solution on the electrode plate and the surface of the electrode plate through a camera during the gravity-driven flow of the precursor solution on the electrode plate, and to analyze the slope flow image data to determine the motion characteristic parameters of the precursor solution. The comparison and adjustment learning module is used to compare the motion characteristic parameters with the set slope flow conditions, and generate adjustment commands for the robotic arm based on the comparison results to dynamically adjust the angle of the electrode plate, thereby making the motion characteristic parameters of the precursor solution match the set control conditions.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program causes a computer to perform the immersion lift-space slope flow control method based on a biomimetic vision system as described in any one of claims 1 to 8.

Images