Winding control method, system and equipment based on pole roll thickness rebound prediction

Abstract

The embodiment of the invention relates to the technical field of battery processing, and discloses a pole roll thickness rebound prediction-based winding control method, system and device and a storage medium, and the method comprises the following steps: obtaining current device data of a target device for rolling a to-be-wound pole roll and a current measurement thickness of the to-be-wound pole roll after rolling; inputting the current equipment data into a trained thickness rebound prediction model to obtain a predicted rebound thickness of the to-be-wound pole roll; and according to the current measurement thickness and the predicted rebound thickness, the to-be-wound pole roll is controlled to be wound. According to the method provided by the embodiment of the invention, the to-be-wound pole roll can be jointly controlled to be wound according to the predicted rebound thickness and the current measured thickness, so that tab dislocation caused by thickness rebound of the to-be-wound pole roll in the winding process is avoided, the rejection rate of a wound battery cell and the solid consumption of a single roll are reduced, and the production cost is saved.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of battery processing, and in particular to a winding control method, system, device and storage medium based on the prediction of the thickness rebound of the electrode roll. Background Art

[0002] With the rapid development of the new energy industry, lithium-ion batteries have been widely used in fields such as electric vehicles and energy storage. In the production and manufacturing process of lithium-ion batteries, the rolling process is a key link in the formation of the electrode sheet, and its quality directly affects the subsequent winding process and the performance and safety of the final battery. However, in the prior art, there is a thickness rebound phenomenon in the electrode roll after rolling, which brings many problems to the winding process.

[0003] Specifically, due to the characteristics of the negative electrode material, the thickness of the rolled negative electrode sheet will rebound within a certain period of time, resulting in a deviation between the actual thickness and the set thickness during rolling. This thickness rebound phenomenon makes the winding operation based on the immediate thickness data after rolling unable to accurately predict the actual state of the electrode sheet in the subsequent process, resulting in inaccurate data sources. Furthermore, during the winding process, due to the uncertainty of the electrode sheet thickness, the problem of pole ear misalignment is likely to occur. Especially when the thickness changes greatly after the negative electrode roll change, the amount of pole ear misalignment is particularly large, which will increase the scrap rate of the wound battery cells and the unit consumption of each roll, greatly increasing the production cost. Summary of the Invention

[0004] The purpose of the embodiments of the present application is to provide a winding control method, system, device and storage medium based on the prediction of the thickness rebound of the electrode roll. By predicting the possible rebound thickness of the electrode roll to be wound during the winding process through a thickness rebound prediction model, and jointly controlling the electrode roll to be wound for winding according to the predicted rebound thickness and the current measured thickness, it is possible to avoid the pole ear misalignment caused by the thickness rebound of the electrode roll to be wound during the winding process, reduce the scrap rate of the wound battery cells and the unit consumption of each roll, and save the production cost.

[0005] To solve the above technical problems, the embodiments of the present application provide a winding control method based on the prediction of the thickness rebound of the electrode roll, including: obtaining the current device data of the target device for rolling the electrode roll to be wound and the current measured thickness of the electrode roll to be wound after rolling; inputting the current device data into a trained thickness rebound prediction model to obtain the predicted rebound thickness of the electrode roll to be wound; and controlling the electrode roll to be wound for winding according to the current measured thickness and the predicted rebound thickness.

[0006] Embodiments of the present application also provide a winding control system based on prediction of the thickness rebound of a pole coil, including: a data acquisition module, configured to acquire current device data of a target device for rolling the pole coil to be wound and the current measured thickness of the pole coil to be wound after rolling; a rebound prediction module, configured to input the current device data into a trained thickness rebound prediction model to obtain the predicted rebound thickness of the pole coil to be wound; and a winding control module, configured to control the winding of the pole coil to be wound according to the current measured thickness and the predicted rebound thickness.

[0007] Embodiments of the present application also provide an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the above-mentioned winding control method based on prediction of the thickness rebound of a pole coil.

[0008] Embodiments of the present application also provide a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the above-mentioned winding control method based on prediction of the thickness rebound of a pole coil is implemented.

[0009] The winding control method based on prediction of the thickness rebound of a pole coil provided by the embodiments of the present application includes: acquiring current device data of a target device for rolling the pole coil to be wound and the current measured thickness of the pole coil to be wound after rolling; inputting the current device data into a trained thickness rebound prediction model to obtain the predicted rebound thickness of the pole coil to be wound; and controlling the winding of the pole coil to be wound according to the current measured thickness and the predicted rebound thickness. By acquiring the current device data and the current measured thickness, using the trained thickness rebound prediction model to predict the rebound thickness of the pole coil, and precisely controlling the winding process according to the prediction result. This method can effectively solve the problem of misalignment of the pole ears of the wound battery core caused by the thickness rebound of the pole coil, reduce the scrap rate of the wound battery core and the unit consumption of a single roll, and save production costs. Description of the Drawings

[0010] One or more embodiments are exemplarily illustrated by pictures in the corresponding drawings, and these exemplary illustrations do not limit the embodiments.

[0011] Figure 1 is the flow of the winding control method based on prediction of the thickness rebound of a pole coil according to an embodiment of the present application Figure 1 ;

[0012] Figure 2 is the flow of the winding control method based on prediction of the thickness rebound of a pole coil in another embodiment of the present application Figure 2 ;

[0013] Figure 3 is the flowchart of the winding control method based on the prediction of the thickness rebound of the pole roll according to another embodiment of the present application Figure 3 ;

[0014] Figure 4 is the flowchart of the winding control method based on the prediction of the thickness rebound of the pole roll according to another embodiment of the present application Figure 4 ;

[0015] Figure 5 is a schematic structural diagram of the winding control system based on the prediction of the thickness rebound of the pole roll according to an embodiment of the present application;

[0016] Figure 6 is a schematic structural diagram of an electronic device according to another embodiment of the present application. Specific embodiments

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the following will elaborate on each embodiment of the present application with reference to the accompanying drawings. However, those of ordinary skill in the art can understand that in each embodiment of the present application, many technical details are provided to help readers better understand the present application. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solutions claimed in the present application can still be implemented. The following division of each embodiment is for convenience of description and should not constitute any limitation to the specific implementation manner of the present application. Each embodiment can be combined and cross-referenced with each other on the premise of not conflicting. An embodiment of the present application relates to a winding control method based on the prediction of the thickness rebound of the pole roll, which is applied to a winding control system based on the prediction of the thickness rebound of the pole roll. Generally speaking, during the production process of lithium batteries, the processes of rolling and winding are both involved. Rolling refers to the rolling process of the pole piece composed of a current collector (such as copper foil or aluminum foil) and the active material coating applied thereon to obtain a pole roll. The pole roll after rolling will be put into a winding machine for winding to wind out the battery core. The solution of the present application is directed to the pole roll about to enter the winding process (the to-be-wound pole roll). The following will specifically describe the implementation details of the winding control method based on the prediction of the thickness rebound of the pole roll in this embodiment. The following content is only provided for convenience of understanding and is not necessary for implementing this solution.

[0018] The specific process of the winding control method based on the prediction of the thickness rebound of the pole roll provided in this embodiment is as follows Figure 1 shown, and specifically includes the following steps:

[0019] Step 101: Obtain the current device data of the target device for rolling the to-be-wound pole roll and the current measured thickness of the to-be-wound pole roll after rolling.

[0020] Specifically, the polar roll to be wound in this embodiment refers to the negative electrode polar roll that has been roll-pressed and is about to enter the winding machine for winding. Obtaining the current device data means obtaining the relevant characteristic data of the target device for roll-pressing the polar roll to be wound, such as the model of the target device, roll-pressing parameters (such as pressure, speed, temperature, etc.), the operating state of the target device, and other parameters related to the subsequent thickness rebound. This embodiment does not make specific limitations on this. After obtaining the current device data of the target device, synchronously obtain the current measured thickness of the polar roll to be wound after roll-pressing. It should be noted that the thickness in this embodiment refers to the thickness of the electrode sheet in the polar roll. For example, the current measured thickness refers to the instantaneous thickness of the electrode sheet that makes up the polar roll to be wound measured by the thickness gauge after roll-pressing the polar roll to be wound, and the predicted rebound thickness refers to the difference between the actual thickness and the current measured thickness when the polar roll to be wound is being wound. The current measured thickness can be the average thickness of the polar roll to be wound within a preset length interval. The preset length interval can be the first 100 - 200 meters before the polar roll to be wound enters the winding machine, or it can also be the measured thickness at any position. This embodiment does not make specific limitations on this.

[0021] Step 102: Input the current device data into the trained thickness rebound prediction model to obtain the predicted rebound thickness of the polar roll to be wound.

[0022] Specifically, after obtaining the current device data of the target device, input it into a pre-trained thickness rebound prediction model. The trained thickness rebound prediction model in this embodiment is trained based on a large amount of historical data and machine learning algorithms, and can predict the thickness rebound situation of the polar roll after roll-pressing according to the input current device data. Among them, the training of the thickness rebound prediction model can be carried out by various algorithms, such as neural network, support vector machine, etc. This embodiment does not make specific limitations on this.

[0023] To achieve accurate prediction of the thickness rebound of the polar roll, a thickness rebound prediction model needs to be trained. In one example, as Figure 2 shown, the steps of training the thickness rebound prediction model can include:

[0024] Step 201: Obtain the device data of the device for roll-pressing the sample polar roll, the measured thickness of the sample polar roll after roll-pressing, and the actual thickness used at the winding location of the sample polar roll.

[0025] Specifically, a sample anode roll refers to the anode roll used as a sample for data collection during model training. Obtain the equipment data of the equipment for rolling the sample anode roll. These equipment data refer to the equipment operating parameters related to the thickness rebound of the anode roll, including but not limited to rolling force, in-out tension, roll gap, and transfer time, etc. Among them, the rolling force is the pressure applied to the anode roll during the rolling process, which has a direct impact on the thickness of the anode roll. Different rolling forces will result in different thickness changes, so it is necessary to record the magnitude of the rolling force; the in-out tension refers to the tension of the anode roll when entering and leaving the rolling equipment. The magnitude of the tension will affect the tension degree of the anode roll, thereby affecting the rolling effect and the thickness change of the anode roll; the roll gap refers to the gap between the two rolls in the rolling equipment. The size of the roll gap determines the compression degree of the anode roll during the rolling process and has an important impact on the thickness of the anode roll; the transfer time refers to the time interval between the output of the anode roll from the rolling equipment to the winding equipment. The length of the transfer time will affect the thickness rebound of the anode roll.

[0026] After obtaining the equipment data, since the thickness gauge will measure the thickness of the anode roll after the rolling process, it is also necessary to obtain the measured thickness of the sample anode roll after rolling and the actual thickness used at the winding of the sample anode roll. Among them, the measured thickness is the thickness data obtained through actual measurement, reflecting the immediate state of the anode roll after rolling; the actual thickness is the thickness data obtained during the actual winding process, reflecting the actual state of the anode roll during winding. It should be noted that the thickness in this embodiment refers to the thickness of the electrode sheet that makes up the anode roll, and will not be elaborated further hereinafter.

[0027] However, due to the limitations of the production site and equipment, the method of manually measuring the thickness usually cannot obtain the actual thickness data in batches when the rolled anode roll reaches the winding, and it is time-consuming and laborious. There will also be errors in manual measurement. In order to improve accuracy and efficiency, we can calculate the actual thickness used at the winding by inversely deducing the misalignment data of the cell tab of the corresponding sample anode roll during winding production, so as to indirectly obtain the batch rebound data.

[0028] Therefore, in one example, as Figure 3 shown, the actual thickness used at the winding of the sample anode roll obtained in step 201 can be carried out by the following method:

[0029] Step 301: Obtain the tab misalignment amount of the sample anode roll after winding. The tab misalignment amount is the tab misalignment amount of the sample anode roll for winding out one cell.

[0030] Step 302: Determine the actual thickness of the sample anode roll according to the tab misalignment amount and the preset calibration coefficient of the machine for winding the sample anode roll.

[0031] Specifically, the tab misalignment amount generally refers to the deviation between the actual position and the theoretical position of the tab during the winding process. This deviation is usually caused by the uneven thickness or rebound of the electrode roll. The tab misalignment amount of the sample electrode roll obtained after winding can be measured by high-precision detection equipment, such as a vision detection system or a laser measurement device, etc. This embodiment does not make specific limitations on this. The tab misalignment amount in this embodiment refers to the tab misalignment amount when winding out a single battery cell. The preset calibration coefficient of the winding machine reflects the influence of the characteristics of the winding equipment and process parameters on the thickness of the electrode roll. This coefficient can be set manually in advance and can be calibrated and adjusted through experiments or historical data. This embodiment does not make specific limitations on this. According to the tab misalignment amount and the preset calibration coefficient, the actual thickness of the sample electrode roll can be determined, where the actual thickness can be the average thickness of each battery cell length of the sample electrode roll after winding.

[0032] Taking the electrode roll A as an example, the tab misalignment amount (denoted as D) is directly related to the actual thickness used during winding (denoted as H1). The relationship is established through a calibration experiment: D = k·(H1 - H2), where k is the preset calibration coefficient of the machine for winding the current electrode roll A, and H2 is the measured thickness of the electrode roll A.

[0033] In actual production, by measuring the tab misalignment amount D of the battery cell, H1 = H2 + D / k is deduced, and then the rebound amount ΔH = H2 - H1 is calculated.

[0034] In this embodiment, by measuring the tab misalignment amount, the thickness rebound situation of the electrode roll can be indirectly deduced. By directly using the method from the end (tab misalignment) to the end (predicting the rebound to eliminate the tab misalignment), the corresponding rebound can be deduced from the misalignment, which can include possible errors to the greatest extent. Such rebound data is directly related to the misalignment, which is more conducive to the prediction model to play a role. Thus, accurate data support is provided for the thickness rebound prediction model, which helps to improve the prediction accuracy of the model, effectively solve the problem of inaccurate data sources caused by the thickness rebound of the negative electrode, and improve the production quality and reliability of the battery.

[0035] Step 202: Calculate the rebound amount based on the actual thickness and the measured thickness to obtain a training sample set corresponding to the equipment data and the rebound amount.

[0036] Specifically, according to the difference between the true thickness and the measured thickness of each sample pole roll, the rebound amount of each sample pole roll is calculated. The rebound amount refers to the change in thickness of the pole roll from after rolling to the winding process, that is, the difference between the true thickness and the measured thickness. By calculating the rebound amount, the change in the thickness of the pole roll can be quantified, providing data support for model training. Associate the equipment data corresponding to each sample pole roll during rolling with the rebound amount corresponding to the sample pole roll to form a training sample set. Each sample includes the equipment data as input data and the rebound amount as output data. By constructing a large number of training samples, a rich data foundation can be provided for model training, improving the accuracy and generalization ability of the model.

[0037] Step 203: Use the equipment data as input data and the rebound amount as output data to train the thickness rebound prediction model.

[0038] Specifically, use the equipment data as input data and the rebound amount as output data to train the thickness rebound prediction model. Multiple machine learning algorithms can be used for training, such as neural networks, support vector machines, etc. During the training process, by adjusting the parameters of the model, the model can accurately predict the rebound thickness of the pole roll.

[0039] Considering that the pole ear misalignment caused by winding mainly occurs in several battery cells after the negative pole roll change, that is, at the head position of the wound negative pole, therefore, in one example, the specific steps of step 203 may include:

[0040] Use the equipment data when rolling the head of the sample pole roll as input data, and the corresponding rebound amount as output data, and use the random forest regression algorithm to train the thickness rebound prediction model;

[0041] Among them, the head of the sample pole roll is a part of the preset length that first enters the machine when the sample pole roll is wound.

[0042] Specifically, considering that the pole ear misalignment caused by winding mainly occurs in several battery cells after the negative pole roll change, in order to more accurately predict the rebound thickness, when using the training samples to train the thickness rebound prediction model, the data input into the model can be screened, and only the equipment data and the rebound amount corresponding to the head position of the sample pole roll entering the winding machine are used to train the thickness rebound prediction model. The head of the sample pole roll is a part of the preset length that first enters the machine when the sample pole roll is wound. Among them, the preset length can be the length of 3 - 5 battery cells, and this embodiment does not make specific limitations on this. The random forest regression algorithm is used to train the thickness rebound prediction model. The random forest regression algorithm is an ensemble learning algorithm that achieves the regression task by constructing multiple decision trees and averaging their prediction results.

[0043] Step 204: Validate the thickness rebound prediction model, and obtain the trained thickness rebound prediction model when the validation result meets the preset conditions.

[0044] Specifically, during the training process, it is necessary to validate the thickness rebound prediction model. Specifically, methods such as cross-validation and holdout validation can be used to evaluate the prediction performance of the thickness rebound prediction model. Only when the validation result meets the preset conditions can it be considered that the model training is completed, and the trained thickness rebound prediction model is obtained. The preset conditions can be that the prediction error is within a certain range or the mean absolute error meets the threshold conditions, etc. This embodiment does not make specific limitations on this.

[0045] In one example, the specific implementation steps of step 204 include:

[0046] Use the method of cross-validation of the training set to validate the thickness rebound prediction model with the training sample set, and calculate the mean absolute error as the validation result;

[0047] When the mean absolute error is less than the first threshold, obtain the trained thickness rebound prediction model.

[0048] Specifically, during the process of training the thickness rebound prediction model, by validating the model, the generalization ability of the model can be ensured, thereby improving the reliability and accuracy of the trained model. In this embodiment, the method of cross-validation of the training set is used to validate the thickness rebound prediction model, and the training sample set is used to validate the thickness rebound prediction model. The specific steps are as follows: randomly divide the training sample set into multiple subsets (for example, 10 subsets). Each time, select one subset as the validation set, and the remaining subsets as the training set for model training and validation. Repeat the above process until each subset has been used as the validation set once. During the validation process, calculate the difference between the predicted rebound amount and the actual rebound amount of each sample, take the absolute value of each difference, sum up all the absolute values, and divide by the number of samples to obtain the mean absolute error (Mean Absolute Error, MAE) during the validation process as the final validation result. Preset a preset condition, that is, the maximum allowable value of the mean absolute error (MAE), denoted as the first threshold. If the calculated mean absolute error is less than or equal to the first threshold, it is considered that the performance of the model meets the requirements, and the trained thickness rebound prediction model is obtained. This embodiment does not make specific limitations on the value of the first threshold.

[0049] Step 103: Control the winding of the to-be-wound pole coil according to the current measured thickness and the predicted rebound thickness.

[0050] Specifically, since the winding process generally sets the winding parameters based on the currently measured thickness, after obtaining the predicted rebound thickness, the winding process parameters can be adjusted according to the currently measured thickness and the predicted rebound thickness to ensure that the thickness change of the pole winding during the winding process is within a predictable range, thereby reducing the occurrence of defects such as pole ear misalignment.

[0051] Compared with the related technology, in this embodiment, by obtaining the characteristic data of the target device and the currently measured thickness of the pole winding to be wound, rich basic data is provided for the thickness rebound prediction. Inputting these data into the trained thickness rebound prediction model can accurately predict the thickness rebound situation of the pole winding during the winding process. Based on the predicted rebound thickness, the winding process is finely controlled, greatly reducing the pole ear misalignment problem caused by thickness rebound, thereby improving the quality of the wound battery cell, reducing the generation of defective products, and saving costs.

[0052] Another embodiment of the present application relates to a winding control based on the prediction of the thickness rebound of the pole winding. This embodiment is a supplement to the foregoing embodiment and specifically relates to the following content.

[0053] In one example, before step 101, the method of this embodiment further includes: uploading the roll number of the pole winding to be wound and the current device data of the target device during the rolling process of the pole winding to be wound to the database.

[0054] At this time, the current device data of the target device for rolling the pole winding to be wound obtained in step 101 includes:

[0055] Obtain the roll number of the pole winding to be wound, and obtain the current device data of the target device corresponding to the roll number from the database according to the roll number.

[0056] Specifically, during the rolling process, the roll number of the pole winding to be wound and the current device data of the target device are synchronously and associatively uploaded to the database. These data include but are not limited to rolling force, incoming and outgoing drawing tension, roll gap, and transfer time, etc. When obtaining the target current device data of the target device for rolling the pole winding to be wound, first obtain the roll number of the pole winding to be wound. Then, obtain the current device data of the target device corresponding to the roll number from the database according to the roll number. These data will be used to input into the trained thickness rebound prediction model later to obtain the predicted rebound thickness.

[0057] Obtaining the current device data of the target device corresponding to the roll number from the database can quickly and accurately obtain the required data. This improves the efficiency of data acquisition and reduces the data processing time.

[0058] In another example, before step 103, the method of this embodiment further includes: obtaining the target calibration coefficient of the target machine table for winding the pole winding to be wound.

[0059] At this time, as Figure 4 shown, the specific steps of step 103 include:

[0060] Step 401: Calculate the actual compensation amount corresponding to the target machine according to the target calibration coefficient and the predicted rebound thickness, and use the actual compensation amount to compensate the current measured thickness.

[0061] Step 402: Control the winding of the to-be-wound pole roll on the target machine according to the thickness data of the to-be-wound pole roll after compensation.

[0062] Specifically, since the equipment parameters, working parameters, etc. of different winding machines are different, the same amount of rebound may result in different pole ear misalignments on the battery cores wound on different winding machines. To unify the relationship between the amount of rebound and the amount of pole ear misalignment, in the process of collecting real thickness data in the foregoing embodiment, calibration was performed for each machine to determine the calibration coefficient corresponding to each machine, so that there is a unified corresponding relationship between the amount of pole ear misalignment and the rebound thickness in the model training process. During the application of the model, after obtaining the predicted rebound thickness, the predicted rebound thickness is further operated with the target calibration coefficient of the target machine to obtain the actual compensation amount corresponding to the target machine. By compensating the current measured thickness with the actual compensation amount, more accurate thickness data can be obtained, providing reliable data support for subsequent winding control, which helps to improve the accuracy and quality of winding.

[0063] Through the above steps, the winding process of the to-be-wound pole roll can be effectively controlled, the possible rebound thickness during the winding of the to-be-wound pole roll can be pre-compensated, and the possibility of pole ear misalignment can be effectively reduced, thereby improving the production quality and reliability of the battery. The step division of the above various methods is only for clear description. During implementation, they can be combined into one step or some steps can be split into multiple steps. As long as the same logical relationship is included, it is within the protection scope of this patent; adding insignificant modifications to the algorithm or process or introducing insignificant designs, but not changing the core design of its algorithm and process, is within the protection scope of this patent.

[0064] Another embodiment of the present application relates to a winding control system based on pole roll thickness rebound prediction. The details of the model training device in this embodiment will be specifically described below. The following content is only the implementation details provided for convenient understanding and is not necessary for implementing this example. Figure 5 is a schematic structural diagram of the winding control system based on pole roll thickness rebound prediction described in this embodiment, including:

[0065] A data acquisition module 501, configured to acquire the target current device data of the target device for rolling the to-be-wound pole roll and the current measured thickness of the to-be-wound pole roll after rolling;

[0066] The bounce prediction module 502 is configured to input the target current device data into a trained thickness bounce prediction model to obtain the predicted bounce thickness of the to-be-wound pole roll.

[0067] The winding control module 503 is configured to control the to-be-wound pole roll to wind according to the current measured thickness and the predicted bounce thickness.

[0068] It is not difficult to find that this embodiment is a system embodiment corresponding to the foregoing method embodiments. The various implementation details recorded in the above method embodiments are still valid in this embodiment, and will not be elaborated herein.

[0069] Another embodiment of the present application relates to an electronic device, as Figure 6 shown, including: at least one processor 601; and a memory 602 communicatively connected to the at least one processor 601; wherein, the memory 602 stores instructions executable by the at least one processor 601, and the instructions are executed by the at least one processor 601 to enable the at least one processor 601 to execute the model training methods in the above embodiments.

[0070] Wherein, the memory and the processor are connected by a bus. The bus may include any number of interconnected buses and bridges, and the bus connects various circuits of one or more processors and memories together. The bus may also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art, and therefore will not be further described herein. The bus interface provides an interface between the bus and the transceiver. The transceiver may be an element or multiple elements, such as multiple receivers and transmitters, and provides a unit for communicating with various other devices on the transmission medium. The data processed by the processor is transmitted over the wireless medium through the antenna. Further, the antenna also receives data and transmits the data to the processor.

[0071] The processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interface, voltage regulation, power management, and other control functions. The memory can be used to store data used by the processor when executing operations.

[0072] Another embodiment of the present application relates to a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the method embodiments described above are implemented.

[0073] That is, those skilled in the art can understand that all or part of the steps in the methods of the above embodiments can be completed by instructing relevant hardware through a program. The program is stored in a storage medium, including several instructions to enable a device (which can be a single-chip microcomputer, a chip, etc.) or a processor to execute all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs.

[0074] Those of ordinary skill in the art can understand that the above embodiments are specific embodiments for implementing the present application. In actual applications, various changes can be made to them in form and details without departing from the spirit and scope of the present application.

Claims

1. A winding control method based on prediction of the rebound of the coil thickness, characterized in that, Including: Obtaining the current device data of the target device for roll pressing the to-be-wound pole roll and the current measured thickness of the to-be-wound pole roll after roll pressing; Inputting the current device data into the trained thickness rebound prediction model to obtain the predicted rebound thickness of the to-be-wound pole roll; Controlling the winding of the to-be-wound pole roll according to the current measured thickness and the predicted rebound thickness.

2. The method according to claim 1, characterized in that, The steps of training the thickness rebound prediction model include: Obtaining the device data of the device for roll pressing the sample pole roll, the measured thickness of the sample pole roll after roll pressing, and the true thickness used at the winding position of the sample pole roll; Calculating the rebound amount according to the true thickness and the measured thickness to obtain a training sample set corresponding to the device data and the rebound amount; Using the device data as input data and the rebound amount as output data to train the thickness rebound prediction model; Verifying the thickness rebound prediction model, and obtaining the trained thickness rebound prediction model when the verification result meets the preset conditions; Wherein, the device data includes any one or any combination of rolling force, inlet and outlet drawing tension, roll gap, and transfer time.

3. The method according to claim 2, wherein Obtaining the true thickness used at the winding position of the sample pole roll includes: Obtaining the pole ear misalignment amount of the sample pole roll after winding, and the pole ear misalignment amount is the pole ear misalignment amount of the sample pole roll when winding out one battery cell; Determining the true thickness of the sample pole roll according to the pole ear misalignment amount and the preset calibration coefficient of the machine for winding the sample pole roll.

4. The method according to claim 2, wherein The step of using the device data as input data and the rebound amount as output data to train the thickness rebound prediction model includes: Using the device data when roll pressing the head of the sample pole roll as the input data, and the corresponding rebound amount as the output data, and training the thickness rebound prediction model by using the random forest regression algorithm; Wherein, the head of the sample pole roll is a preset length part that first enters the machine when the sample pole roll is wound.

5. The method according to claim 4, characterized in that, Verifying the thickness rebound prediction model, and obtaining the trained thickness rebound prediction model when the verification result meets the preset conditions includes: Verifying the thickness rebound prediction model by using the training sample set by using the method of training set cross-validation, and calculating the mean absolute error as the verification result; When the mean absolute error is less than the first threshold, obtaining the trained thickness rebound prediction model.

6. The method according to claim 1, wherein Before obtaining the current device data of the target device for roll pressing the to-be-wound pole roll, the method further includes: Uploading the roll number of the to-be-wound pole roll and the current device data of the target device to the database during the roll pressing process of the to-be-wound pole roll; Obtaining the current device data of the target device for roll pressing the to-be-wound pole roll includes: Obtaining the roll number of the to-be-wound pole roll, and obtaining the corresponding current device data from the database according to the roll number.

7. The method according to claim 1, characterized in that, Before controlling the winding of the to-be-wound pole roll according to the current measured thickness and the predicted rebound thickness, it further includes: Obtain the target calibration coefficient of the target machine to wind the to-be-wound pole roll; The controlling the to-be-wound pole roll to wind according to the current measured thickness and the predicted rebound thickness includes: Calculate the actual compensation amount corresponding to the target machine according to the target calibration coefficient and the predicted rebound thickness, and use the actual compensation amount to compensate the current measured thickness; Control the to-be-wound pole roll to wind on the target machine according to the thickness data of the to-be-wound pole roll after compensation.

8. A winding control system based on prediction of the rebound of the coil thickness, characterized in that, It includes: A data acquisition module, configured to acquire the target current device data of the target device for rolling the to-be-wound pole roll and the current measured thickness of the to-be-wound pole roll after rolling; A rebound prediction module, configured to input the target current device data into a trained thickness rebound prediction model to obtain the predicted rebound thickness of the to-be-wound pole roll; A winding control module, configured to control the to-be-wound pole roll to wind according to the current measured thickness and the predicted rebound thickness.

9. An electronic device, characterized in that, It includes: At least one processor; And, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the winding control method based on pole roll thickness rebound prediction according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the winding control method based on pole roll thickness rebound prediction according to any one of claims 1 to 7.

Images