Lithium battery diaphragm process parameter self-correction method and system based on big data analysis

Abstract

The invention provides a lithium battery diaphragm process parameter self-correction method and system based on big data analysis, and the method comprises the following steps: in a preparation process of a lithium battery diaphragm, obtaining thickness values measured at a plurality of continuous time nodes after a raw material is stretched; a thickness data set corresponding to each time node is constructed, the thickness data set comprises thickness data used for representing the thickness condition of each region of the raw material, and the thickness data is generated according to the thickness values; comparing each group of thickness data set with other thickness data sets of adjacent time nodes, and determining a target thickness data set in the thickness data sets according to a data difference condition; and correcting the process parameters of the lithium battery diaphragm according to the target thickness data set. The method has the beneficial effect that the process parameter correction in the manufacturing process of the lithium battery diaphragm can be more accurate.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of lithium battery preparation, in particular to a lithium battery diaphragm process parameter self-correction method and system based on big data analysis. BACKGROUND

[0002] Under the background of rapid development of lithium battery industry, the quality of lithium battery diaphragm, as a core component of the battery, directly affects the safety, cycle life and energy density of the battery. In the diaphragm production process, the raw material stretching process is the core link of determining the key performance of diaphragm thickness uniformity, porosity, etc. This link involves the coordinated control of multiple parameters such as stretching temperature, rate, clamping force, etc. However, there is a technical problem of inaccurate process parameter correction in the current lithium battery diaphragm manufacturing process.

[0003] Traditional diaphragm production relies on manual experience. The technician mainly judges whether the thickness meets the standard by periodically sampling diaphragm samples and combining his own experience, and then adjusts the process parameters. This method has obvious disadvantages: for example, the heating device, stretching device, etc. need to run continuously for a period of time to reach stability, and when manual detection is performed, only thickness data at a single time node is collected to control the equipment, while the working state of the production equipment in the lithium battery production process is ignored. Fluctuations and changes exist, which makes the correction and adjustment of the process parameters of the equipment not accurate enough. SUMMARY

[0004] Embodiments of the present application provide a lithium battery diaphragm process parameter self-correction method and system based on big data analysis, which can improve the technical problem of inaccurate process parameter correction in the lithium battery diaphragm manufacturing process.

[0005] In a first aspect, embodiments of the present application provide a lithium battery diaphragm process parameter self-correction method based on big data analysis, comprising the following steps: In the preparation process of the lithium battery diaphragm, the thickness values measured at multiple consecutive time nodes after stretching of the raw material are obtained; A thickness data set corresponding to each of the time nodes is constructed, wherein the thickness data set includes thickness data for representing the thickness condition of each region of the raw material, and the thickness data is generated according to the thickness values; Compare each of the thickness data sets with other thickness data sets of adjacent time nodes, and determine a target thickness data set in the thickness data set according to the data difference; Correct the process parameters of the lithium battery diaphragm according to the target thickness data set.

[0006] In an embodiment, the process parameters include clamping force of a clamping device and rotating speed of a stretching device, and the correction of the process parameters of the lithium battery diaphragm according to the target thickness data set comprises: determining a first thickness variation of the raw material in a transverse direction and a second thickness variation of the raw material in a longitudinal direction according to the target thickness data set; correcting the clamping force of the clamping device and the rotating speed of the stretching device in combination with the first thickness variation and the second thickness variation.

[0007] In an embodiment, the first thickness variation includes a transverse thickness data deviation value of two adjacent areas of the raw material in the transverse direction, the second thickness variation includes a longitudinal thickness data deviation value of two adjacent areas of the raw material in the longitudinal direction, and the correcting the clamping force of the clamping device and the rotating speed of the stretching device in combination with the first thickness variation and the second thickness variation includes: if the transverse thickness data deviation value is greater than a preset first deviation threshold value and the longitudinal thickness data deviation value is less than or equal to a preset second deviation threshold value, correcting the clamping force of the clamping device according to the transverse thickness data deviation value; if the longitudinal thickness data deviation value is greater than the second deviation threshold value and the transverse thickness data deviation value is less than or equal to the first deviation threshold value, correcting the rotating speed of the stretching device according to the longitudinal thickness data deviation value; if the transverse thickness data deviation value is greater than the first deviation threshold value and the longitudinal thickness data deviation value is greater than the second deviation threshold value, correcting the clamping force of the clamping device and the rotating speed of the stretching device in combination with the transverse thickness data deviation value and the longitudinal thickness data deviation value.

[0008] In an embodiment, the correcting the clamping force of the clamping device according to the transverse thickness data deviation value includes: locating all thickness difference abnormal areas of the raw material according to the transverse thickness data deviation value; calculating abnormal area distances between all the thickness difference abnormal areas; determining a clamping force initial value according to the transverse thickness data deviation value and by using regression analysis; correcting the clamping force initial value according to the abnormal area distances to obtain a clamping force target value; correcting the clamping force of the clamping device according to the clamping force target value.

[0009] In an embodiment, the correcting the clamping force of the clamping device and the rotating speed of the stretching device in combination with the transverse thickness data deviation value and the longitudinal thickness data deviation value includes: inputting the lateral thickness data deviation value and the longitudinal thickness data deviation value into a pre-constructed deviation coupling correction model, outputting a clamping force correction amplitude and a rotating speed correction amplitude by the deviation coupling correction model, the deviation coupling correction model being constructed based on a gradient boosting regression tree; correcting the clamping force of the clamping device and the rotating speed of the stretching device according to the clamping force correction amplitude and the rotating speed correction amplitude respectively.

[0010] In an embodiment, the comparing each of the thickness data sets with other thickness data sets of adjacent time nodes and determining a target thickness data set in the thickness data sets according to data difference includes: determining a distance value of thickness data in each of the thickness data sets and thickness data in other thickness data sets of adjacent time nodes in a data space, wherein the distance value includes Euclidean distance or Manhattan distance; determining the target thickness data set according to the distance value.

[0011] In an embodiment, the thickness data includes an average value and a dispersion degree value of the thickness value in each of the regions, and the distance value includes a first distance value calculated by each of the average values corresponding to each of the regions in each of the thickness data sets and other thickness data sets of adjacent time nodes, and a second distance value calculated by the dispersion degree value; The determining the target thickness data set according to the distance value includes: determining a distance comprehensive index according to the first distance value and the second distance value, wherein a determination formula of the distance comprehensive index is: D = w 1 D 1 +w 2 D 2 D1 is the first distance value, D2 is the second distance value, w1 is a first preset weight, w2 is a second preset weight, and w2 is greater than w1; determining the target thickness data set according to the distance comprehensive index.

[0012] In an embodiment, the distance comprehensive index includes a first distance comprehensive index of the thickness data set of the current time node relative to a thickness data set of a previous time node, and a second distance comprehensive index of the thickness data set of the current time node relative to a thickness data set of a next time node. When the first distance comprehensive index and the second distance comprehensive index are both less than or equal to a preset index threshold, the thickness data set of the current time node is determined as the target thickness data set.

[0013] In an embodiment, the correcting the process parameters of the lithium battery separator according to the target thickness data set comprises: According to the average value corresponding to each region in the target thickness data set, the thickness variation of the raw material along the transverse and longitudinal directions is determined. When the thickness variation represents that the thickness of the raw material uniformly changes in the same direction, whether the variation of the dispersion degree value of each region in the same direction is adapted to the thickness variation is compared; If yes, the equipment needing process parameter correction is located according to the thickness variation and the dispersion degree value, and the process parameter is corrected; If no, the step of determining the target thickness data set is re-performed.

[0014] In a second aspect, embodiments of the present application provide a lithium battery separator process parameter self-correction system based on big data analysis, comprising: One or more processors; Memory; And one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the processor to implement the above method.

[0015] The beneficial effects of embodiments of the present application are: In embodiments of the present application, when the lithium battery separator is prepared, a large number of thickness values reflecting the thickness of the stretched raw material are obtained at a plurality of continuous time nodes after the raw material is stretched, and then the thickness value data of each time node is processed to obtain a thickness data set, which includes thickness data reflecting the thickness of each region, and the thickness data is generated from the thickness value to reduce the data amount. By comparing the data difference of the thickness data of adjacent time nodes, whether the equipment operation reaches a relatively stable state is determined, and then the target thickness data set in the stable state is determined for data analysis, and the process parameters of the lithium battery separator are corrected, so that the parameter correction is more accurate and fast, the labor and material costs are saved, and misjudgment is prevented. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments description. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0017] Figure 1 is a flowchart of the lithium battery diaphragm process parameter self-correction method based on big data analysis provided by the embodiments of the present application Figure 1 ; Figure 2 is a flowchart of the lithium battery diaphragm process parameter self-correction method based on big data analysis provided by the embodiments of the present application Figure 2 ; Figure 3 is a flowchart of the lithium battery diaphragm process parameter self-correction method based on big data analysis provided by the embodiments of the present application Figure 3 ; Figure 4 is a structural block diagram of the lithium battery diaphragm process parameter self-correction system based on big data analysis provided by the embodiments of the present application. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present application will be described clearly and completely in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0019] In the description of the present application, it should be understood that the terms "first", "second" are only used for description purpose, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features with "first", "second" can explicitly or implicitly include one or more features. In the description of the present application, the meaning of "multiple" is two or more, unless otherwise specifically limited.

[0020] In this application, the word "exemplary" is used to mean "serving as an example, instance, or illustration." Any implementation described as "exemplary" in this application is not necessarily to be construed as preferred or advantageous over other implementations. The following description is presented to enable any person skilled in the art to make and use the application. In the following description, for purposes of explanation, specific details are set forth. It is apparent to those skilled in the art that the present application can be practiced without the specific details presented. In other instances, well-known structures and processes are not shown in detail to avoid obscuring this application. Thus, the present application is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features presented.

[0021] To solve the problem of inaccurate process parameter correction in the manufacturing process of lithium battery separator, an embodiment of the present application provides a lithium battery separator process parameter self-correction method based on big data analysis, please refer to Figure 1 , Figure 1 is a flowchart of a lithium battery separator process parameter self-correction method based on big data analysis provided by an embodiment of the present application. The lithium battery separator process parameter self-correction method based on big data analysis includes the following steps: Step 100: In the preparation process of the lithium battery separator, the thickness values measured at multiple continuous time nodes after stretching of the raw material are obtained.

[0022] In the preparation process of the lithium battery separator, the stretching equipment continuously stretches the raw material. For example, in the wet production process of the lithium battery separator, the stretching includes longitudinal stretching and transverse stretching of the raw material film. In longitudinal stretching, the film is stretched after heating by the pressure of the stretching roller and the speed ratio of the front and rear rollers to make the molecular chain longitudinally oriented to achieve the specified thickness. The transverse stretching is also similar to the operation. After the longitudinal stretching of the raw material film, the film is stretched transversely by the pressure of the stretching roller after heating.

[0023] In the embodiment of the present application, the thickness values of the stretched raw material are obtained continuously multiple times. High-precision equipment such as laser thickness gauge and capacitance thickness gauge can be used to continuously scan and collect real-time thickness data of the separator. A large number of thickness values are collected at each time node to reflect the thickness of the stretched raw material in the current collection area.

[0024] Step 200: Construct a thickness data set corresponding to each time node, wherein the thickness data set includes thickness data for representing the thickness of each region of the raw material, and the thickness data is generated according to the thickness value; It can be understood that for a piece of stretched raw material, the thickness value collected includes data of multiple points. In the embodiment of the present application, for a large amount of thickness value data, a thickness data set can be generated from the thickness values of multiple points to represent the overall thickness of the coverage area of these points, thereby reducing the amount of data while accurately representing the thickness of the raw material.

[0025] For the stretching preparation process of the raw material, it may generally face the thickness deviation of the center area, the edge area and the transition area between the center and the edge. Therefore, for the division of the raw material area, it can be divided according to the center, the transition and the edge, so that the thickness data can represent the overall thickness of these areas.

[0026] For the generation of thickness data, the original measurement value, i.e. the thickness value, can be derived, for example, by calculating the average value of the thickness value to represent the overall thickness of the area and reduce the amount of data for subsequent comparison.

[0027] Step 300: comparing each set of thickness data set with other thickness data sets of adjacent time nodes, and determining a target thickness data set in the thickness data set according to the data difference; Step 400: correcting the process parameters of the lithium battery separator according to the target thickness data set.

[0028] It can be understood that for the continuous stretching preparation of the lithium battery separator, after stretching for a certain period of time, the equipment is in a relatively stable working stage, for example, the heating of the heating equipment reaches a relatively stable stage, and the operation of the stretching roller reaches a relatively stable stage. At this time, the thickness parameters of the stretched raw material can better match the actual process parameters. Therefore, in the embodiment of the present application, for a plurality of sets of thickness data sets measured in sequence according to time nodes, by comparing the thickness data set of each time node with the thickness data set of the adjacent time node, the difference between the data sets is judged to indirectly judge whether the operation of each device reaches stability. In this way, the thickness data set of the part of the raw material stretched when the device reaches stable operation is determined as the target thickness data set. Then, the target thickness data set obtained at present matches the process parameters of the current device. When the target thickness data set is abnormal, it can more accurately indicate that some process parameters have problems. Therefore, by analyzing the target thickness data set, more accurate correction of the process parameters of the lithium battery separator can be realized, and a more qualified lithium battery separator can be produced.

[0029] Thus, in the lithium battery separator process parameter self-correction method based on big data analysis in the embodiments of the present application, a large number of thickness values reflecting the thickness of the stretched raw material are obtained at a plurality of continuous time nodes after the raw material is stretched in the preparation process of the lithium battery separator, and then the thickness value data of each time node is processed to obtain a thickness data set, which includes thickness data reflecting the thickness of each region. The thickness data is generated from the thickness value to reduce the data volume. By comparing the data differences of the thickness data of adjacent time nodes, it is determined whether the equipment operation reaches a relatively stable state, so as to determine the target thickness data set in the stable state for data analysis, and then the process parameters of the lithium battery separator are corrected to make the parameter correction more accurate and fast, save manpower and material costs, and prevent misjudgment.

[0030] In addition, in the embodiments of the present application, the target thickness data set meeting the needs is determined in real time by obtaining the thickness data of a plurality of continuous time nodes. After the process parameters are corrected, the material meeting the needs can be directly produced. The judgment process has higher real-time and accuracy, and compared with the judgment by manpower and experience, the raw material cost consumed in the early stage can be saved as much as possible, thereby reducing the overall material cost in the production process.

[0031] Referring to Figure 2 In some embodiments of the present application, the process parameters include the clamping force of the clamping device and the rotating speed of the stretching device. The correction of the process parameters of the lithium battery separator according to the target thickness data set includes: Step 411: determining the first thickness variation of the raw material from the center to the edge in the transverse direction and the second thickness variation of the raw material in the longitudinal direction according to the target thickness data set; Step 412: correcting the clamping force of the clamping device and the rotating speed of the stretching device in combination with the first thickness variation and the second thickness variation.

[0032] In this embodiment, for the stretching of the raw material, the transverse direction refers to the direction parallel to the horizontal plane and perpendicular to the transmission direction of the raw material preparation, and the longitudinal direction refers to the transmission direction of the raw material preparation. In the stretching process of the raw material, the raw material is clamped by the clamping devices on both sides and stretched by the stretching rollers after heating. For the transverse direction of the raw material, i.e., the width direction of the raw material, there may be a first thickness variation from the edge to the center region. When the thickness variation meets the requirements, it is within the allowable error range for preparation. When the thickness variation is too large and exceeds the expectation, it indicates that the current preparation does not meet the requirements.

[0033] Generally, the edge and the center thickness change with a large difference is related to the clamping force of the clamping device, such as the clamping force of the clamp is large, which causes the edge and the center to have a large deviation in the stretching amount, thereby causing a large thickness change. Therefore, in the embodiments of the present application, the clamping device and the corresponding clamping force that need to be corrected are determined according to the first thickness change condition, such as the thickness difference and the position where the thickness difference occurs, so as to realize accurate adjustment.

[0034] For the embodiments of the present application, the thickness data in the target thickness data set is obtained by actual thickness value data processing, so for the first thickness change condition of the raw material from the center to the edge in the transverse direction determined by the target thickness data set, the difference between the data is more a magnitude of the thickness difference than an actual thickness difference, so when setting the error value for comparison and judgment with the first thickness change condition, adaptive adjustment can be made according to the actual situation and experience, which will not be described here.

[0035] In addition, in the preparation process of the lithium battery separator, the raw material may have a periodic or random fluctuation in thickness along the length direction in the longitudinal direction. In the longitudinal direction, the thickness fluctuation exceeding the error does not meet the production requirements.

[0036] For example, for periodic thickness fluctuation, it is usually related to the periodic and regular deviation of the stretching device speed, such as the encoder signal of the driving motor is disturbed to cause periodic loss of pulse signal. The encoder pulse signal can be analyzed by algorithm to eliminate abnormal pulses and correct the speed calculation result to ensure the accuracy of the speed feedback, thereby realizing automatic correction of the stretching device speed and further adjusting the thickness of the raw material obtained by subsequent stretching to make the second thickness change condition meet the error.

[0037] In some embodiments of the present application, the lithium battery separator process parameter self-correction method based on big data analysis, characterized in that the first thickness change condition includes the transverse thickness data deviation value of the raw material in the transverse direction of the adjacent two regions, and the second thickness change condition includes the longitudinal thickness data deviation value of the raw material in the longitudinal direction of the adjacent two regions, and the correction of the clamping force of the clamping device and the speed of the stretching device based on the first thickness change and the second thickness change condition includes: If the transverse thickness data deviation value is greater than the preset first deviation threshold value, and the longitudinal thickness data deviation value is less than or equal to the preset second deviation threshold value, then the clamping force of the clamping device is corrected according to the transverse thickness data deviation value; If the longitudinal thickness data deviation value is greater than the second deviation threshold value, and the transverse thickness data deviation value is less than or equal to the first deviation threshold value, then the speed of the stretching device is corrected according to the longitudinal thickness data deviation value; If the lateral thickness data deviation value is greater than the first deviation threshold value and the longitudinal thickness data deviation value is greater than the second deviation threshold value, the clamping force of the clamping device and the rotation speed of the stretching device are simultaneously corrected in combination with the lateral thickness data deviation value and the longitudinal thickness data deviation value.

[0038] In the present embodiment, since the lateral stretching step and the longitudinal stretching step of the lithium battery separator are not completely independent, the longitudinal stretching performed first changes the material state of the raw material, such as molecular orientation, local density, ductility, etc., and these material states also indirectly affect the subsequent lateral stretching. Therefore, when it is necessary to simultaneously correct the clamping force of the clamping device and the rotation speed of the stretching device, the synergistic effect of the two on the lateral thickness data deviation value and the longitudinal thickness data deviation value, especially the lateral thickness data deviation value, needs to be considered to prevent the lateral thickness data deviation value from being still too large due to overcorrection of the clamping force of the clamping device during the lateral stretching process. Therefore, if it is necessary to simultaneously correct the clamping force of the clamping device and the rotation speed of the stretching device, the synergistic effect of the two needs to be considered.

[0039] If the lateral thickness data deviation value is greater than the preset first deviation threshold value, and the longitudinal thickness data deviation value is less than or equal to the preset second deviation threshold value, it indicates that only the clamping force of the clamping device needs to be corrected to restore the lateral thickness data deviation value to normal, so the synergistic effect of correcting the clamping force of the clamping device and the rotation speed of the stretching device does not need to be considered, and only the rotation speed of the stretching device needs to be corrected according to the longitudinal thickness data deviation value. The specific correction steps include: regarding two adjacent regions with a lateral thickness data deviation value greater than the preset first deviation threshold value as thickness difference abnormal regions. After determining the thickness difference abnormal regions, the edge of the raw material can be used as a reference to record the vertical distance between the center point of each thickness difference abnormal region and the edge of the raw material. According to the vertical distance, the center point distance of each thickness difference abnormal region, i.e., the abnormal region distance, is further calculated. If two abnormal regions are close to each other, for example, the left and right sides of a thickness abnormal region are both thickness abnormal regions, the adjustment range of the clamping force of the clamping device at the positions of the three thickness abnormal regions needs to be appropriately reduced to reduce interference, because the clamping force has a certain diffusion range, and when it is adjacent to two abnormal regions, the force transmission will interfere with each other. The historical lateral thickness data deviation values and the corresponding historical clamping device clamping forces are obtained as sample data. The historical lateral thickness data deviation values are used as dependent variables, and the historical clamping device clamping forces are used as independent variables for nonlinear regression analysis to obtain a regression equation between the lateral thickness data deviation value and the clamping device clamping force. The lateral thickness data deviation value between the current thickness difference abnormal region and the adjacent thickness difference abnormal region is input into the regression equation to obtain the initial value of the clamping force of the clamping device. If the current thickness difference abnormal region has adjacent thickness abnormal regions, the initial value of the clamping force needs to be multiplied by a preset correction coefficient to obtain the target value of the clamping force, and the determination of the correction coefficient can be determined through multiple clamping device clamping force adjustment experiments. Finally, the clamping force of the clamping device corresponding to the thickness abnormal region is corrected to the target value of the clamping force.

[0040] In the preparation process of the lithium battery separator, for the stretching of the raw material, the stretching equipment usually includes a plurality of spaced-apart stretching rollers, different stretching rollers will finally affect the thickness of different areas of the raw material, when the longitudinal thickness data deviation value is greater than the preset deviation value, in addition to indicating that the thickness deviation of the adjacent two areas is large to make the preparation not meet the actual production demand, it also reflects to a certain extent that the speed of the stretching roller corresponding to the two areas has a problem, for example, when the thickness data of a certain adjacent area shows that the previous area is too thin and the next area is too thick, it is possible that the corresponding stretching roller speed is too fast or too slow, resulting in excessive or insufficient stretching of the raw material in that area. At this time, through the mapping relationship between the area and the stretching roller, the target stretching equipment causing the deviation can be quickly located, the stretching roller with abnormal speed can be accurately locked, and rapid adjustment can be realized, effectively improving the stability of the production process and the yield of the separator product, and avoiding the decline of production efficiency and waste of resources caused by long time of manual investigation. In the process of correcting the stretching equipment speed according to the longitudinal thickness data deviation value, the adjustment range of the longitudinal thickness data deviation value for correcting the stretching equipment speed can be determined by regression analysis or by establishing and training a neural network model. Taking regression analysis as an example, a regression model is established using historical longitudinal thickness data deviation values and stretching equipment speeds. By inputting the current longitudinal thickness data deviation value into the regression model, the target value of the stretching equipment speed can be directly obtained, and the stretching equipment speed can be corrected according to the target value of the stretching equipment speed.

[0041] If the transverse thickness data deviation value is greater than the first deviation threshold value, and the longitudinal thickness data deviation value is greater than the second deviation threshold value, it indicates that both the longitudinal stretching process and the transverse stretching process are abnormal during the stretching of the raw material. Generally, the raw material stretching process is first longitudinal stretching, and then transverse stretching. At this time, if the clamping force of the clamping equipment is corrected according to the transverse thickness data deviation value, and the speed of the stretching equipment is corrected according to the longitudinal thickness data deviation value, without considering the coupling between the clamping force of the clamping equipment and the speed of the stretching equipment, the speed of the stretching equipment may be corrected to the target value, but the raw material will change its stress state and microstructure when stretched longitudinally using the corrected speed of the stretching equipment. The subsequent transverse stretching process requires different clamping force of the clamping equipment, but since the correction of the clamping force of the clamping equipment is still based on the raw material without correction of the speed of the stretching equipment, it may further increase the transverse thickness data deviation value. Therefore, the clamping force of the clamping equipment and the speed of the stretching equipment need to be corrected in coordination with the transverse thickness data deviation value and the longitudinal thickness data deviation value. The correction amplitude of the clamping force and the correction amplitude of the speed can be predicted based on gradient boosting regression trees, and the clamping force and the speed of the stretching equipment can be corrected according to the correction amplitude of the clamping force and the correction amplitude of the speed, respectively.

[0042] Gradient boosting regression tree is a regression model based on the idea of ensemble learning. Its core principle is to gradually build multiple decision trees (base learners) and let each new tree focus on fitting the residual (error) between the prediction results of all previous trees and the true values, and continuously optimize the residual using the idea of gradient descent. Finally, the prediction results of all trees are added up as the final output. When processing regression problems, gradient boosting regression tree can not only retain the ability of decision trees to capture nonlinear relationships, but also reduce the overfitting risk of a single tree through ensemble strategies, thereby achieving high prediction accuracy. In the step-by-step stretching process of lithium battery separators, gradient boosting regression tree can efficiently handle nonlinear relationships, and the clamping force of the clamping device, the speed of the stretching device, the lateral thickness data deviation value, and the longitudinal thickness data deviation value. Gradient boosting regression tree does not need to manually preset the function form, and can automatically capture this nonlinear relationship through the segmented fitting of decision trees.

[0043] In some embodiments of the present application, correcting the clamping force of the clamping device according to the lateral thickness data deviation value includes: locating all thickness difference abnormal area regions of the raw material according to the lateral thickness data deviation value; calculating the abnormal area distance between all thickness difference abnormal area regions; determining the initial value of the clamping force according to the lateral thickness data deviation value and using regression analysis; correcting the clamping force of the clamping device according to the clamping force target value. correcting the clamping force of the clamping device according to the clamping force target value.

[0044] In this embodiment, the lateral thickness data deviation value refers to the difference between the lateral thickness data of two adjacent regions of the raw material in the lateral direction. When the lateral thickness data deviation value is greater than a preset first deviation threshold, it means that one of the two adjacent regions of the raw material in the lateral direction is too thick and the other is too thin, and both regions are thickness difference abnormal area regions. According to the lateral thickness data deviation value, the over-thick or over-thin region of the raw material is marked as a thickness difference abnormal area region. For the thickness difference abnormal area region, it may be caused by excessive clamping force or insufficient clamping force of the clamping device corresponding to the region, so the clamping force of the clamping device of the thickness difference abnormal area region needs to be corrected.

[0045] The lateral thickness data deviation value of the thickness difference abnormal area can be directly input into a preset neural network model to obtain a clamping force target value, wherein the preset neural network model is trained by a training data set and an initial neural network model, the training data set includes historical lateral thickness data deviation values and corresponding historical clamping device clamping forces, and the clamping device clamping force is corrected according to the clamping force target value. The preset neural network model is used to estimate the clamping force, which is trained by the initial neural network model, and the training data set includes historical lateral thickness data deviation values and corresponding historical clamping device clamping forces. The initial neural network model can be a multilayer perceptron or a convolutional neural network, and the network parameters are continuously adjusted by a back propagation algorithm. The historical lateral thickness data deviation values in the training data set are input into the model, the error between the predicted clamping force and the real historical clamping force is calculated, the gradient descent optimization algorithm is used to minimize the error, and the model is iteratively trained to accurately learn the complex mapping relationship between the thickness deviation and the clamping force. When the model reaches a low prediction error on the validation set, the training is completed. In actual application, the real-time lateral thickness data deviation value is input into the trained model, and the model can quickly output the corresponding clamping force target value, providing accurate basis for the clamping force adjustment of the diaphragm production equipment, thereby ensuring the uniformity of the diaphragm lateral thickness and the stability of the production process.

[0046] The neural network model can accurately output the clamping force target value, but the process of training the neural network model is relatively complex, requires a large amount of computing power, and requires a large amount of historical lateral thickness data deviation values and corresponding historical clamping device clamping forces, which is highly dependent on the amount and quality of data. Therefore, regression analysis can be used to calculate the clamping force target value, but considering that the accuracy of regression analysis is lower than that of the output results of the neural network model, in order to further improve the accuracy of regression analysis, the abnormal area distance is introduced to correct the clamping force initial value obtained by regression analysis to obtain a more accurate clamping force target value. By this method, the accuracy of the final result can be guaranteed while balancing the computing power.

[0047] Specifically, for two adjacent regions with a transverse thickness data deviation value greater than a preset first deviation threshold, the two adjacent regions are regarded as thickness difference abnormal regions. After determining the thickness difference abnormal regions, the center point of each thickness difference abnormal region can be recorded in the raw material between the raw material edge and the vertical distance of the raw material edge. According to the vertical distance, the center point distance of each thickness difference abnormal region, i.e., the abnormal region distance, is further calculated. If two abnormal regions are close to each other, for example, the left and right sides of a thickness abnormal region are both thickness abnormal regions, the clamping force has a certain diffusion range, and when the clamping force is adjacent to the two abnormal regions, the force conduction will superimpose or interfere with each other. Therefore, the adjustment range of the clamping force of the clamping device at the positions of the three thickness abnormal regions needs to be appropriately reduced to reduce the interference. The historical transverse thickness data deviation value and the corresponding historical clamping device clamping force are obtained as sample data, the historical transverse thickness data deviation value is taken as the dependent variable, and the historical clamping device clamping force is taken as the independent variable, and nonlinear regression analysis is performed. This is because the relationship between the transverse thickness data deviation value and the clamping force of the clamping device is not linear. When the clamping force of the clamping device is too low, the transverse thickness data deviation value decreases rapidly with the increase of the clamping force. When the clamping force of the clamping device reaches a reasonable range, the transverse thickness data deviation value changes slowly. Through quadratic regression fitting, a regression equation between the transverse thickness data deviation value and the clamping force of the clamping device is obtained. The transverse thickness data deviation value between the current thickness difference abnormal region and the adjacent thickness difference abnormal region is input into the regression equation to obtain the initial value of the clamping force of the clamping device. If there are still adjacent thickness abnormal regions in the current thickness difference abnormal region, for example, the thickness distribution of multiple continuous regions in a raw material is A region thin-B region thick-C region thick-D region thin, and the C region is adjacent to two thickness difference abnormal regions, i.e., the B region and the D region, the initial value of the clamping force of the C region needs to be multiplied by a preset correction coefficient to obtain the target value of the clamping force to compensate for the interference of the B region and the D region on the C region. The determination of the correction coefficient can be determined through multiple clamping force adjustment tests of the clamping device. Finally, the clamping force of the clamping device corresponding to the thickness abnormal region is corrected to the target value of the clamping force.

[0048] Through the above steps, the thickness of the raw material can be accurately controlled during the transverse stretching process, and the raw material is divided into regions, and the deviation of adjacent regions is analyzed, so that the clamping device that needs to be corrected can be quickly located, the correction process is more targeted, the thickness is accurately controlled, and quality assurance is provided for the subsequent production of lithium battery separators.

[0049] In some embodiments of the present application, the clamping force of the clamping device and the rotation speed of the stretching device are corrected in combination with the transverse thickness data deviation value and the longitudinal thickness data deviation value. The transverse thickness data deviation value and the longitudinal thickness data deviation value are input into a pre-constructed deviation coupling correction model, and a clamping force correction amplitude and a rotating speed correction amplitude are output by the deviation coupling correction model, the deviation coupling correction model being constructed based on a gradient boosting regression tree; The clamping force of the clamping device and the rotating speed of the stretching device are corrected according to the clamping force correction amplitude and the rotating speed correction amplitude respectively.

[0050] In the embodiment, the gradient boosting regression tree can process nonlinear relationships, capture feature interactions (such as the synergistic effect of the transverse thickness data deviation value and the longitudinal thickness data deviation value), and has strong robustness to noise in process data, is suitable for learning complex mapping relationships from historical data, and outputs continuous correction amplitudes, i.e., the clamping force correction amplitude and the rotating speed correction amplitude. A sufficient number of historical transverse thickness data deviation values, historical longitudinal thickness data deviation values, and corresponding historical clamping device clamping forces and historical stretching device rotating speeds are collected and integrated into a sample set. After data cleaning and normalization processing of the sample set, it is divided into a training set (used for model learning) and a test set (used for verifying the generalization ability) according to a certain proportion, for example, a 7:3 proportion, to ensure that the deviation distribution and process condition distribution of the two groups of samples are consistent (such as including high / medium / low deviation scenarios). Then, a deviation coupling correction model is constructed based on a gradient boosting regression tree. The deviation coupling correction model is a multi-output gradient boosting regression tree model. The training set is input into the deviation coupling correction model for iterative training, and multiple decision trees are generated by gradient boosting iteration, each tree fitting the prediction residual of the previous model, gradually reducing the error, and continuously optimizing model hyperparameters such as the number of trees, tree depth, and learning rate during the training process. The test set is used to verify the model error, and when the model error is less than a preset error threshold or reaches a preset maximum number of iterations, the model training is completed, and a trained deviation coupling correction model is obtained. The current transverse thickness data deviation value and the longitudinal thickness data deviation value are input into the trained deviation coupling correction model, and the clamping force correction amplitude and the rotating speed correction amplitude are output by the trained deviation coupling correction model. The clamping force of the clamping device and the rotating speed of the stretching device are corrected according to the clamping force correction amplitude and the rotating speed correction amplitude.

[0051] Gradient boosting regression tree is a regression model based on the idea of ensemble learning. Its core principle is to gradually build multiple decision trees (base learners) and let each new tree focus on fitting the residual (error) between the prediction results of all previous trees and the true values, and continuously optimize the residual using the idea of gradient descent. Finally, the prediction results of all trees are added up as the final output. When processing regression problems using gradient boosting regression trees, both the ability to capture nonlinear relationships and the risk of overfitting of a single tree can be reduced through ensemble strategies, thereby achieving high prediction accuracy. In the step-by-step stretching process of lithium battery separators, gradient boosting regression trees can efficiently handle nonlinear relationships, and the clamping force of the clamping device, the speed of the stretching device, the lateral thickness data deviation value, and the longitudinal thickness data deviation value. Gradient boosting regression trees do not require manual preset function forms and can automatically capture this nonlinear relationship through segmented fitting of decision trees.

[0052] Referring to Figure 3 In some embodiments of the present application, by comparing each set of thickness data set with other thickness data set of adjacent time node, the target thickness data set is determined in the thickness data set according to the data difference. Step 301: Determine the distance value of the thickness data in each thickness data set and the thickness data in the other thickness data set of the adjacent time node in the data space, wherein the distance value includes Euclidean distance or Manhattan distance. Step 302: Determine the target thickness data set according to the distance value.

[0053] In this embodiment, by comparing the corresponding data in the thickness data set, the data difference between the two data sets is determined by judging the distance value in the data space.

[0054] For example, for two thickness data sets of adjacent time nodes, the two data sets are A [ a 1, a 2,..., a n ] and B [ b 1, b 2,..., b n ]The calculation formula of the Euclidean distance in the data space is: ; That is, the square root of the sum of the difference values of the corresponding elements. The smaller the distance value, the smaller the data difference.

[0055] For these two data sets, the distance value can also be Manhattan distance. The calculation formula of the Manhattan distance in the data space is: ; i.e. the sum of the absolute values of the difference values of the corresponding elements, and the smaller the distance value is, the smaller the data difference is.

[0056] In some embodiments, determining the target thickness data set according to the distance value can include determining the data difference between the thickness data set of each time node and the thickness data set of the previous time node, and determining the thickness data set of the current time node as the target thickness data set when the determined distance value is less than a preset threshold value.

[0057] It can be understood that, for the stretched raw material in the embodiments of the present application, it is a plane, and when the data set is divided, it can be a data array of the thickness data in the divided area of each row and each column on the raw material plane.

[0058] In some embodiments of the present application, the thickness data includes the average value and the dispersion value of the thickness value in each area, and the distance value includes a first distance value calculated by each set of thickness data set and the average value corresponding to each area in the other thickness data set of the adjacent time node, and a second distance value calculated by the dispersion value. Determining the target thickness data set according to the distance value includes: Determining a distance comprehensive index according to the first distance value and the second distance value, wherein the determination formula of the distance comprehensive index is: D = w 1 D 1+ w 2 D 2 , D1 is the first distance value, D2 is the second distance value, w 1 is the first preset weight, w 2 is the second preset weight, w 2 is greater than w 1; Determining the target thickness data set according to the distance comprehensive index.

[0059] In the embodiment, in each thickness data set determined at each time node, the thickness data of each area pair has two types, one of which is thickness data representing the average thickness of the area. In the raw material stretching process, if the overall equipment has no major failure, the thickness of adjacent time nodes will not have major fluctuations under normal circumstances. If major fluctuations occur, the average value can also accurately indicate the overall thickness of the area. Thus, after a large number of thickness values are collected, the embodiment calculates the dispersion degree value in each group of data by taking the average value, and does not need to calculate the distance of each thickness value in the adjacent two groups of data one by one, thereby reducing the amount of data, which is more conducive to quickly comparing the differences between the data of adjacent time nodes. In addition, the dispersion degree value of the thickness value of the area is further compared, and the dispersion degree value can be a standard deviation or a variance, etc. The dispersion degree value can reflect the uniformity of the thickness of the area at the same time. The smaller the value, the closer the thickness of each point to the average value, and the better the thickness uniformity. Conversely, the larger the value, the more drastic the thickness fluctuation, and the worse the uniformity.

[0060] At adjacent two time nodes, the average value and the dispersion degree value can reflect the difference between the two groups of data. For example, the average values are similar, but the dispersion degree values differ greatly, indicating that there is a region with a large thickness fluctuation change in the stretching process, and the change continues. Only using the average value may lead to misjudgment, so combining the dispersion degree value makes the judgment more accurate. For the average value, the dispersion degree value is not large, indicating that the running state of the equipment may be a stable change process in the continuous stretching process. The thickness of each area of the raw material stretched at adjacent time nodes changes to some extent, causing the average value to change. However, the thickness fluctuation in the area at the same time node is relatively uniform, so the dispersion degree values of adjacent time nodes are similar. At this time, the conclusion of the difference between the two groups of data is mainly based on the change of the average value. It can be understood that when the average value and the dispersion degree value are similar, the difference between the two groups of data is relatively small.

[0061] In the embodiment, the spatial distance of the average value of the same area of the adjacent two groups of thickness data sets is the first distance value, and the spatial distance of the dispersion degree value is the second distance value. The first distance value and the second distance value are combined to construct a distance comprehensive index formula for judgment. Specifically, when comparing and judging, the second distance value corresponding to the dispersion degree value has a greater judgment weight to comprehensively reflect the change of the collected large amount of data. Finally, the target thickness data set is determined by the distance comprehensive index. For example, when the distance comprehensive index determined above is less than a preset index threshold, the thickness data set of the current time node is determined as the target thickness data set.

[0062] In some embodiments of the present application, the distance comprehensive index comprises a first distance comprehensive index of the thickness data set of the current time node relative to the thickness data set of the previous time node, and a second distance comprehensive index of the thickness data set of the current time node relative to the thickness data set of the next time node. When the first distance comprehensive index and the second distance comprehensive index are both less than or equal to a preset index threshold, the thickness data set of the current time node is determined as the target thickness data set.

[0063] In the present embodiment, the differences between the thickness data set of the current time node and the thickness data sets of the previous and next time nodes are combined to more accurately determine whether the equipment operation has reached a stable state, that is, the first distance comprehensive index and the second distance comprehensive index of the current time node relative to the previous and next time nodes are calculated by the above-mentioned Euclidean distance or Manhattan distance, so that only when the first distance comprehensive index and the second distance comprehensive index are both less than or equal to the preset index threshold, it is indicated that the data of the current time node has little difference relative to the data of the previous and next time nodes, that is, the data fluctuation is not large, and thus it is more accurately indicated that the current equipment operation has reached a stable state.

[0064] The preset index threshold can be set according to actual conditions, which will not be described here.

[0065] In some embodiments of the present application, the thickness data sets of multiple time nodes and the first distance comprehensive index and the second distance comprehensive index of adjacent time nodes are determined within a preset time period, and an index average value of the first distance comprehensive index and the second distance comprehensive index is determined. When the first distance comprehensive index and the second distance comprehensive index of multiple time nodes are both less than or equal to the preset index threshold, the thickness data set of the time node with the minimum index average value is determined as the target thickness data set, so that the target thickness data set determined in this way has the minimum difference relative to the data of the previous and next time nodes and the minimum data fluctuation, and thus it is more accurately indicated that the current equipment operation has reached a stable state.

[0066] In some embodiments of the present application, the thickness value is obtained by scanning the raw material by a thickness measuring device; and the step of constructing the thickness data set corresponding to each time node comprises the steps of: dividing the range scanned by the thickness measuring device on the raw material into multiple regions; determining the average value of the thickness value in each region; generating the thickness data according to the average value.

[0067] In the embodiment, a plurality of thickness values are obtained by scanning the stretched raw material with the thickness measuring device. The raw material scanned by the thickness measuring device is divided into a plurality of regions. It can be understood that for each region, there are a plurality of thickness value data. For a large number of thickness value data, the average value of the thickness values in each region is calculated. The average value can represent the overall thickness of the region to a certain extent. Subsequently, the thickness data in the thickness data set can be constructed based on the average value. For example, in one embodiment, the thickness data is the average value of the thickness values in the region, and in another embodiment, the thickness data is the average value and the dispersion value of the thickness values in the region.

[0068] In some embodiments of the present application, the process parameters of the lithium battery separator are corrected according to the target thickness data set, which includes: According to the average value corresponding to each region in the target thickness data set, the thickness variation of the raw material along the transverse and longitudinal directions is determined. When the thickness variation represents that the thickness of the raw material uniformly changes in the same direction, it is compared whether the change of the dispersion value of each region in the same direction matches the thickness variation. If yes, the equipment that needs to be corrected in the process parameters is located according to the thickness variation and the dispersion value, and the process parameters are corrected. If no, the determination step of the target thickness data set is performed again.

[0069] In the embodiment, the target thickness data set including the average value and the dispersion value is used to correct the process parameters of the lithium battery separator.

[0070] The thickness variation includes the first thickness variation and the second thickness variation.

[0071] In the preparation of the lithium battery separator, the raw material can be divided into a center region, a transition region and an edge region arranged in sequence along the longitudinal and transverse directions. The average value of these regions is calculated by the target thickness data set, so that the thickness variation of these regions in sequence along the transverse and longitudinal directions can be determined by the average value. When the thickness variation represents that the thickness of the raw material uniformly changes in the same direction, for example, when there is a thickness variation in the stretching process, the raw material may gradually become thinner or thicker along the stretching direction, or the thickness of the raw material center is uniform but gradually becomes thinner or thicker towards the edge. At this time, the corresponding dispersion value should change relatively small on the adjacent regions or in the same direction, so as to match the thickness variation. If there is no matching, it indicates that the currently determined target thickness data set may have a problem or the equipment has a fault, resulting in a non-uniform thickness fluctuation in each region. Therefore, the determination step of the target thickness data set can be performed again, for example, including steps 100-300, so as to improve the correction accuracy.

[0072] If the change condition matches, the equipment that needs to be corrected in process parameters can be located according to the thickness change condition and the dispersion degree value. For example, the thickness change condition reflects that the thickness fluctuates in adjacent areas or in the same transverse or longitudinal direction, and the stretching equipment processing this direction is directly located for correction of process parameters. If the dispersion degree value is too high, there is a problem in the area corresponding to the dispersion degree value, and the stretching equipment processing this area is located for correction of process parameters.

[0073] For example, when the second thickness change condition of the raw material in the longitudinal direction is determined according to the target thickness data set, and the rotation speed of the stretching equipment is corrected according to the second thickness change condition, on the one hand, the second thickness change condition of the overall raw material can be indicated by the average value of adjacent two areas or in the same direction, and on the other hand, the thickness change in each area is indicated by the dispersion degree value. In some specific application scenarios, the relatively large range of thickness fluctuation corresponding to multiple adjacent areas in the longitudinal direction may meet the production requirements, but the thickness change is caused by problems in equipment process parameters, such as rotation speed fluctuation at some time points. Therefore, the small range change in a single area is further reflected by combining the dispersion degree value, so that the detection of the stretched raw material is more accurate, and the process parameters can be better adjusted or corrected.

[0074] In some embodiments of the present application, the range of the raw material scanned by the thickness measuring equipment is divided into a plurality of regions according to the device size parameter of the stretching equipment. The device size parameter of the stretching equipment used to stretch the raw material is obtained, and the device size parameter includes the size of the stretching roller and the distance between adjacent stretching rollers. The region size parameter is determined according to the device size parameter, and the range of the raw material scanned by the thickness measuring equipment is divided into a plurality of regions according to the region size parameter.

[0075] In this embodiment, the above-mentioned region division is performed by the device size parameter of the stretching equipment. For the stretching of the raw material, it is usually realized by a plurality of stretching rollers arranged in sequence along the longitudinal direction. After the raw material passes through all the stretching rollers, one stretching is completed. Therefore, the thickness of each region in the stretched raw material has a certain corresponding relationship with the arrangement and size of the stretching rollers. Therefore, the device size includes the size of the stretching roller and the distance between adjacent stretching rollers in the longitudinal direction. The size of the stretching roller specifically can include the width of the stretching roller in the longitudinal direction. Based on this device parameter, the region size parameter is determined. Specifically, the region size parameter can be the region boundary length in the longitudinal direction. Finally, the plurality of regions divided can correspond to the position of the stretching equipment. When analyzing the thickness data set in the subsequent process to determine the position that needs to be adjusted and corrected, the position can be more accurately determined.

[0076] In one specific embodiment, the above-described scheme in the embodiments of this application is performed after the raw material is longitudinally stretched to facilitate subsequent transverse stretching preparation. In other embodiments, the above-described scheme of this application can also be implemented after transverse stretching.

[0077] This application also provides a self-calibration system for lithium battery separator process parameters based on big data analysis. The self-calibration system for lithium battery separator process parameters includes: One or more processors; Memory; and one or more applications, wherein the one or more applications are stored in memory and configured to be operated by the processor in any of the methods in any of the above method embodiments.

[0078] like Figure 4 As shown, it illustrates a structural schematic diagram of the electronic device involved in the embodiments of this application, specifically: The electronic device may include components such as a processor 501 with one or more processing cores, a storage unit 502 with one or more computer-readable storage media, a power supply 503, and an input unit 504. Those skilled in the art will understand that... Figure 4 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein: The processor 501 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs and / or modules stored in the storage unit 502, and by calling data stored in the storage unit 502, it performs various functions and processes data, thereby providing overall monitoring of the electronic device. Optionally, the processor 501 may include one or more processing cores; preferably, the processor 501 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may also not be integrated into the processor 501.

[0079] The storage unit 502 can be used to store software programs and modules, and the processor 501 executes various functional applications and data processing by running the software programs and modules stored in the storage unit 502. The storage unit 502 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, application programs required by at least one function (such as a sound playing function, an image playing function, etc.), etc.; and the data storage area can store data created according to the use of the electronic device, etc. In addition, the storage unit 502 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other volatile solid-state memory device. Accordingly, the storage unit 502 can also include a memory controller to provide access for the processor 501 to the storage unit 502.

[0080] The electronic device also includes a power supply 503 for powering various components. Preferably, the power supply 503 can be logically connected to the processor 501 through a power management system, so as to realize functions such as management of charging, discharging, and power consumption management through the power management system. The power supply 503 can also include one or more direct current or alternating current power supplies, a recharging system, a power supply fault detection circuit, a power supply converter or inverter, a power supply status indicator, etc.

[0081] The electronic device can also include an input unit 504, which can be used to receive input digital or character information, and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0082] Although not shown, the electronic device can also include a display unit, etc., which will not be described here. Specifically in the embodiments of the present application, the processor 501 in the electronic device will load the executable file corresponding to the process of one or more application programs into the storage unit 502 according to the following instructions, and run the application programs stored in the storage unit 502 by the processor 501, so as to realize various functions, as follows: In the preparation process of the lithium battery separator, the thickness values measured at multiple continuous time nodes after stretching of raw materials are obtained; A thickness data set corresponding to each of the time nodes is constructed, wherein the thickness data set includes thickness data for representing the thickness conditions of each region of the raw material, and the thickness data is generated according to the thickness values; Each of the thickness data sets is compared with other thickness data sets of adjacent time nodes, and a target thickness data set is determined in the thickness data set according to the data difference; The process parameters of the lithium battery separator are corrected according to the target thickness data set.

[0083] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by programs or instructions, or by relevant hardware controlled by the programs or instructions, and the programs or instructions can be stored in a computer readable storage medium and loaded and executed by a processor.

[0084] In the above embodiments, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the detailed description of other embodiments above, which will not be repeated here.

[0085] In the implementation, the above various units or structures can be implemented as independent entities, or can be combined as the same or several entities, and the specific implementation of the above various units or structures can be referred to the method embodiments above, which will not be repeated here.

[0086] The specific implementation of the above various operations can be referred to the above embodiments, which will not be repeated here.

[0087] The embodiments of the present application are described in detail above, and the principle and implementation mode of the present application are described by applying specific examples; the above embodiment description is only used to help understand the method of the present application and its core idea; meanwhile, according to the idea of the present application, the specific implementation mode and application range will be changed by those skilled in the art, and the above description should not be understood as the limitation of the present application.

Claims

1. A method for lithium battery separator process parameter self-correction based on big data analysis, characterized in that, The method comprises the following steps: During the preparation of a lithium battery separator, the thickness values of a raw material at multiple continuous time nodes after stretching are obtained; A thickness data set corresponding to each of the time nodes is constructed, wherein the thickness data set comprises thickness data for representing the thickness conditions of each region of the raw material, and the thickness data is generated according to the thickness values; Each of the thickness data sets is compared with other thickness data sets of adjacent time nodes, and a target thickness data set is determined in the thickness data set according to the data difference conditions; The process parameters of the lithium battery separator are corrected according to the target thickness data set.

2. The big data analytics based lithium battery separator process parameter self-correction method of claim 1, wherein, The process parameters comprise clamping force of a clamping device and rotating speed of a stretching device, and the correction of the process parameters of the lithium battery separator according to the target thickness data set comprises: The first thickness variation condition of the raw material from the center to the edge in the transverse direction and the second thickness variation condition of the raw material in the longitudinal direction are determined according to the target thickness data set; The clamping force of the clamping device and the rotating speed of the stretching device are corrected in combination with the first thickness variation and the second thickness variation.

3. The big data analytics based lithium battery separator process parameter self-correction method of claim 2, wherein, The first thickness variation condition comprises a transverse thickness data deviation value of adjacent two regions of the raw material in the transverse direction, and the second thickness variation condition comprises a longitudinal thickness data deviation value of adjacent two regions of the raw material in the longitudinal direction, and the correction of the clamping force of the clamping device and the rotating speed of the stretching device in combination with the first thickness variation and the second thickness variation comprises: If the transverse thickness data deviation value is greater than a preset first deviation threshold value and the longitudinal thickness data deviation value is less than or equal to a preset second deviation threshold value, the clamping force of the clamping device is corrected according to the transverse thickness data deviation value; If the longitudinal thickness data deviation value is greater than the second deviation threshold value and the transverse thickness data deviation value is less than or equal to the first deviation threshold value, the rotating speed of the stretching device is corrected according to the longitudinal thickness data deviation value; If the transverse thickness data deviation value is greater than the first deviation threshold value and the longitudinal thickness data deviation value is greater than the second deviation threshold value, the clamping force of the clamping device and the rotating speed of the stretching device are corrected in combination with the transverse thickness data deviation value and the longitudinal thickness data deviation value.

4. The big data analytics based lithium battery separator process parameter self-correction method of claim 3, wherein, The correction of the clamping force of the clamping device according to the transverse thickness data deviation value comprises: All thickness difference abnormal regions of the raw material are located according to the transverse thickness data deviation value; An abnormal region distance between all the thickness difference abnormal regions is calculated; A clamping force initial value is determined according to the transverse thickness data deviation value and by using regression analysis; The clamping force initial value is corrected according to the abnormal region distance to obtain a clamping force target value; The clamping force of the clamping device is corrected according to the clamping force target value.

5. The big data analytics based lithium battery separator process parameter self-correction method of claim 3, wherein, The correction of the clamping force of the clamping device and the rotating speed of the stretching device in combination with the transverse thickness data deviation value and the longitudinal thickness data deviation value comprises: inputting the lateral thickness data deviation value and the longitudinal thickness data deviation value into a pre-constructed deviation coupling correction model, outputting a clamping force correction amplitude and a rotating speed correction amplitude by the deviation coupling correction model, the deviation coupling correction model being constructed based on a gradient boosting regression tree; correcting the clamping force of the clamping device and the rotating speed of the stretching device according to the clamping force correction amplitude and the rotating speed correction amplitude respectively.

6. The big data analytics based lithium battery separator process parameter self-correction method according to any one of claims 1-5, wherein, the comparing each of the thickness data sets with other thickness data sets of adjacent time nodes, and determining a target thickness data set in the thickness data sets according to data difference conditions includes: determining a distance value of thickness data in each of the thickness data sets and thickness data in other thickness data sets of adjacent time nodes in a data space, wherein the distance value includes an Euclidean distance or a Manhattan distance; determining the target thickness data set according to the distance value.

7. The big data analytics based lithium battery separator process parameter self-correction method of claim 6, wherein, the thickness data includes an average value and a dispersion degree value of the thickness value in each of the regions, and the distance value includes a first distance value calculated from the average value of each region in each of the thickness data sets and other thickness data sets of adjacent time nodes, and a second distance value calculated from the dispersion degree value; the determining the target thickness data set according to the distance value includes: determine a distance comprehensive index according to the first distance value and the second distance value, wherein a determination formula of the distance comprehensive index is: D = w 1 D 1+ w 2 D 2 , D1 is the first distance value, D2 is the second distance value, w 1 is a first preset weight, w 2 is a second preset weight, w 2 is greater than w 1. determining the target thickness data set according to the distance comprehensive index.

8. The big data analytics based lithium battery separator process parameter self-correction method of claim 7, wherein, the distance comprehensive index includes a first distance comprehensive index of the thickness data set of the current time node relative to a thickness data set of a previous time node, and a second distance comprehensive index of the thickness data set of the current time node relative to a thickness data set of a next time node; when the first distance comprehensive index and the second distance comprehensive index are both less than or equal to a preset index threshold, determining the thickness data set of the current time node as the target thickness data set.

9. The big data analytics based lithium battery separator process parameter self-correction method of claim 7, wherein, the correcting the process parameters of the lithium battery separator according to the target thickness data set includes: determining thickness variation of the raw material along the lateral direction and the longitudinal direction according to the average value of each region in the target thickness data set; when the thickness variation represents uniform thickness variation of the raw material in the same direction, comparing whether the variation of the dispersion degree value of each region in the same direction is adapted to the thickness variation; if yes, positioning the device requiring process parameter correction according to the thickness variation and the dispersion degree value, and correcting the process parameters; if no, re-performing the determining step of the target thickness data set.

10. A lithium battery separator process parameter self-correction system based on big data analysis, characterized in that, the lithium battery separator process parameter self-correction system based on big data analysis includes: one or more processors; a memory; and one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the processor to implement the method of any one of claims 1 to 9.

Images