Tab alignment method, device, equipment, medium and product

Abstract

The invention discloses a tab alignment method, device and equipment, a medium and a product in the technical field of battery production. The method comprises the following steps: acquiring materials of a pole piece and a diaphragm as well as the thickness of the pole piece and the thickness of the diaphragm at a rolling moment; selecting a corresponding thickness prediction model constructed based on a variable length memory (LSTM) model according to the pole piece and the diaphragm; respectively inputting the pole piece thickness and the diaphragm thickness at the rolling moment into the corresponding thickness prediction models, and obtaining the pole piece thickness and the diaphragm thickness at the future target moment; calculating the tab spacing according to the pole piece thickness and the diaphragm thickness at the future target moment; carrying out tab alignment operation at a future target moment according to the tab spacing; by accurately acquiring the thickness of the pole piece, the accuracy of the input data for calculating the distance between the tabs is improved, and the accurate positions of the tabs are finally calculated, so that accurate tab alignment operation is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of battery production, and in particular to a tab alignment method, device, equipment, medium and product. Background Art

[0002] In the process of lithium battery manufacturing, tab misalignment may lead to a series of potential problems, such as increased internal resistance, thermal management problems, uneven battery capacity, unstable mechanical structure, poor welding, battery consistency problems, etc. These problems affect the performance, safety and life of the battery. Therefore, tab alignment is a key research in the industrial and academic fields with practical application value.

[0003] Although there are some technical means for aligning the tabs in the prior art, the following problems generally exist: (1) The deviation of cell thickness data in actual production conditions is not taken into account.

[0004] The key parameter used in calculating the position of the tabs by laser cutting is the thickness of the pole piece, which is measured during rolling several hours ago. During this period, the pole piece will rebound after rolling, and the actual thickness will change. Whether the calculated tab spacing can be aligned can only be reflected in the subsequent winding process. Therefore, the grasp of the thickness change is both the key point and the difficulty in determining the alignment.

[0005] (2) The existing time series prediction model LSTM lacks attention to the size of the time window.

[0006] When a larger time window is used, the LSTM model can see more historical information and thus better capture long-distance dependencies. Although this is very important for some tasks that require understanding long-term context, too long a sequence may also lead to gradient vanishing or gradient exploding problems. On the other hand, when the time window is shorter, it helps the model focus more on current information, respond quickly to changes, and reduce the computational burden of processing long-term sequences. This is more effective for tasks with strong local dependencies, but it also inevitably leads to a reduction in long-term dependencies, reducing the model's ability to handle tasks that require long-term dependencies.

[0007] (3) The automation level of the spacing correction process is low.

[0008] The existing spacing correction process gradually measures the spacing between two adjacent tabs and adjusts them in turn, which requires a lot of manpower and material resources. The process of continuous trial and error not only causes waste of materials and manpower, but also causes production stoppages due to continuous trial and error in related processes, which seriously affects the overall production efficiency. Summary of the invention

[0009] The purpose of the present invention is to overcome the deficiencies in the prior art and provide a method, device, equipment, medium and product for pole lug alignment, which predicts the thickness of the pole piece and diaphragm at any time after rolling through a variable-length memory LSTM model, improves the accuracy of the input data for calculating the pole lug spacing, and finally calculates the exact position of the pole lug, thereby achieving precise pole lug alignment operation.

[0010] To achieve the above object, the present invention is implemented by adopting the following technical solutions: In a first aspect, the present invention provides a tab alignment method, comprising: Obtain the materials of the pole piece and the diaphragm and the thickness of the pole piece and the diaphragm at the time of rolling; Selecting a corresponding thickness prediction model based on a variable length memory (LSTM) model according to the pole piece and the diaphragm; Inputting the electrode piece thickness and the diaphragm thickness at the rolling moment into the corresponding thickness prediction model respectively to obtain the electrode piece thickness and the diaphragm thickness at the future target moment; Calculate the distance between the pole tabs according to the pole piece thickness and the diaphragm thickness at a future target moment; The tab alignment operation is performed at a future target time according to the tab spacing.

[0011] The above technical solution aims to solve the problem that the pole piece will rebound after rolling and the actual thickness will change. The relationship between the thickness of the pole piece and time after rolling is modeled based on the variable-length memory LSTM model, and a thickness prediction model that can be used to derive the thickness of the pole piece at any time is obtained, thereby obtaining the accurate thickness of the pole piece, improving the accuracy of the input data for calculating the spacing between the pole tabs, and finally calculating the exact position of the pole tabs, thereby achieving precise pole tab alignment operations.

[0012] Optionally, the thickness prediction model includes a positive electrode sheet thickness prediction model, a negative electrode sheet thickness prediction model and a diaphragm thickness prediction model constructed based on a variable length memory LSTM model.

[0013] Since the materials of the positive electrode sheet, negative electrode sheet and separator are different, the amount of rebound after rolling is different. Therefore, in order to improve the accuracy of thickness prediction, the positive electrode sheet, negative electrode sheet and separator are modeled according to their materials. After modeling and recording different negative electrode sheet materials, positive electrode sheet materials and separator materials, if different types of batteries are used in the future, they can be used only by combining the recorded materials. This improves the flexibility of model use.

[0014] Optionally, the variable-length memory LSTM model includes a first LSTM module, a second LSTM module and a third LSTM module; Each layer of the variable length memory LSTM model inputs the input data , input data , cell status And output data , for the moment; The first LSTM module is based on the input data and input data Get the second branch Cell state at a moment and output data ; The second LSTM module is based on the input data Get cell status and output data ; The third LSTM module is based on the cell state , output data And input data Get the first branch Cell state at a moment and output data ; The output data and the output data The output data is obtained by weighted fusion through the channel attention mechanism , the cell state and the cell state The cell state is obtained by weighted fusion through the channel attention mechanism .

[0015] The above technical solution uses a variable-length memory LSTM model to build a thickness prediction model. Compared with the traditional LSTM model, the variable-length memory LSTM model uses a different time window. The time window size of each layer on the input is 2; one of the branches in the layer uses a time window size of 2, and the other branch uses a time window of 1. The cell state and output at the current moment are transformed with different step sizes, so it contains memory patterns of different lengths; the upper and lower branches output their respective cell states respectively. and , output results and Finally, the channel attention mechanism is applied to the cell state and output results for weighted fusion, which retains the information that is effective for the task and finally realizes the adaptive selection of different memory modes. This solves the problem that the traditional LSTM model lacks attention to the time window size.

[0016] Optionally, the thickness prediction model is deployed and used after training, and the training process of the thickness prediction model includes: After the calibrated position of the target object is rolled, the calibrated position of the target object is periodically sampled to obtain the thickness of the target object at each sampling time; The training set is constructed by taking the target object thickness as the true label at the sampling moment; The thickness prediction model is trained by using the training set. In each round of training, the target object thickness at a future target moment is obtained and used as a prediction label, the loss is calculated according to the true label and the prediction label, and the model parameters of the thickness prediction model are optimized according to the loss until a maximum number of iterations is reached or the loss converges, thereby completing the training; Among them, when training the thickness prediction model of the pole piece, the target object is the pole piece, and when training the thickness prediction model of the diaphragm, the target object is the diaphragm.

[0017] In the above technical solution, the sampling frequency is usually once an hour, and the thickness measurement is tracked to 24 times a day; the larger the number of samples, the more accurate the fitted thickness change. Considering the cost factor, the number of samples should be reduced as much as possible, but for the sake of generalization, it should be at least 10 or more, so as to ensure the accuracy of the trained model.

[0018] Optionally, the calculating the pole tab spacing according to the pole piece thickness and the diaphragm thickness at the future target moment includes: Calculate the distance between negative electrode tabs : ; ; In the formula, For the The distance between the negative electrode tabs, is the circumference of the winding needle, is the negative electrode position increment, The thickness of the negative electrode sheet, the positive electrode sheet and the separator at the future target time; Calculate the positive electrode tab spacing : ; In the formula, For the The distance between the positive electrode ears.

[0019] In the above technical solution, the number of positive electrode spacings and the number of negative electrode spacings can be set in advance. The variables in the formula are mainly the negative electrode sheet thickness, the positive electrode sheet thickness and the diaphragm thickness. After accurately obtaining the negative electrode sheet thickness, the positive electrode sheet thickness and the diaphragm thickness at the future target moment through the thickness prediction model, the negative electrode tab spacing and the positive electrode tab spacing can also be accurately calculated, thereby serving as the basis for subsequent tab alignment operations.

[0020] In a second aspect, the present invention provides a tab alignment device, characterized in that the device comprises: A data acquisition module is configured to acquire the material of the pole piece and the diaphragm and the thickness of the pole piece and the diaphragm at the time of rolling; A model selection module is configured to select a corresponding thickness prediction model constructed based on a variable length memory (LSTM) model according to the pole piece and the diaphragm; A thickness prediction module is configured to input the electrode thickness and the diaphragm thickness at the rolling moment into the corresponding thickness prediction model to obtain the electrode thickness and the diaphragm thickness at a future target moment; A spacing calculation module is configured to calculate the tab spacing according to the pole piece thickness and the diaphragm thickness at a future target moment; The tab alignment module is configured to perform tab alignment operations at a future target time according to the tab spacing.

[0021] Optionally, the thickness prediction model includes a positive electrode sheet thickness prediction model, a negative electrode sheet thickness prediction model and a diaphragm thickness prediction model constructed based on a variable length memory LSTM model.

[0022] In a third aspect, the present invention provides an electronic device, characterized in that it includes a processor and a storage medium; The storage medium is used to store instructions; The processor is used to operate according to the instructions to execute the steps according to the above method.

[0023] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above method when executed by a processor.

[0024] In a fifth aspect, the present invention provides a computer program product, comprising a computer program / instruction, which implements the steps of the above method when executed by a processor.

[0025] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a method, device, equipment, medium and product for aligning a pole piece. In order to solve the problem that the pole piece will rebound after rolling and the actual thickness will change, a variable-length memory LSTM model is used to model the relationship between the thickness of the pole piece and time after rolling, so as to obtain a thickness prediction model that can derive the thickness of the pole piece at any time, thereby obtaining the accurate thickness of the pole piece, improving the accuracy of the input data for calculating the spacing between the pole pieces, and finally calculating the accurate position of the pole piece, so as to achieve accurate alignment of the pole pieces. The memory mode of the variable-length memory LSTM model is adopted to adaptively adjust tasks that require long-term dependence or short-term response flexibility, thereby improving the prediction accuracy of the pole piece thickness. At the same time, compared with the traditional manual step-by-step measurement and adjustment method, the present invention models the relationship between the spacing and key factors such as thickness and winding needle circumference, automatically predicts the spacing between all pole pieces, and can efficiently adjust according to the deviation state. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a schematic flow chart of a tab alignment method provided by an embodiment of the present invention; Figure 2 It is a schematic diagram of the operation logic of the variable length memory LSTM model provided by an embodiment of the present invention; Figure 3 is a schematic diagram of a human-computer interaction interface of a tab alignment method provided by an embodiment of the present invention; Figure 4 is a verification data diagram for predicting the tab spacing provided by an embodiment of the present invention; Figure 5 is a verification data diagram of the actual tab spacing provided by an embodiment of the present invention; Figure 6 is a verification data diagram of the tab spacing error provided by an embodiment of the present invention; Figure 7 It is a schematic diagram of the structure of the tab alignment device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0028] Embodiment 1:

[0029] like Figure 1 As shown, an embodiment of the present invention provides a method for aligning a tab, comprising the following steps: Step S1, obtaining the materials of the pole piece and the diaphragm and the thickness of the pole piece and the diaphragm at the time of rolling.

[0030] Since the materials of the positive electrode sheet, negative electrode sheet and separator are different, the amount of rebound generated after rolling is different. Therefore, in order to improve the accuracy of thickness prediction, the positive electrode sheet, negative electrode sheet and separator are modeled according to their materials. After modeling and recording different negative electrode sheet materials, positive electrode sheet materials and separator materials, if different types of battery cells are used in the future, they only need to be combined and used. The flexibility of model use is improved. Therefore, step S1 gives priority to obtaining the materials of the electrode sheet and separator in order to reasonably select the thickness prediction model from the recorded materials.

[0031] Step S2: Select a corresponding thickness prediction model based on a variable length memory (LSTM) model according to the pole piece and the diaphragm.

[0032] The thickness prediction model includes a positive electrode sheet thickness prediction model, a negative electrode sheet thickness prediction model and a diaphragm thickness prediction model based on a variable-length memory LSTM model.

[0033] Since the traditional LSTM model lacks attention to the size of the time window, various problems will arise when the time window is long or short.

[0034] like Figure 2 As shown, specifically in this embodiment, a variable length memory LSTM model is adopted, which includes: A first LSTM module, a second LSTM module, and a third LSTM module; Each layer of the variable length memory LSTM model inputs the input data , input data , cell status And output data , for the moment; The first LSTM module is based on the input data and input data Get the second branch Cell state at a moment and output data ; The second LSTM module is based on the input data Get cell status and output data ; The third LSTM module is based on the cell state , output data And input data Get the first branch Cell state at a moment and output data ; Output Data and output data The output data is obtained by weighted fusion through the channel attention mechanism , cell status and cell status The cell state is obtained by weighted fusion through the channel attention mechanism .

[0035] Compared with the traditional LSTM model, the variable-length memory LSTM model uses a different time window. The time window size of each layer in the input is 2; one branch in the layer uses a time window size of 2, and the other branch uses a time window of 1. The cell state and output at the current moment are transformed with different step sizes, so it contains memory patterns of different lengths; the upper and lower branches output their respective cell states. and , output results and Finally, the channel attention mechanism is applied to the cell state and output results for weighted fusion, which retains the information that is effective for the task and finally realizes the adaptive selection of different memory modes. This solves the problem that the traditional LSTM model lacks attention to the time window size.

[0036] The thickness prediction model is deployed and used after training. Specifically in this embodiment, the training process of the thickness prediction model includes: After the calibrated position of the target object is rolled, the calibrated position of the target object is periodically sampled to obtain the thickness of the target object at each sampling time; The training set is constructed by taking the target object thickness as the true label at the sampling moment; The thickness prediction model is trained using the training set. In each round of training, the target object thickness at the future target moment is obtained and used as the prediction label. The loss is calculated based on the true label and the predicted label, and the model parameters of the thickness prediction model are optimized based on the loss until the maximum number of iterations is reached or the loss converges, completing the training. Among them, when training the thickness prediction model of the pole piece, the target object is the pole piece, and when training the thickness prediction model of the diaphragm, the target object is the diaphragm.

[0037] In the above technical solution, the sampling frequency is usually once an hour, and the thickness measurement is tracked to 24 times a day; the larger the number of samples, the more accurate the fitted thickness change. Considering the cost factor, the number of samples should be reduced as much as possible, but for the sake of generalization, it should be at least 10 or more, so as to ensure the accuracy of the trained model.

[0038] Step S3: input the electrode thickness and the diaphragm thickness at the rolling moment into the corresponding thickness prediction model respectively to obtain the electrode thickness and the diaphragm thickness at the future target moment.

[0039] Step S4, calculating the distance between the pole tabs according to the pole piece thickness and the diaphragm thickness at the future target moment.

[0040] Calculate the distance between negative electrode tabs : ; ; In the formula, For the The distance between the negative electrode tabs, is the circumference of the winding needle, is the negative electrode position increment, The thickness of the negative electrode sheet, the positive electrode sheet and the separator at the future target time; Calculate the positive electrode tab spacing : ; In the formula, For the The distance between the positive electrode ears.

[0041] In the above technical solution, the number of positive electrode spacings and the number of negative electrode spacings can be set in advance. The variables in the formula are mainly the negative electrode sheet thickness, the positive electrode sheet thickness and the diaphragm thickness. After accurately obtaining the negative electrode sheet thickness, the positive electrode sheet thickness and the diaphragm thickness at the future target moment through the thickness prediction model, the negative electrode tab spacing and the positive electrode tab spacing can also be accurately calculated, thereby serving as the basis for subsequent tab alignment operations.

[0042] Step S5: performing the tab alignment operation at a future target time according to the tab spacing.

[0043] like Figure 3 As shown, specifically in this implementation, in order to facilitate the staff to directly use the pole ear alignment method provided in this embodiment, the relevant parameters in this embodiment can be concentrated on the same human-computer interaction interface. The staff only needs to input the relevant parameters, and the software automatically runs the pole ear alignment method provided in this embodiment to obtain the final pole ear spacing.

[0044] The above technical solution aims to solve the problem that the pole piece will rebound after rolling and the actual thickness will change. The relationship between the thickness of the pole piece and time after rolling is modeled based on the variable-length memory LSTM model, and a thickness prediction model that can be used to derive the thickness of the pole piece at any time is obtained, thereby obtaining the accurate thickness of the pole piece, improving the accuracy of the input data for calculating the spacing between the pole tabs, and finally calculating the exact position of the pole tabs, thereby achieving precise pole tab alignment operations.

[0045] like Figure 4 , Figure 5 as well as Figure 6As shown, this embodiment can derive a spacing prediction with good tab alignment when the thickness data is accurate, and the error can be controlled within a very small range.

[0046] Embodiment 2: like Figure 7 As shown, an embodiment of the present invention provides a tab alignment device, characterized in that the device comprises: A data acquisition module is configured to acquire the material of the pole piece and the diaphragm and the thickness of the pole piece and the diaphragm at the time of rolling; The model selection module is configured to select the corresponding thickness prediction model constructed based on the variable-length memory LSTM model according to the pole piece and the diaphragm; the thickness prediction model includes a positive pole piece thickness prediction model, a negative pole piece thickness prediction model and a diaphragm thickness prediction model constructed based on the variable-length memory LSTM model.

[0047] A thickness prediction module is configured to input the electrode thickness and the diaphragm thickness at the rolling moment into corresponding thickness prediction models to obtain the electrode thickness and the diaphragm thickness at a future target moment; A spacing calculation module is configured to calculate the spacing between the pole tabs according to the thickness of the pole piece and the thickness of the diaphragm at a future target moment; The tab alignment module is configured to perform tab alignment operations at a future target time according to the tab spacing.

[0048] Embodiment three:

[0049] Based on the tab alignment method provided in the first embodiment, an embodiment of the present invention provides an electronic device, characterized in that it includes a processor and a storage medium; The storage medium is used to store instructions; The processor is used to operate according to the instructions to execute the steps according to the above method.

[0050] Embodiment 4:

[0051] Based on the tab alignment method provided in the first embodiment, an embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps of the above method are implemented.

[0052] Embodiment five:

[0053] Based on the tab alignment method provided in the first embodiment, an embodiment of the present invention provides a computer program product, including a computer program / instruction, which implements the steps of the above method when executed by a processor.

[0054] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0055] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0056] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0057] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0058] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for aligning a tab, characterized in that: include: Obtain the materials of the pole piece and the diaphragm and the thickness of the pole piece and the diaphragm at the time of rolling; Selecting a corresponding thickness prediction model based on a variable length memory (LSTM) model according to the pole piece and the diaphragm; Inputting the electrode piece thickness and the diaphragm thickness at the rolling moment into the corresponding thickness prediction model respectively to obtain the electrode piece thickness and the diaphragm thickness at the future target moment; Calculate the distance between the pole tabs according to the pole piece thickness and the diaphragm thickness at a future target moment; The tab alignment operation is performed at a future target time according to the tab spacing.

2. The method for aligning the tabs according to claim 1, characterized in that: The thickness prediction model includes a positive electrode sheet thickness prediction model, a negative electrode sheet thickness prediction model and a diaphragm thickness prediction model constructed based on a variable length memory (LSTM) model.

3. The method for aligning the tabs according to claim 1, characterized in that: The variable length memory LSTM model includes a first LSTM module, a second LSTM module and a third LSTM module; Each layer of the variable length memory LSTM model inputs the input data , input data , cell status And output data , for the moment; The first LSTM module is based on the input data and input data Get the second branch Cell state at a moment and output data ; The second LSTM module is based on the input data Get cell status and output data ; The third LSTM module is based on the cell state , output data And input data Get the first branch Cell state at a moment and output data ; The output data and the output data The output data is obtained by weighted fusion through the channel attention mechanism , the cell state and the cell state The cell state is obtained by weighted fusion through the channel attention mechanism .

4. The method for aligning the tabs according to claim 1, characterized in that: The thickness prediction model is deployed and used after training. The training process of the thickness prediction model includes: After the calibrated position of the target object is rolled, the calibrated position of the target object is periodically sampled to obtain the thickness of the target object at each sampling time; The training set is constructed by taking the target object thickness as the true label at the sampling moment; The thickness prediction model is trained by using the training set. In each round of training, the target object thickness at a future target moment is obtained and used as a prediction label, the loss is calculated according to the true label and the prediction label, and the model parameters of the thickness prediction model are optimized according to the loss until a maximum number of iterations is reached or the loss converges, thereby completing the training; Among them, when training the thickness prediction model of the pole piece, the target object is the pole piece, and when training the thickness prediction model of the diaphragm, the target object is the diaphragm.

5. The tab alignment method according to claim 1, characterized in that: The calculating of the distance between the pole tabs according to the pole piece thickness and the diaphragm thickness at the future target moment comprises: Calculate the distance between negative electrode tabs : ； ； In the formula, For the The distance between the negative electrode tabs, is the circumference of the winding needle, is the negative electrode position increment, The thickness of the negative electrode sheet, the positive electrode sheet and the separator at the future target time; Calculate the positive electrode tab spacing : ； In the formula, For the The distance between the positive electrode ears.

6. A tab alignment device, characterized in that: The device comprises: A data acquisition module is configured to acquire the material of the pole piece and the diaphragm and the thickness of the pole piece and the diaphragm at the time of rolling; A model selection module is configured to select a corresponding thickness prediction model constructed based on a variable length memory (LSTM) model according to the pole piece and the diaphragm; A thickness prediction module is configured to input the electrode thickness and the diaphragm thickness at the rolling moment into the corresponding thickness prediction model to obtain the electrode thickness and the diaphragm thickness at a future target moment; A spacing calculation module is configured to calculate the tab spacing according to the pole piece thickness and the diaphragm thickness at a future target moment; The tab alignment module is configured to perform tab alignment operations at a future target time according to the tab spacing.

7. The tab alignment device according to claim 6, characterized in that: The thickness prediction model includes a positive electrode sheet thickness prediction model, a negative electrode sheet thickness prediction model and a diaphragm thickness prediction model constructed based on a variable length memory (LSTM) model.

8. An electronic device, characterized in that: including processor and storage medium; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1-5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images