Near-field electrostatic spinning accurate control method and device based on auxiliary electrode and visual learning

Abstract

The invention belongs to the field of micro-nano manufacturing and additive manufacturing, and particularly relates to a near-field electrostatic spinning accurate control method and device based on an auxiliary electrode and visual learning. A pair of parallel auxiliary electrodes is mounted in the X-axis direction and the Y-axis direction of the spray head respectively, and it is ensured that uniform horizontal electric fields are generated at the bottom end of the spray head and the area below the spray head to act on a jet flow generation area; setting process parameters of near-field electrostatic spinning, and starting processing; auxiliary electrode voltage in the same direction as the moving direction of the receiving device is set, and a horizontal electric field is formed to counteract the stretching effect of substrate moving on the jet flow; a visual learning system is started, images are collected in real time through a high-speed camera, key features are extracted through an image recognition algorithm, and information such as the jet flow angle, the fiber deposition path and the fiber diameter is collected in real time; according to original data collected by a visual learning system, a machine learning algorithm is applied to realize self-adaptive adjustment of auxiliary electrode voltage so as to improve fiber deposition precision. The device is used for solving the problem of fiber deposition errors caused by high-speed movement of a receiving device in the traditional near-field electrostatic spinning process.

Description

Technical Field

[0001] The present invention belongs to the technical field of micro-nano manufacturing and additive manufacturing, and specifically relates to a near-field electrostatic spinning precision control method and device based on auxiliary electrodes and visual learning. Background Art

[0002] In the field of micro-nano fiber preparation, near-field electrospinning technology has attracted much attention as a precise fiber deposition method. Compared with traditional electrospinning, near-field electrospinning can achieve precise positioning and directional deposition of fibers due to the smaller distance between the electrode and the receiving substrate, and has shown great potential in the fields of flexible electronic devices, tissue engineering scaffolds and sensors. However, due to the fast injection speed and small fiber size, near-field electrospinning still faces the challenge of insufficient control accuracy in high-precision pattern preparation and three-dimensional structure construction, especially in the production of complex patterns. There are considerable limitations.

[0003] Current near-field electrospinning systems mostly rely on simple electric field control methods, which indirectly control fiber deposition by adjusting parameters such as voltage or electrode position. However, during operation, environmental factors (such as humidity and temperature) and small fluctuations in equipment often lead to fiber deviation and accumulation, making it difficult to achieve structural uniformity. In terms of rapidly generating uniform and highly repeatable fiber structures, traditional technologies still cannot meet the needs, and more intelligent control methods are urgently needed.

[0004] In order to improve the precise control of the spinning process, auxiliary electrode technology has been introduced into the field of near-field electrospinning. By adding an auxiliary electrode to the main electrode, the electric field strength can be locally regulated, thereby more accurately controlling the fiber deposition position. This method enhances the spinning stability and uniformity of fiber distribution to a certain extent, but the existing auxiliary electrode solutions are mostly static and difficult to respond to changes in the deposition process in real time.

[0005] In addition, the application of visual learning technology makes the spinning process more intelligent. Visual learning combines image recognition with machine learning algorithms to monitor the jet state and fiber deposition state in real time, and adaptively adjust parameters to ensure accurate deposition and consistency of the fiber. The introduction of this technology effectively improves the convenience of fiber morphology control and the repeatability of spinning effects. Summary of the invention

[0006] The present invention provides a near-field electrospinning precision control method based on auxiliary electrodes and visual learning, which is used to solve the fiber deposition error problem caused by the high-speed movement of a receiving device in the process of traditional near-field electrospinning.

[0007] The present invention provides a near-field electrostatic spinning precision control device based on auxiliary electrodes and visual learning, which is used to realize a near-field electrostatic spinning precision control method based on auxiliary electrodes and visual learning.

[0008] The present invention is achieved through the following technical solutions:

[0009] A near-field electrospinning precision control method based on auxiliary electrodes and visual learning, the method comprising the following steps:

[0010] Step 1: A pair of parallel auxiliary electrodes 3 are installed in the X-axis and Y-axis directions of the nozzle 1, respectively, and a uniform horizontal electric field is generated at the bottom of the nozzle and the area below it to act on the jet generation area;

[0011] Step 2: Based on the settings of the nozzle 1 and the auxiliary electrode 3 in step 1, the process parameters of near-field electrospinning are set and the process is started. The process parameters include applied voltage, working distance between the nozzle and the receiving device, feeding speed and moving speed of the receiving device;

[0012] Step 3: Based on the process parameters of step 2, the auxiliary electrode voltage is set in the same direction as the moving direction of the receiving device to form a horizontal electric field to offset the stretching effect of the substrate movement on the jet;

[0013] Step 4: Start the visual learning system, collect images in real time through a high-speed camera, extract key features through an image recognition algorithm, and collect information on the jet angle, fiber deposition path, and fiber diameter in real time;

[0014] Step 5: Based on the raw data collected by the visual learning system, a machine learning algorithm is applied to adaptively adjust the auxiliary electrode voltage to improve the fiber deposition accuracy.

[0015] Furthermore, the step 1 is specifically as follows: a pair of insulating frames 2 are respectively installed in the X-axis and Y-axis directions of the nozzle 1, and the angle between two adjacent insulating frames 2 is 90 degrees. Each of the insulating frames 2 is inserted into the top of the auxiliary electrode 3, and the auxiliary electrode 3 is on the center line of the width direction of the insulating frame 2. The auxiliary electrode 3 and the axis of the nozzle 1 on the X-axis are located in the same X-ZP plane, and the auxiliary electrode 3 and the axis of the nozzle 1 on the Y-axis are located in the same Y-ZP plane, and an annular connector 7 is set at the bottom end of the four auxiliary electrodes 3.

[0016] Furthermore, in step 1, the auxiliary electrode 3 is made of a material with good conductivity and high voltage resistance, and has a thickness of 0.1 mm. The shape of the auxiliary electrode is processed using laser cutting technology to ensure accurate electrode size;

[0017] The spacing between a pair of parallel auxiliary electrodes 3 is 1.5 mm, which is precisely controlled by the frame structure for fixing the auxiliary electrodes.

[0018] Furthermore, the outer diameter of the nozzle used for near-field electrospinning in step 1 is 0.2 mm; the width of the auxiliary electrode 3 is twice the outer diameter of the nozzle, i.e. 0.4 mm, to ensure that the electric field formed by it can effectively cover a sufficient area and reduce the impact on the nozzle electric field.

[0019] Furthermore, in step 1, the height of the auxiliary electrode 3 is 2 mm; the auxiliary electrode fixed in the frame, whose bottommost position not blocked by the frame, is precisely set to 0.2 mm below the bottom of the nozzle, to ensure that the electric field generated by the auxiliary electrode can act on the jet in the controlled stretching stage to offset the deviation caused by the movement of the substrate.

[0020] Furthermore, the step 2 is specifically as follows: the material used is a polyvinyl pyrrolidone solution with a mass fraction of 5%, and the solvent is propylene glycol methyl ether; the setting range of the applied voltage is 1500-1800V; the working distance is set to 1mm; the setting range of the feeding speed is 0.2-0.5μL / min; and the setting range of the movement speed of the receiving device is 20mm / s-50mm / s.

[0021] Furthermore, in step 3, the auxiliary electrode that needs to be connected to the high-voltage power supply is controlled by the host computer according to the movement direction of the receiving device;

[0022] The setting range of the voltage applied between the pair of parallel auxiliary electrodes 3 is 500 to 800V.

[0023] Furthermore, in step 4, the visual learning system acquires images through two high-speed cameras;

[0024] Among them, the horizontally placed camera is used to collect jet state images, and the vertically placed camera is used to collect deposition images of spun fibers; the collected images are subjected to binarization, noise removal, edge detection and feature extraction operations, and finally key information such as the jet angle, fiber deposition diameter, and fiber offset distance are extracted.

[0025] Furthermore, in step 5, the visual learning system uses a machine learning algorithm, takes the jet angle, the distance between the fiber deposition diameter and the optimal value, and the fiber offset distance as input, and takes the voltage deviation value as output, and establishes a relationship model between input and output data through algorithms such as random forests, neural networks, or multi-layer perceptrons, so as to adaptively adjust the voltage applied to the auxiliary electrode in real time through the model.

[0026] A near-field electrostatic spinning precision control device based on auxiliary electrodes and visual learning, the device is adjusted using the near-field electrostatic spinning precision control method based on auxiliary electrodes and visual learning as described above, the device comprises a nozzle 1, an insulating frame 2 and an auxiliary electrode 3, a pair of insulating frames 2 are respectively installed in the X-axis and Y-axis directions of the nozzle 1, the angle between two adjacent insulating frames 2 is 90 degrees, each of the insulating frames 2 is inserted into the top of the auxiliary electrode 3, the auxiliary electrode 3 is on the center line of the width direction of the insulating frame 2, the axis of the auxiliary electrode 3 and the nozzle 1 on the X-axis is located in the same X-ZP plane, the axis of the auxiliary electrode 3 and the nozzle 1 on the Y-axis is located in the same Y-ZP plane, an annular connector 7 is provided at the bottom end of the four auxiliary electrodes 3, the annular connector 7 is used to fix the relative position of the insulating frame 2 and the nozzle 1, the nozzle 1 is connected to the feeding barrel 4, the feeding barrel 4 is fixed to the connecting plate 6 through the fixing connector 5, and the connecting plate 6 is fixed to the Z-axis motion module, so as to accurately control the position and movement of the nozzle 1 in the Z direction.

[0027] The beneficial effects of the present invention are:

[0028] In the present invention, especially when the fiber structure has specific arrangement requirements (such as parallel or vertical relationship), the jet stretching effect caused by the movement of the substrate may cause the deposition accuracy to decrease, affecting the integrity and consistency of the fiber structure. To this end, the present invention proposes a near-field electrospinning precision control method combining auxiliary electrodes and visual learning. By configuring auxiliary electrodes around the print head and adjusting the electric field distribution in real time, the jet path can be precisely controlled to meet the needs of high-precision fiber structure manufacturing.

[0029] The present invention realizes real-time monitoring and precise control of the near-field electrospinning process by combining auxiliary electrodes with visual learning.

[0030] The auxiliary electrode of the present invention is used to enhance the electric field regulation capability, while visual learning provides intelligent monitoring and feedback adjustment, thereby overcoming the shortcomings of traditional methods and ensuring the accuracy and stability of the fiber structure.

[0031] The present invention not only improves the spinning efficiency and fiber quality, but also provides a new method for the orientation and patterning of micro-nano fibers. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a structural schematic diagram of the present invention.

[0033] Figure 2 This is the near-field electrospinning jet state when the auxiliary electrode is not activated in the second embodiment of the present invention.

[0034] Figure 3This is the near-field electrospinning jet state in which the auxiliary electrode is enabled in the second embodiment of the present invention but is not analyzed and regulated by the visual learning system.

[0035] Figure 4 This is the state of the near-field electrospinning jet in which the auxiliary electrode is enabled and the jet is analyzed and regulated by the visual learning system in the second embodiment of the present invention.

[0036] Figure 5 It is a schematic diagram of the processing trajectory of near-field electrospinning set by the host computer of the present invention.

[0037] Figure 6 This is the fiber deposition state obtained when the auxiliary electrode is not activated in the third embodiment of the present invention.

[0038] Figure 7 The auxiliary electrode is activated in the third embodiment of the present invention, and the fiber deposition state is obtained through analysis and regulation by the visual learning system. DETAILED DESCRIPTION

[0039] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0040] It should be understood that when used in this specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0041] It should also be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, the singular forms "a", "an" and "the" are intended to include plural forms unless the context clearly indicates otherwise.

[0042] The following is attached to this application specification Figure 1-7 , the technical solutions in the embodiments of the present application are clearly and completely described. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0043] In the following description, many specific details are set forth to facilitate a full understanding of the present application, but the present application may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present application. Therefore, the present application is not limited to the specific embodiments disclosed below.

[0044] Implementation Method 1

[0045] This embodiment provides a near-field electrospinning precision control method based on auxiliary electrodes and visual learning, the method comprising the following steps:

[0046] Step 1: Install a pair of parallel auxiliary electrodes 3 in the X-axis and Y-axis directions of the nozzle 1, and ensure that a uniform horizontal electric field is generated at the bottom of the nozzle and the area below it to act on the jet generation area, such as Figure 1 As shown;

[0047] A micro-terminal is arranged on the frame, one end of the terminal is connected and fixed to the auxiliary electrode through a conductive silver glue, and the other end is connected to the high-voltage power supply through a high-voltage insulated wire; in each pair of parallel auxiliary electrodes, one auxiliary electrode is connected to the high-voltage output end of the high-voltage power supply, and the other is connected to the ground end of the high-voltage power supply;

[0048] Step 2: Based on the settings of the nozzle 1 and the auxiliary electrode 3 in step 1, the process parameters of near-field electrospinning are set and the process is started. The process parameters include applied voltage, working distance between the nozzle and the receiving device, feeding speed and moving speed of the receiving device;

[0049] Step 3: Based on the process parameters of step 2, the auxiliary electrode voltage is set in the same direction as the moving direction of the receiving device to form a horizontal electric field to offset the stretching effect of the substrate movement on the jet;

[0050] Step 4: Start the visual learning system, collect images in real time through a high-speed camera, extract key features through an image recognition algorithm, and collect information differences such as jet angle, fiber deposition path, and fiber diameter in real time;

[0051] Step 5: Adaptively adjust the auxiliary electrode voltage based on the visual learning system feedback to improve the fiber deposition accuracy.

[0052] Furthermore, the step 1 is specifically as follows: a pair of insulating frames 2 are respectively installed in the X-axis and Y-axis directions of the nozzle 1, and the angle between two adjacent insulating frames 2 is 90 degrees. Each of the insulating frames 2 is inserted into the top of the auxiliary electrode 3, and the auxiliary electrode 3 is on the center line of the width direction of the insulating frame 2. The auxiliary electrode 3 and the axis of the nozzle 1 on the X-axis are located in the same X-ZP plane, and the auxiliary electrode 3 and the axis of the nozzle 1 on the Y-axis are located in the same Y-ZP plane. An annular connector 7 is set at the bottom end of the four auxiliary electrodes 3, and the annular connector 7 is used to fix the relative position of the insulating frame 2 and the nozzle 1.

[0053] Furthermore, in step 1, the auxiliary electrode 3 is made of a material with good electrical conductivity and high voltage resistance, such as a copper-plated thin metal rectangular sheet with a thickness of 0.1 mm, and the shape of the auxiliary electrode is processed using laser cutting technology to ensure accurate electrode size;

[0054] The spacing between a pair of parallel auxiliary electrodes 3 is 1.5 mm, which is precisely controlled by the frame structure for fixing the auxiliary electrodes.

[0055] Furthermore, the outer diameter of the nozzle used for near-field electrospinning in step 1 is 0.2 mm; the width of the auxiliary electrode 3 is twice the outer diameter of the nozzle, i.e. 0.4 mm, to ensure that the electric field formed by it can effectively cover a sufficient area and reduce the impact on the nozzle electric field.

[0056] Furthermore, in step 1, the height of the auxiliary electrode 3 is 2 mm; the auxiliary electrode fixed in the frame, whose bottommost position not blocked by the frame, is precisely set to 0.2 mm below the bottom of the nozzle, to ensure that the electric field generated by the auxiliary electrode can act on the jet in the controlled stretching stage to offset the deviation caused by the movement of the substrate.

[0057] The frame structure used to fix the auxiliary electrode is made of insulating materials such as polytetrafluoroethylene (PTFE); the frame structure consists of 4 separate frames of the same size, and each frame is assembled from 4 strips of material with a square cross-section; the inner and outer circle widths of each separate frame are 0.4mm and 1mm respectively, the inner and outer circle heights are 1.8mm and 2.4mm respectively, and the thickness is 0.3mm; the inner circle of the frame, at the center line in the width direction, is processed with a groove with a width of 0.15mm and a depth of 0.1mm for placing and fixing the auxiliary electrode.

[0058] Furthermore, the step 2 is specifically as follows: the material used is a polyvinyl pyrrolidone (PVP) solution with a mass fraction of 5%, and the solvent is propylene glycol methyl ether; the setting range of the applied voltage is 1500-1800V; the working distance is set to 1mm; the setting range of the feeding speed is 0.2-0.5μL / min; and the setting range of the movement speed of the receiving device is 20mm / s-50mm / s.

[0059] Furthermore, in step 3, the auxiliary electrode that needs to be connected to the high-voltage power supply is controlled by the host computer according to the movement direction of the receiving device;

[0060] The setting range of the voltage applied between the pair of parallel auxiliary electrodes 3 is 500 to 800V.

[0061] Furthermore, in step 4, the visual learning system acquires images through two high-speed cameras;

[0062] Among them, the horizontally placed camera is used to collect jet state images, and the vertically placed camera is used to collect deposition images of spun fibers; the collected images are subjected to binarization, noise removal, edge detection and feature extraction operations, and finally key information such as the jet angle, fiber deposition diameter and fiber offset distance are extracted.

[0063] The visual learning system includes two high-speed cameras and machine learning algorithm software.

[0064] Furthermore, the machine learning algorithm built into the visual learning system in step 5 takes the jet angle, the distance between the fiber deposition diameter and the optimal value, and the fiber offset distance as input, and the voltage deviation value as output, and establishes a relationship model between input and output data through a random forest, neural network or multi-layer perceptron algorithm, thereby adaptively adjusting the voltage applied to the auxiliary electrode in real time through the model.

[0065] Implementation Method 2

[0066] Based on the first embodiment, the following embodiments are given:

[0067] Step 1: Dissolve 5g of polyvinyl pyrrolidone (PVP) in 95g of propylene glycol methyl ether solvent to prepare a near-field electrospinning solution with a concentration of 5wt%. Add the solution to the near-field electrospinning device, set the voltage to 1500V, the working distance to 1mm, the feed rate to 0.2μL / min, and the moving speed of the receiving device to 40mm / s. Perform near-field electrospinning, and use a horizontal camera to collect the jet state when the auxiliary electrode is not enabled, such as Figure 2 shown.

[0068] Step 2: Enable the auxiliary electrode, and set the voltage applied between a pair of parallel auxiliary electrodes to 500 V. The jet state at this time is captured by a horizontal camera, such as Figure 3 shown.

[0069] Step 3: The visual learning system analyzes and processes the jet state collected in step 2, and adjusts the voltage applied between the auxiliary electrodes in real time. The adjusted voltage is 640 V. The jet state at this time is collected by a horizontal camera, such as Figure 4 shown.

[0070] contrast Figure 2 to Figure 4 It can be found that after the auxiliary electrode is enabled, the deposition accuracy of the jet is significantly improved. Figure 2 ), the jet is significantly deflected by the high-speed moving substrate, causing the deposition path to deviate from the target trajectory. After applying an auxiliary electrode voltage of 500V in step 2 ( Figure 3 ), the jet deviation is reduced and the deposition path tends to be stable, but there is still a certain error. After the auxiliary electrode voltage is adjusted to 640V ( Figure 4 ), the jet further tends to be stable, and the deposition path is highly consistent with the predetermined trajectory, which proves the effective regulation of the electric field on the jet.

[0071] Implementation Method 3

[0072] Based on the first embodiment, the following embodiments are given:

[0073] Step 1: The configuration of the near-field electrospinning solution is the same as that described in Example 1. The solution is added to the near-field electrospinning device, and the voltage is set to 1700 V, the working distance is 1 mm, the feeding speed is 0.2 μL / min, and the moving speed of the receiving device is 50 mm / s.

[0074] Step 2: Set the processing trajectory of near-field electrospinning through the host computer, such as Figure 5 The fiber deposition state on the receiving substrate is captured by a vertical camera, as shown in Figure 6 shown.

[0075] Step 3: Enable the auxiliary electrode and visual learning system, set the same processing trajectory for processing, and use the vertical camera to collect the fiber deposition state on the receiving substrate, such as Figure 7 shown.

[0076] contrast Figure 6 and Figure 7In the fiber deposition state in the process, when the auxiliary electrode is not enabled, the deposition path of the fiber at a high moving speed has a large deviation, and it is difficult to completely match the predetermined trajectory, showing a trend of decreased deposition accuracy; after the auxiliary electrode and the visual learning system are enabled, the system adaptively adjusts the voltage value of the auxiliary electrode according to the real-time feedback information, so that the jet gradually tends to the preset trajectory during the deposition process, and the deposition error is greatly reduced. The above results show that the auxiliary electrode and the visual learning system significantly improve the fiber deposition accuracy in the near-field electrospinning process using a high-speed receiving device, verifying the effectiveness of this method.

[0077] Implementation Method 4

[0078] The present embodiment provides a near-field electrospinning precision control device based on auxiliary electrodes and visual learning, the device is adjusted using the near-field electrospinning precision control method based on auxiliary electrodes and visual learning as described in the first embodiment, the device includes a nozzle 1, an insulating frame 2 and an auxiliary electrode 3, a pair of insulating frames 2 are respectively installed in the X-axis and Y-axis directions of the nozzle 1, the angle between two adjacent insulating frames 2 is 90 degrees, each of the insulating frames 2 is inserted into the top of the auxiliary electrode 3, the auxiliary electrode 3 is on the center line of the width direction of the insulating frame 2, the auxiliary electrode 3 and the axis of the nozzle 1 on the X-axis are located in the same X-ZP plane, the auxiliary electrode 3 and the axis of the nozzle 1 on the Y-axis are located in the same Y-ZP plane, an annular connector 7 is provided at the bottom end of the four auxiliary electrodes 3, the annular connector 7 is used to fix the relative position of the insulating frame 2 and the nozzle 1, the nozzle 1 is connected to the feed barrel 4, the feed barrel 4 is fixed to the connecting plate 6 through the fixed connector 5, and the connecting plate 6 is fixed to the Z-axis motion module, so as to accurately control the position and movement of the nozzle 1 in the Z direction.

Claims

1. A near-field electrospinning precision control method based on auxiliary electrodes and visual learning, characterized in that: The method comprises the following steps: Step 1: Install a pair of parallel auxiliary electrodes (3) in the X-axis and Y-axis directions of the nozzle (1) respectively, and ensure that a uniform horizontal electric field is generated at the bottom of the nozzle and the area below it to act on the jet generation area; Step 2: Based on the settings of the nozzle (1) and the auxiliary electrode (3) in step 1, the process parameters of near-field electrospinning are set and the process is started, wherein the process parameters include applied voltage, working distance between the nozzle and the receiving device, feeding speed and moving speed of the receiving device; Step 3: Based on the process parameters of step 2, the auxiliary electrode voltage is set in the same direction as the moving direction of the receiving device to form a horizontal electric field to offset the stretching effect of the substrate movement on the jet; Step 4: Start the visual learning system, collect images in real time through a high-speed camera, extract key features through an image recognition algorithm, and collect information on the jet angle, fiber deposition path, and fiber diameter in real time; Step 5: Based on the raw data collected by the visual learning system, a machine learning algorithm is applied to adaptively adjust the auxiliary electrode voltage to improve the fiber deposition accuracy.

2. According to claim 1, a near-field electrospinning precision control method based on auxiliary electrodes and visual learning is characterized in that: Specifically, step 1 comprises: installing a pair of insulating frames (2) in the X-axis and Y-axis directions of the nozzle (1), respectively, with the angle between two adjacent insulating frames (2) being 90 degrees, each insulating frame (2) being inserted into the top end of an auxiliary electrode (3), the axis of the auxiliary electrode (3) and the nozzle (1) on the X-axis being located in the same X-ZP plane, the axis of the auxiliary electrode (3) and the nozzle (1) on the Y-axis being located in the same Y-ZP plane, and an annular connecting piece (7) being provided at the bottom ends of the four auxiliary electrodes (3).

3. According to claim 1, a near-field electrospinning precision control method based on auxiliary electrodes and visual learning is characterized in that: In step 1, the auxiliary electrode (3) is made of a material with good conductivity and high voltage resistance, and has a thickness of 0.1 mm. The shape of the auxiliary electrode is processed using laser cutting technology to ensure accurate electrode size; The spacing between a pair of parallel auxiliary electrodes (3) is 1.5 mm, which is precisely controlled by a frame structure for fixing the auxiliary electrodes.

4. According to claim 1, a near-field electrospinning precision control method based on auxiliary electrodes and visual learning is characterized in that: The outer diameter of the nozzle used for near-field electrospinning in step 1 is 0.2 mm; the width of the auxiliary electrode (3) is twice the outer diameter of the nozzle, i.e. 0.4 mm, to ensure that the electric field formed by it can effectively cover a sufficient area and reduce the impact on the nozzle electric field.

5. According to claim 1, a near-field electrospinning precision control method based on auxiliary electrodes and visual learning is characterized in that: The height of the auxiliary electrode (3) in step 1 is 2 mm; the auxiliary electrode fixed in the frame, whose bottom position not blocked by the frame, is precisely set to 0.2 mm below the bottom of the nozzle, to ensure that the electric field generated by the auxiliary electrode can act on the jet in the controlled stretching stage to offset the deviation caused by the movement of the substrate.

6. According to claim 1, a near-field electrospinning precision control method based on auxiliary electrodes and visual learning is characterized in that: Specifically, the material used in step 2 is a 5% polyvinyl pyrrolidone solution, and the solvent is propylene glycol methyl ether; the setting range of the applied voltage is 1500-1800V; the working distance is set to 1mm; the setting range of the feeding speed is 0.2-0.5μL / min; and the setting range of the movement speed of the receiving device is 20mm / s-50mm / s.

7. According to claim 1, a near-field electrospinning precision control method based on auxiliary electrodes and visual learning is characterized in that: In step 3, the auxiliary electrode that needs to be connected to the high-voltage power supply is controlled by the host computer according to the movement direction of the receiving device; The setting range of the voltage applied between a pair of parallel auxiliary electrodes (3) is 500 to 800V.

8. According to claim 1, a near-field electrospinning precision control method based on auxiliary electrodes and visual learning is characterized in that: In step 4, the visual learning system acquires images through two high-speed cameras; Among them, the horizontally placed camera is used to collect the jet state image, and the vertically placed camera is used to collect the deposition image of the spun fiber; the collected images are subjected to binarization processing, noise removal, edge detection and feature extraction operations, and finally the information of the jet angle, fiber deposition diameter and fiber offset distance is extracted.

9. The near-field electrospinning precision control method based on auxiliary electrodes and visual learning according to claim 1 is characterized in that: The machine learning algorithm built into the visual learning system in step 5 takes the jet angle, the distance between the fiber deposition diameter and the optimal value, and the fiber offset distance as input, and the voltage deviation value as output, and establishes a relationship model between input and output data through a random forest, neural network, or multi-layer perceptron algorithm, thereby adaptively adjusting the voltage applied to the auxiliary electrode in real time through the model.

10. A near-field electrospinning precision control device based on auxiliary electrodes and visual learning, characterized in that: The device is adjusted using the near-field electrospinning precision control method based on auxiliary electrodes and visual learning as described in any one of claims 1 to 9. The device comprises a nozzle (1), an insulating frame (2) and an auxiliary electrode (3). A pair of insulating frames (2) are respectively installed in the X-axis and Y-axis directions of the nozzle (1). The angle between two adjacent insulating frames (2) is 90 degrees. Each insulating frame (2) is inserted into the top of the auxiliary electrode (3). The axis of the auxiliary electrode (3) and the nozzle (1) on the X-axis are located in the same X-ZP plane. The axis of the auxiliary electrode (3) and the nozzle (1) on the Y-axis are located in the same Y-ZP plane. An annular connector (7) is arranged at the bottom end of the four auxiliary electrodes (3). The nozzle (1) is connected to the feed barrel (4). The feed barrel (4) is fixed to the connecting plate (6) through a fixed connector (5). The connecting plate (6) is fixed to the Z-axis motion module, so as to accurately control the position and movement of the nozzle (1) in the Z direction.

Images