CN-121997130-A - Electrode material dislocation adjusting method, device, equipment and storage medium

Abstract

The disclosure provides a method, a device, equipment and a storage medium for adjusting dislocation of electrode materials, and relates to the technical field of battery production. The method comprises the steps of obtaining pre-process data of an electrode material to be wound, wherein the pre-process data are process parameter values of a process of the electrode material to be wound before a winding process, processing the pre-process data through a process matching model to obtain target process parameters corresponding to ideal alignment states in the electrode material to be wound, processing the target process parameters and the pre-process data through a tab dislocation prediction model to obtain dislocation predicted values of tabs in the electrode material to be wound, and generating recommended instructions based on the target process parameters if the dislocation predicted values meet preset conditions, so that closed-loop control of winding and pre-process is achieved, and tab dislocation adjustment is effectively achieved.

Inventors

• LIN XUJIE

Assignees

• 厦门海辰储能科技股份有限公司

Dates

Publication Date

20260508

Application Date

20260119

Claims (16)

1. 1. A method for adjusting misalignment of an electrode material, comprising: Acquiring pre-process data of an electrode material to be wound, wherein the pre-process data is a process parameter value of a working procedure of the electrode material to be wound before a winding working procedure; processing the pre-process data through a process matching model to obtain target process parameters corresponding to the ideal alignment state of the electrode material to be wound; Processing the target process parameters and the pre-process data through a tab dislocation prediction model to obtain a dislocation prediction value of the tab in the electrode material to be coiled; And if the dislocation predicted value meets a preset condition, generating a recommended instruction based on the target process parameter.
2. 2. The method of claim 1, wherein the target process parameters include pinch roll pressure and/or variable caliper servo position; When the target process parameters comprise the pressure of the compaction roller, the pre-process data comprise first process data, the process matching model comprises a first matching model, and the first matching model is obtained through training of historical first process data of the target tab and corresponding pressure label of the compaction roller; When the target process parameters comprise the variable-roll-diameter servo position, the pre-process data comprise second process data, the process matching model comprises a second matching model, and the second matching model is obtained through training of historical second process data of the target tab and corresponding variable-roll-diameter servo position labels.
3. 3. The method of claim 1, wherein before processing the pre-process data by the process matching model to obtain a target process parameter corresponding to a tab in the electrode material to be wound being in an ideal alignment state, the method further comprises: acquiring historical first process data of a target tab and corresponding process parameter labels, and constructing a training set and a verification set based on a first preset proportion; Performing iterative training on the process matching model to be trained based on the training set to obtain an initial process matching model; Performing model evaluation on the initial process matching model based on the verification set to obtain a model evaluation index; If the model evaluation index does not meet the preset performance condition, adjusting the model parameters of the initial process matching model; and if the model evaluation index meets the preset performance condition, determining the initial process matching model as the process matching model.
4. 4. A method according to claim 3, wherein prior to iteratively training the process matching model to be trained based on the training set to obtain an initial process matching model, the method further comprises: Initializing super parameters of a process matching model to be trained; After the initial process matching model is determined to be the process matching model if the model evaluation index meets the preset performance condition, the method further comprises: Based on the verification set, performing super-parameter tuning on the super-parameters of the initial process matching model, determining an optimal super-parameter combination, and configuring the process matching model according to the optimal super-parameter combination.
5. 5. A method according to claim 3, characterized in that the method further comprises: calculating the characteristic contribution degree of each historical first process data based on the process matching model; Sorting the feature contribution degree, screening the pre-process parameters corresponding to the historical first process data with the feature contribution degree larger than a preset contribution degree threshold value, and/or screening the pre-process parameters corresponding to the historical first process data with the preset quantity, which are in front of the sorting, as the input quantity of the process matching model.
6. 6. The method of claim 3, wherein the model evaluation index comprises at least one of a mean square error, an average absolute error, and a decision coefficient; Wherein the model evaluation index satisfies the preset performance condition, including at least one of: the mean square error is less than or equal to a first error threshold; the average absolute error is less than or equal to a second error threshold; the decision coefficient is greater than or equal to a preset coefficient threshold.
7. 7. The method of claim 1, wherein the pre-process data comprises second process data, and wherein the tab misalignment prediction model is trained from the target process parameters and the second process data and corresponding tab misalignment amount labels.
8. 8. The method of claim 1, wherein the process matching model and the tab misalignment prediction model comprise at least one of a lightweight gradient hoist, an extreme gradient hoist decision tree model, a random forest, a support vector machine.
9. 9. The method of claim 2, wherein the first process data comprises at least one of a cathode coated on-surface density average, an anode coated on-surface density average, a roll-on-anode roll pre-slit in-process time, an anode roll thickness average, an anode roll speed, a roll-on-cathode roll pre-slit in-process time, a cathode roll thickness average, a cathode roll speed.
10. 10. The method of claim 2 or 8, wherein the second process data comprises at least one of an anode die-anode roll pre-cut in process time, a cathode die-cathode roll pre-cut in process time, a cathode roll thickness average, a cathode roll speed, a cathode double sided density average, an anode roll thickness average, and an anode double sided density average.
11. 11. The method of claim 1, wherein the preset condition comprises a preset misalignment threshold; if the dislocation predicted value meets a preset condition, generating a recommended instruction based on the target process parameter, including: If the dislocation predicted value is smaller than or equal to the preset dislocation threshold value, judging that the dislocation predicted value meets the preset condition; And if the dislocation predicted value is larger than the preset dislocation threshold value, judging that the dislocation predicted value does not meet the preset condition.
12. 12. The method of claim 11, wherein the misalignment prediction value comprises at least one of a tab misalignment amount, a tab misalignment probability, or a tab misalignment level.
13. 13. The method of claim 11, wherein the method further comprises: and if the dislocation predicted value does not meet the preset condition, not recommending the target process parameter.
14. 14. An electrode material misalignment adjustment apparatus, comprising: The data acquisition module is used for acquiring pre-process data of the electrode material to be wound, wherein the pre-process data is a process parameter value of a process before a winding process of the electrode material to be wound; The parameter determining module is used for processing the pre-process data through a process matching model to obtain target process parameters corresponding to ideal alignment states in the electrode material to be wound; the dislocation detection module is used for processing the target process parameters and the pre-process data through a tab dislocation prediction model to obtain a dislocation prediction value of the tab in the electrode material to be coiled; and the parameter recommendation module is used for generating a recommendation instruction based on the target process parameter if the dislocation predicted value meets a preset condition.
15. 15. An electronic device, comprising: A processor; a memory for storing executable instructions of the processor; wherein the processor is configured to perform the electrode material misalignment adjustment method of any one of claims 1-13 via execution of the executable instructions.
16. 16. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements the electrode material misalignment adjustment method of any one of claims 1 to 13.

Description

Electrode material dislocation adjusting method, device, equipment and storage medium Technical Field The present disclosure relates to the field of battery production technology, and in particular, to an electrode material misalignment adjustment method, an electrode material misalignment adjustment apparatus, an electronic device, and a computer-readable storage medium. Background Battery winding preparation is one of the key steps in the battery production process, directly affecting the performance, safety and life of the battery. In the battery winding process, the positive electrode sheet, the negative electrode sheet, and the separator are alternately laminated and wound into a winding core structure. The tab is a connection point between the electrode and an external circuit, and the alignment accuracy of the tab is critical to the overall performance of the battery. In the related art, the adjustment of tab dislocation is realized mainly depending on the automatic control capability of a winding machine. However, the above adjustment method is easy to cause the problems of large fluctuation of product yield and low overall process controllability. Disclosure of Invention The disclosure provides a method, a device, equipment and a storage medium for adjusting electrode material dislocation, which at least overcome the problems of large product yield fluctuation and low process overall controllability of a pole ear dislocation adjusting mode in the related technology to a certain extent. Other features and advantages of the present disclosure will be apparent from the following detailed description, or may be learned in part by the practice of the disclosure. According to one aspect of the disclosure, a method for adjusting dislocation of a production electrode material is provided, and the method comprises the steps of obtaining pre-process data of the electrode material to be coiled, wherein the pre-process data are process parameter values of a working procedure of the electrode material to be coiled before a coiling working procedure, processing the pre-process data through a process matching model to obtain target process parameters corresponding to the ideal alignment state of the electrode material to be coiled, processing the target process parameters and the pre-process data through a tab dislocation prediction model to obtain dislocation predicted values of tabs in the electrode material to be coiled, and generating recommended instructions based on the target process parameters if the dislocation predicted values meet preset conditions. In one embodiment of the disclosure, the target process parameters include pinch roll pressure and/or a roll-to-roll servo position, the pre-process data includes first process data when the target process parameters include pinch roll pressure, the process matching model includes a first matching model trained from historical first process data of the target tab and a corresponding pinch roll pressure label, and the pre-process data includes second process data when the target process parameters include a roll-to-roll servo position, the process matching model includes a second matching model trained from historical second process data of the target tab and a corresponding roll-to-roll servo position label. In one embodiment of the disclosure, before the pre-process data is processed through the process matching model to obtain target process parameters corresponding to the ideal alignment state of the tab in the electrode material to be wound, the method further comprises the steps of obtaining historical first process data of the target tab and corresponding process parameter labels, constructing a training set and a verification set based on a first preset proportion, performing iterative training on the process matching model to be trained based on the training set to obtain an initial process matching model, performing model evaluation on the initial process matching model based on the verification set to obtain a model evaluation index, adjusting model parameters of the initial process matching model if the model evaluation index does not meet preset performance conditions, and determining the initial process matching model as the process matching model if the model evaluation index meets the preset performance conditions. In one embodiment of the disclosure, before the training set is used for carrying out iterative training on the process matching model to be trained to obtain an initial process matching model, the method further comprises initializing the super parameters of the process matching model to be trained, after the initial process matching model is determined to be the process matching model if the model evaluation index meets the preset performance condition, carrying out super parameter tuning on the super parameters of the initial process matching model based on the verification set, determining an optimal super parameter combination