A closed-loop control system and method for winding across multiple processes

Abstract

This invention discloses a cross-process winding closed-loop control system and method, including a roll forming electrode thickness measuring device installed in the roll forming machine, a cutting and rolling thickness measuring device installed in the cutting and rolling equipment, and an edge calculator. The roll forming electrode thickness measuring device is used to measure the thickness of the electrode sheet during roll forming; the cutting and rolling thickness measuring device is used to measure the thickness of the electrode sheet in the Y direction; the edge calculator is configured with a neural network self-learning model, which outputs tension adjustment values, tab spacing adjustment values, and embossing roller pressure adjustment values ​​based on the electrode sheet roll forming thickness and the electrode sheet thickness in the Y direction. This invention finds the relationship between the incoming material thickness in two dimensions of the previous process, the tension value of the current process, and the tab misalignment value per turn and the electrode roll thickness deviation value. This enables timely adjustment of the tab spacing value, embossing roller start / stop, and embossing roller pressure value when the electrode sheet thickness or equipment tension changes, thereby ensuring the tab alignment and core thickness of the core during the winding process.

Description

Technical Field

[0001] This invention relates to the field of lithium battery manufacturing technology, and in particular to a winding closed-loop control system and method that spans multiple processes. Background Technology

[0002] Inconsistencies in the longitudinal thickness of electrode sheets from different or the same electrode roll ultimately manifest in the winding equipment, causing problems such as poor tab alignment and inconsistent core thickness in core manufacturing. Furthermore, data measured during coating and rolling cannot be correlated with the core product, hindering the traceability of information such as tab alignment and thickness. Tab alignment and thickness in core manufacturing are affected by the electrode sheet thickness and laser-cut tab spacing in the preceding rolling process, as well as by the tension and embossing roller pressure during the winding process. Currently, when alignment problems occur in the winding equipment, manual adjustment of the winding needle diameter using Teflon is employed. This method requires highly skilled personnel, is time-consuming, and delays production. Inconsistent core thickness manifests as inability to be encapsulated in subsequent processes or low single-cell capacity. Existing technologies mostly rely on adjusting the thickness and tab spacing before encapsulation, lacking a multi-faceted approach. Therefore, situations arise where the tabs are aligned, but the core thickness is too high to be encapsulated, or the thickness meets requirements but the tab spacing is the same, yet tab alignment is sometimes satisfactory and sometimes inadequate. Summary of the Invention

[0003] Based on the problems existing in the prior art, the present invention provides a cross-process winding closed-loop control system and method. By establishing model relationships, the relationship between the two dimensions of the incoming material thickness of the previous process, the tension value of the current process and the electrode misalignment value per turn and the electrode roll thickness deviation value is found. When the electrode thickness or equipment tension changes, the electrode spacing value, the start and stop of the embossing roller and the embossing roller pressure value are adjusted in time, thereby ensuring the electrode alignment and core thickness of the core during the winding process.

[0004] To achieve the above objectives, the present invention employs the following technical solution:

[0005] In a first aspect, the present invention provides a closed-loop control system for winding lithium batteries across processes, including a roll forming electrode thickness measuring device installed in the roll forming process, a winding thickness measuring device installed in the winding cutting equipment, and an edge calculator.

[0006] The roller-pressed electrode thickness measuring device is used to measure the roller-pressed thickness of the electrode.

[0007] The cutting and rolling thickness measuring device is used to measure the thickness of the electrode sheet in the Y direction.

[0008] The edge calculator is configured with a neural network self-learning model, which is used to output tension adjustment values, tab spacing adjustment values, and embossing roller pressure adjustment values ​​based on the electrode roll thickness and the electrode Y-direction thickness.

[0009] Preferably, the roller-pressed electrode thickness measuring device is mounted on a moving module, and the moving module drives the roller-pressed electrode thickness measuring device to move along the X direction of the electrode.

[0010] Preferably, the roller-pressed electrode thickness measuring device is used for,

[0011] The measured electrode roll thickness data is uploaded to the edge calculator to calculate the covariance and standard deviation of the electrode roll thickness, which are then input into the constructed neural network model.

[0012] Preferably, the roll cutting and thickness measuring device is located after the unwinding equipment and in front of the laser cutting equipment, and is fixed on the large plate of the roll cutting equipment.

[0013] Preferably, the roll cutting and thickness measuring device includes: a servo module, a laser rangefinder, a lower stabilizing roller, and an upper stabilizing roller;

[0014] The lower stabilizing roller and the upper stabilizing roller are used to fix and transfer the electrode sheet. The laser rangefinder is located on the upper and lower sides of the electrode sheet. The laser rangefinder is used to upload the measured thickness data of the electrode sheet in the Y direction to the edge calculator.

[0015] The servo module is used to drive the laser rangefinder.

[0016] Preferably, the slitting and rolling equipment further includes: an electrode spacing testing device, a tension control system, and a tension detection system;

[0017] The electrode spacing testing device is used to measure the electrode spacing and upload the data to the edge calculator;

[0018] The tension detection system is used to detect the tension value during electrode conveying and upload it to the edge calculator;

[0019] The tension control system is used to adjust the tension value.

[0020] Preferably, the electrode spacing testing device includes: a servo motor, a laser sensor, an encoder, and a winding needle;

[0021] The winding needle is used to wind small electrode pieces, the servo motor is used to drive the winding needle to rotate, and the laser sensor is installed at the center line aligned with the winding electrode tab, with a distance of h from the electrode piece; an encoder is installed on the winding needle.

[0022] There are at least two laser sensors, which are respectively set before the positive and negative electrodes are wound. The two laser sensors are used to mark the positive electrode Mark position, the start and end positions of the electrode tab, and the negative electrode Mark position, the start and end positions of the electrode tab, respectively.

[0023] Preferably, the closed-loop control system further includes a thickness gauge installed in the assembly and testing machine, which is used to detect the overall thickness of the electrode roll and determine whether it can be inserted into the aluminum shell.

[0024] Secondly, the present invention provides a closed-loop control method for cross-process winding of lithium batteries, implemented based on the aforementioned closed-loop control system for cross-process winding of lithium batteries, the method comprising:

[0025] A neural network self-learning model is constructed and trained. The neural network self-learning model takes the covariance and standard deviation of the electrode alignment thickness and the covariance and standard deviation of the electrode thickness as inputs, and takes the tension adjustment value, the electrode spacing adjustment value and the embossing roller pressure adjustment value as outputs.

[0026] The tension value is detected during the electrode conveying process to determine the change in tension of the electrode during the conveying process. If the tension change value is greater than the threshold ΔK, the tension value is adjusted.

[0027] The X-direction electrode thickness measured by the roller pressing electrode thickness measuring device and the Y-direction electrode thickness measured by the cutting and rolling thickness measuring device are obtained. It is then determined whether the deviation between the measured thickness and the theoretical thickness is greater than the electrode tab spacing adjustment thickness threshold △T1. If it is not greater, no adjustment is made. If it is greater than the electrode tab spacing adjustment thickness threshold △T1, it is further determined whether it is greater than the embossing roller adjustment threshold △T2. If it is greater than both △T1 and △T2, the embossing roller is turned on, and the pressure of the embossing roller is output based on the neural network self-learning model. If it is greater than △T1 and less than △T2, the electrode tab spacing is adjusted based on the neural network self-learning model.

[0028] Preferably, the construction and training of the neural network self-learning model includes:

[0029] Obtain a training dataset, which includes: acquiring the tension value, thickness measurement value in the X direction of the roller pressing, and thickness measurement value in the Y direction of the electrode sheet during the conveying process; calculating the covariance and standard deviation of the electrode tab alignment thickness, as well as the covariance and standard deviation of the electrode sheet thickness, based on the acquired data; combining the calculated covariance and standard deviation of the electrode tab alignment thickness, the covariance and standard deviation of the electrode sheet thickness, and the tension adjustment value, electrode tab spacing adjustment value, and embossing roller pressure adjustment value under the current working conditions as a set of training data; and obtaining the training dataset based on the measurement data and adjustment data during the electrode sheet conveying process in the above manner.

[0030] The training dataset is input into the constructed neural network self-learning model. The model learns to find the relationship between tab alignment, core thickness deviation and tension, tab spacing and embossing roller pressure adjustment value. The model parameters are then corrected based on the predicted deviation value and the actual detected deviation value to obtain the trained neural network self-learning model.

[0031] Preferably, the method further includes:

[0032] The adjusted tab spacing and tape thickness are detected, and the tension adjustment value, tab spacing adjustment value, and embossing roller pressure adjustment value are fed back to optimize the neural network self-learning model.

[0033] The beneficial effects of this invention are as follows:

[0034] This invention provides a cross-process winding closed-loop control system and method. By establishing model relationships, it identifies the relationships between the incoming material thickness in the previous process, the tension value in the current process, and the electrode misalignment value per turn and the electrode roll thickness deviation value. This enables real-time adjustment of the electrode spacing, embossing roller start / stop, and embossing roller pressure values ​​when the electrode thickness or equipment tension changes, thereby ensuring the electrode alignment and core thickness of the core during winding. This invention replaces manual inspection, analysis, and adjustment, accurately understands and obtains the location information of electrode misalignment, achieves pre-control, and reduces problems such as core misalignment defects, excessive thickness rework defects, and poor capacity consistency. Attached Figure Description

[0035] Figure 1 A diagram of a cross-process winding closed-loop control system provided by the present invention;

[0036] Figure 2 The cross-process winding closed-loop control logic diagram provided by the present invention;

[0037] Figure 3 The diagram of the closed-loop control neural network provided by this invention;

[0038] Figure 4 A front view of the winding thickness measuring device in the cross-process winding closed-loop control system provided by the present invention;

[0039] Figure 5 The overall structural diagram of the winding thickness measuring device in the cross-process winding closed-loop control system provided by the present invention;

[0040] Figure 6 A schematic diagram of the electrode spacing testing device in the cross-process winding closed-loop control system provided by the present invention;

[0041] Figure 7 This is a schematic diagram of the electrode detection position in this invention. Detailed Implementation

[0042] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The embodiments described below with reference to the accompanying drawings are illustrative and intended to explain the present invention, and should not be construed as limiting the present invention.

[0043] In the description of this invention, it should be noted that the terms "upper", "lower", "front", "rear", "end", "bottom", "side", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or component referred to must have a specific orientation. Therefore, they should not be construed as limiting this invention.

[0044] In this invention, unless otherwise explicitly specified and limited, the terms "connection," "installation," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a direct connection, or a connection through an intermediate medium. Those skilled in the art can understand the specific meaning of these terms in this invention according to the specific circumstances.

[0045] Secondly, the term "an embodiment" or "embodiment" as used in this invention refers to a specific feature, structure, or characteristic that can be included in at least one implementation of this invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that excludes other embodiments.

[0046] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention. Example 1

[0047] This embodiment 1 provides a closed-loop control system for lithium battery winding across processes, which is installed in the lithium battery production process, such as... Figure 1 As shown, the control system includes:

[0048] Thickness measuring device 1-1 for the roller pressing electrode installed in roller pressing 1;

[0049] The positive and negative unwinding device 2-1, the winding thickness measuring device 2-2, the embossing roller device 2-3, the laser cutting device 2-4, the electrode tab spacing testing device 2-5, the tension control system 2-6, the tension detection system 2-7, the roller guide 2-8, and the winding mechanism 2-9 are installed in the cutting and rolling equipment 2.

[0050] And a thickness gauge 3-1 installed in the assembly and testing machine 3.

[0051] Roller pressing 1, cutting and rolling 2, and assembly and testing machine 3 are different processes and equipment.

[0052] In this embodiment, the electrode sheet is uniformly cut into several rolls of small electrode sheets by the roller pressing 1. The small electrode sheets start from the positive and negative electrode unwinding device 2-1 installed on the cutting and rolling equipment 2, and move through the cutting and rolling thickness measuring device 2-2, the embossing roller device 2-3, and several passing rollers 2-8 to the laser cutting device 2-4. After the electrode tabs 5 are cut by the laser cutting device 2-4, the small electrode sheets pass through several passing rollers 2-8 to the winding mechanism 2-9. The electrode tab spacing detection device 2-5 is installed on the outside of the winding mechanism 2-9. During the conveying of the small electrode sheets, the tension control system 2-6 ensures continuous movement. The tension detection system 2-7 and the tension control system 2-6 form a closed-loop control to maintain tension balance.

[0053] Specifically, in this embodiment, the roller-pressed electrode thickness measuring device 1-1 moves along the X direction of the electrode to measure the lateral thickness of the "Z"-shaped electrode, i.e., the roller-pressed thickness. The measured thickness value is uploaded to the edge calculator, which outputs the thickness covariance and standard deviation data to the constructed neural network self-learning model.

[0054] It should be noted that, Figure 1 The control equipment is divided into two groups, left and right, which are used for cutting and rolling the positive and negative electrode sheets respectively.

[0055] It should be noted that the roller-pressed electrode thickness measuring device 1-1 is installed on the moving module, and is driven by the moving module to move along the X direction of the electrode; the roller-pressed electrode thickness measuring device is implemented using existing thickness measuring instruments.

[0056] It should be noted that the neural network self-learning model is configured on the edge calculator.

[0057] See Figure 4 and Figure 5 In this embodiment, the roll cutting and thickness measuring device 2-2 includes: a servo module 2-21, a laser rangefinder 2-22, a lower stabilizing roller 2-23, and an upper stabilizing roller 2-24;

[0058] Specifically,

[0059] The coil cutting and thickness measuring device 2-2 is set after the unwinding equipment and in front of the laser cutting equipment. The distance between the electrode and the laser cutting position needs to be greater than 2 meters. It is fixed on the large plate of the coil cutting equipment.

[0060] The laser rangefinder measures the thickness data of the electrode in the Y direction and uploads the measurement data to the edge calculator.

[0061] It should be noted that during the measurement process, the positive and negative electrodes are measured independently.

[0062] It should be noted that the distance from the thickness measuring position of the electrode sheet from the thickness measuring device 2-2 to the laser cutting device 2-4 should not be less than the maximum distance between the three electrode tabs. This is to allow time for communication and action execution.

[0063] In this embodiment, the embossing roller device 2-3 includes a pressure control unit and an embossing roller.

[0064] In this embodiment, the laser cutting device 2-4 is used to cut the electrode tabs. The size, shape, and spacing of the electrode tabs can be adjusted. The adjustment time interval should be less than the time interval between the small electrode moving from the thickness measuring position of the roll-cutting and thickness measuring device 2-2 to the cutting position of the laser cutting device 2-4.

[0065] See Figure 6 In this embodiment, the tab spacing testing device 2-5 includes: a servo motor, a laser sensor 2-52, an encoder 2-53, and a winding needle 2-54;

[0066] Specifically, the winding needle 2-54 is used to wind the small electrode sheet, the servo motor 2-51 is used to drive the winding needle to rotate, and the laser sensor 2-52 is installed at the center line aligned with the winding tab 5, with a distance of h from the electrode sheet. An encoder 2-53 is installed on the winding needle 2-54.

[0067] It should be noted that there are at least two laser sensors 2-52, which are respectively set before the positive and negative electrodes are wound. The laser sensors 2-52 are used to mark the positive electrode Mark position, the start and end positions of the electrode tab, and the negative electrode Mark position, the start and end positions of the electrode tab.

[0068] A mark hole is machined at the initial winding position. When the laser sensor 2-52 detects the mark hole, the encoder 2-53 records the number of turns i, where i ranges from 0 to N, and N equals the designed number of turns for the core. When winding to the positive / negative electrode tab position, the laser sensor 2-52 detects the straight line at the center of the tab. At this time, the encoder 2-53 records the starting rotation angle WZi1 and ending rotation angle WZi2 of the positive electrode tab, and the starting rotation angle WFi1 and ending rotation angle WFi2 of the negative electrode tab. Where: Z represents the positive electrode, F represents the negative electrode, and i represents the i-th turn. The encoder uploads the recorded rotation angle data to the edge calculator. Figure 7 This is a schematic diagram of the detection location.

[0069] The thickness gauge 3-1 installed in the assembly and testing machine 3 is used to detect the overall thickness of the electrode roll and determine whether it can be inserted into the aluminum shell. Example 2

[0070] Based on the winding closed-loop control system of Embodiment 1 above, Embodiment 2 provides a winding closed-loop control method to achieve real-time adjustment of the tab spacing, embossing roller start / stop, and embossing roller pressure when the electrode thickness or equipment tension changes, thereby ensuring the tab alignment and core thickness of the core during the winding process. (See also...) Figure 2 The method includes:

[0071] Step S1, establish as follows Figure 3 The neural network self-learning model shown obtains a training dataset, which includes:

[0072] The tension value, thickness measurement value in the X direction of the roller, and thickness measurement value in the Y direction of the electrode sheet were obtained during the conveying process.

[0073] Based on the acquired data, calculate the covariance and standard deviation of the electrode alignment thickness, as well as the covariance and standard deviation of the electrode thickness;

[0074] The calculated covariance and standard deviation of electrode alignment thickness, covariance and standard deviation of electrode thickness, as well as the tension adjustment value, electrode spacing adjustment value and embossing roller pressure adjustment value under the current working conditions are used together as a set of training data.

[0075] Obtain the training dataset as described above.

[0076] Step S2: Train the model. Input the data obtained in step S1 into the constructed neural network self-learning model. Through model self-learning, find the relationship between tab alignment, core thickness deviation and tension, tab spacing and embossing roller pressure adjustment value. And correct the model parameters based on the predicted deviation value and the actual detected deviation value to ensure prediction accuracy.

[0077] Step S3-1: Detect the tension value and determine the change in tension of the electrode during the conveying process. If the change in tension is greater than the threshold ΔK, adjust the tension value to ensure tension stability.

[0078] Step S3-2: Obtain the X-direction roll thickness measured by the roll forming electrode thickness measuring device and the Y-direction electrode thickness measured by the cutting and rolling thickness measuring device. Determine whether the deviation between the measured thickness and the theoretical thickness is greater than the electrode tab spacing adjustment threshold △T1. If it is not greater, there is no need to turn on the embossing roller or adjust the electrode tab spacing. If it is greater than △T1, further determine whether it is greater than the embossing roller adjustment threshold △T2. If it is greater than both △T1 and △T2, turn on the embossing roller and adjust the pressure of the embossing roller according to the actual thickness value. If it is greater than △T1 and less than △T2, adjust the electrode tab spacing according to the actual thickness value.

[0079] It should be noted that the thickness threshold △T1 for adjusting the tab spacing and the threshold △T2 for adjusting the embossing roller were obtained through model training.

[0080] Step S4: Detect the adjusted tab spacing and coil thickness, output the tab alignment deviation value and coil thickness deviation value, and provide feedback on whether the tension adjustment value, tab spacing adjustment value and embossing roller pressure adjustment value are too large or too small.

[0081] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the specification of this invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "or / and" as used herein includes any and all combinations of one or more of the associated listed items.

[0082] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0083] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A closed-loop control system for cross-process winding of lithium batteries, characterized in that, This includes a roll forming electrode thickness measuring device installed in the roll forming process, a roll cutting thickness measuring device installed in the roll cutting equipment, and an edge calculator; The roller-pressed electrode thickness measuring device is used to measure the roller-pressed thickness of the electrode. The cutting and rolling thickness measuring device is used to measure the thickness of the electrode sheet in the Y direction. The edge calculator is configured with a neural network self-learning model. Based on the tension value, roller pressing thickness, and Y-direction thickness of the electrode sheet during the conveying process, the edge calculator calculates the covariance and standard deviation of the electrode tab alignment thickness, as well as the covariance and standard deviation of the electrode sheet roller pressing thickness. The neural network self-learning model takes the covariance and standard deviation of the electrode tab alignment thickness and the covariance and standard deviation of the electrode sheet thickness as inputs, and outputs the tension adjustment value, the electrode tab spacing adjustment value, and the embossing roller pressure adjustment value. The winding closed-loop control system is used for, The tension value during electrode transport is detected to determine the change in tension during transport. If the tension change value exceeds a threshold... If so, adjust the tension value; The electrode thickness measured by the rolling electrode thickness measuring device and the electrode thickness in the Y direction measured by the slitting and rolling thickness measuring device are obtained. It is then determined whether the deviation between the measured thickness and the theoretical thickness is greater than the thickness threshold for adjusting the electrode tab spacing. If the value is not greater than the threshold, no adjustment is needed; if it is greater than the tab spacing threshold, adjust the thickness. Then, it is further determined whether the value exceeds the adjustment threshold of the embossing roller. If it is greater than And greater than If the pressure is greater than the set pressure, the embossing roller will be activated, and the pressure will be adjusted based on the neural network self-learning model. and less than The electrode spacing is then adjusted based on the output of the neural network self-learning model.

2. The lithium battery cross-process winding closed-loop control system according to claim 1, characterized in that, The roller-pressed electrode thickness measuring device is mounted on a moving module, which drives the roller-pressed electrode thickness measuring device to move along the X direction of the electrode.

3. The lithium battery cross-process winding closed-loop control system according to claim 2, characterized in that, The roller-pressed electrode thickness measuring device is used for, The measured electrode roll thickness data is uploaded to the edge calculator to calculate the covariance and standard deviation of the electrode roll thickness, which are then input into the constructed neural network self-learning model.

4. The lithium battery cross-process winding closed-loop control system according to claim 1, characterized in that, The roll cutting and thickness measuring device is located after the unwinding equipment and in front of the laser cutting equipment, and is fixed on the large plate of the roll cutting equipment.

5. A closed-loop control system for cross-process winding of lithium batteries according to claim 4, characterized in that, The roll cutting and thickness measuring device includes: a servo module, a laser rangefinder, a lower stabilizing roller, and an upper stabilizing roller; The lower stabilizing roller and the upper stabilizing roller are used to fix and transfer the electrode sheet. The laser rangefinder is located on the upper and lower sides of the electrode sheet. The laser rangefinder is used to upload the measured thickness data of the electrode sheet in the Y direction to the edge calculator. The servo module is used to drive the laser rangefinder.

6. A closed-loop control system for cross-process winding of lithium batteries according to claim 5, characterized in that, The cutting and rolling equipment also includes: an electrode spacing testing device, a tension control system, and a tension detection system; The electrode spacing testing device is used to measure the electrode spacing and upload the data to the edge calculator; The tension detection system is used to detect the tension value during the electrode conveying process and upload it to the edge calculator; The tension control system is used to adjust the tension value.

7. A closed-loop control system for cross-process winding of a lithium battery according to claim 6, characterized in that, The electrode spacing testing device includes: a servo motor, a laser sensor, an encoder, and a winding needle; The winding needle is used to wind small electrode pieces, the servo motor is used to drive the winding needle to rotate, and the laser sensor is installed at the center line aligned with the winding electrode tab, with a distance of h from the electrode piece; an encoder is installed on the winding needle. There are at least two laser sensors, which are respectively set before the positive and negative electrodes are wound. The two laser sensors are used to mark the positive electrode Mark position, the start and end positions of the electrode tab, and the negative electrode Mark position, the start and end positions of the electrode tab, respectively.

8. A closed-loop control system for cross-process winding of lithium batteries according to claim 1, characterized in that, The closed-loop control system also includes a thickness gauge installed in the assembly and testing machine. The thickness gauge is used to detect the overall thickness of the electrode roll and determine whether it can be inserted into the aluminum shell.

9. A closed-loop control method for cross-process winding of lithium batteries, characterized in that, Based on the lithium battery cross-process winding closed-loop control system according to any one of claims 1 to 8, the method includes: A neural network self-learning model is constructed and trained. The neural network self-learning model takes the covariance and standard deviation of the electrode alignment thickness and the covariance and standard deviation of the electrode thickness as inputs, and takes the tension adjustment value, the electrode spacing adjustment value and the embossing roller pressure adjustment value as outputs. The tension value during electrode transport is detected to determine the change in tension during transport. If the tension change value exceeds a threshold... If so, adjust the tension value; The X-direction thickness of the electrode sheet measured by the roll forming electrode sheet thickness measuring device and the Y-direction thickness measured by the slitting and rolling thickness measuring device are obtained. It is then determined whether the deviation between the measured thickness and the theoretical thickness exceeds the thickness threshold for adjusting the electrode tab spacing. If the value is not greater than the threshold, no adjustment is needed; if it is greater than the tab spacing threshold, adjust the thickness. Then, it is further determined whether the value exceeds the adjustment threshold of the embossing roller. If it is greater than And greater than If the pressure is greater than the set pressure, the embossing roller will be activated, and the pressure will be adjusted based on the neural network self-learning model. and less than The electrode spacing is then adjusted based on the output of the neural network self-learning model.

10. A closed-loop control method for cross-process winding of lithium batteries according to claim 9, characterized in that, The construction and training of the neural network self-learning model includes: Obtain a training dataset, which includes: acquiring the tension value, roller-pressed thickness, and Y-direction thickness measurement values ​​of the electrode sheet during the conveying process; calculating the tab alignment thickness covariance and standard deviation, as well as the electrode sheet thickness covariance and standard deviation values, based on the acquired data; combining the calculated tab alignment thickness covariance and standard deviation, electrode sheet thickness covariance and standard deviation values, and the tension adjustment value, tab spacing adjustment value, and embossing roller pressure adjustment value under the current operating conditions as a set of training data; and obtaining the training dataset based on the measurement and adjustment data during the electrode sheet conveying process in the above manner. The training dataset is input into the constructed neural network self-learning model. The model learns to find the relationship between tab alignment, tab thickness deviation and tension, tab spacing and embossing roller pressure adjustment value. The model parameters are then corrected based on the predicted deviation value and the actual detected deviation value to obtain the trained neural network self-learning model.

11. The lithium battery cross-process winding closed-loop control method according to claim 10, characterized in that, The method further includes: The adjusted tab spacing and tape thickness are detected, and the tension adjustment value, tab spacing adjustment value, and embossing roller pressure adjustment value are fed back to optimize the neural network self-learning model.

Images