Cyclic volt-ampere curve prediction method, device, equipment, medium and product
By constructing a sample set based on historical cyclic current-voltage curves and a deep learning model, the problems of high computational cost and low accuracy in fuel cell life prediction were solved, achieving efficient and accurate fuel cell life assessment and optimization.
Patent Information
- Application Number
- CN202510916019.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-03
- Publication Date
- 2025-11-14
AI Technical Summary
Existing technologies suffer from high computational complexity and limited prediction accuracy in fuel cell life prediction. Furthermore, experimental measurement of CV curves is costly and lacks real-time performance, making it difficult to achieve efficient fuel cell life assessment and optimization.
A sample set is constructed based on historical cyclic voltammetry curves. A time series prediction model based on deep learning is adopted. The model is fused by stacking CatBoost, RF and lightGBM models. The prediction models of the upper and lower halves of the curves are trained separately to achieve accurate prediction.
It significantly reduced experimental time, improved prediction accuracy, and enabled efficient assessment and optimization of fuel cell lifespan, with a prediction accuracy of 98.7%.
Smart Images
Figure CN120951525A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of battery performance prediction technology, and in particular to a method, apparatus, equipment, medium and product for predicting cyclic voltammetry curves. Background Technology
[0002] A proton exchange membrane fuel cell (PEMFC) is a highly efficient, low-carbon energy conversion device. Its lifespan is affected by various factors, including electrode material degradation, catalyst decay, membrane damage, operating conditions (temperature, humidity, pressure), and load variations. Cyclic voltammetry (CV) is an important electrochemical testing method used to evaluate fuel cell catalyst activity, electrode degradation, and electrochemical reaction kinetics. By analyzing the changes in CV curves at different stages of the lifespan, the health status of the PEMFC can be effectively determined, and its lifespan decline trend can be predicted.
[0003] Traditional fuel cell lifetime prediction methods primarily rely on physical models and experimental data fitting. While these methods can reflect key electrochemical processes to some extent, they often suffer from high computational costs and limited prediction accuracy when dealing with complex nonlinear dynamic characteristics. Furthermore, experimentally measuring cyclic voltammetry (CV) curves requires significant time and cost, limiting their real-time performance and scalability in practical applications. Therefore, how to utilize efficient data-driven methods to accurately predict cyclic voltammetry curve changes, thereby enabling fuel cell lifetime assessment and optimization, has become a current research hotspot. Summary of the Invention
[0004] The purpose of this application is to provide a method, apparatus, device, medium, and product for predicting cyclic voltammetry curves, so as to achieve accurate prediction of cyclic voltammetry curve changes using an efficient data-driven method, thereby enabling fuel cell life assessment and optimization.
[0005] To achieve the above objectives, this application provides the following solution.
[0006] Firstly, this application provides a method for predicting cyclic voltammetry curves, including:
[0007] Based on historical cyclic voltammetry curves, a first sample set and a second sample set are constructed. The cyclic voltammetry curves include an upper half and a lower half, both of which are polarimetric curves. The first sample set includes multiple first sample data points, which are input data based on historical second time points, one or more first time points, and feature point data on the upper half of the first time point's curve, with the feature point data on the upper half of the second time point serving as the label. The second sample set includes multiple second sample data points, which are input data based on historical second time points, one or more first time points, and feature point data on the lower half of the first time point's curve, with the feature point data on the lower half of the second time point serving as the label.
[0008] The fusion model is trained using the first sample set and the second sample set respectively, and the fusion model trained with the first sample set and the fusion model trained with the second sample set are used as the upper half curve prediction model and the lower half curve prediction model respectively.
[0009] Cyclic voltammetry curves are predicted using the upper half curve prediction model and the lower half curve prediction model.
[0010] Secondly, this application provides a cyclic voltammetry curve prediction device, which applies the above-mentioned cyclic voltammetry curve prediction method, and the cyclic voltammetry curve prediction device includes:
[0011] A sample set construction module is used to construct a first sample set and a second sample set based on historical cyclic voltammetry curves. The cyclic voltammetry curves include an upper half and a lower half, both of which are polarimetric curves. The first sample set includes multiple first sample data points, which take historical second time points, one or more first time points, and feature point data on the upper half of the first time point as input data, and feature point data on the upper half of the second time point as labels. The second sample set includes multiple second sample data points, which take historical second time points, one or more first time points, and feature point data on the lower half of the first time point as input data, and feature point data on the lower half of the second time point as labels.
[0012] The training module is used to train the fusion model using the first sample set and the second sample set respectively, to obtain the fusion model trained on the first sample set and the fusion model trained on the second sample set, which are used as the upper half curve prediction model and the lower half curve prediction model respectively.
[0013] The prediction module is used to predict cyclic voltammetry curves using the upper half curve prediction model and the lower half curve prediction model.
[0014] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described cyclic current-voltage curve prediction method.
[0015] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method for predicting cyclic current-voltage curves.
[0016] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the above-described cyclic current-voltage curve prediction method.
[0017] According to the specific embodiments provided in this application, this application has the following technical effects.
[0018] This application provides a method, apparatus, device, medium, and product for predicting cyclic voltammetry curves. Based on historical cyclic voltammetry curves, this application constructs a first sample set and a second sample set. A fusion model is trained using both the first and second sample sets to obtain fusion models trained on the first and second sample sets, respectively serving as the upper and lower half of the curve prediction model. The cyclic voltammetry curve is then predicted using these two models. This application, through sample construction and model training, obtains the upper and lower half of the curve prediction models, avoiding the unpredictable situation caused by multiple currents corresponding to the same voltage. It achieves accurate prediction of cyclic voltammetry curve changes using an efficient data-driven method, thereby enabling fuel cell life assessment and optimization. Attached Figure Description
[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0020] Figure 1 This is a flowchart illustrating a method for predicting cyclic voltammetry curves according to an embodiment of this application.
[0021] Figure 2 An example diagram of cyclic voltammetry curves provided in an embodiment of this application.
[0022] Figure 3 This is a schematic diagram of the division of cyclic voltammetry curves provided in an embodiment of this application.
[0023] Figure 4 This is a schematic diagram illustrating the principle of model training provided in one embodiment of this application.
[0024] Figure 5 This is a schematic diagram of the prediction results provided in one embodiment of this application.
[0025] Figure 6 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation
[0026] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.
[0027] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0028] Machine learning technology has shown great potential in fuel cell performance prediction and lifetime analysis. Deep learning-based time-series prediction models can automatically extract key features from large amounts of experimental data, capture complex nonlinear relationships in the evolution of cyclic voltammetry (CV) curves, and improve prediction accuracy and generalization ability. This application proposes a method, apparatus, device, medium, and product for predicting cyclic voltammetry (CV) curves over a lifetime, involving a process for predicting CV curves over a certain lifetime and a machine learning model (fusion model). This fusion model can accurately predict the CV curve over a certain lifetime, and the method can significantly reduce experimental time.
[0029] In one exemplary embodiment, this application provides a method for predicting cyclic voltammetry curves, such as... Figure 1 As shown, it includes the following steps 101-103.
[0030] Step 101: Based on historical cyclic voltammetry curves, construct a first sample set and a second sample set; the cyclic voltammetry curve includes an upper half curve and a lower half curve, both of which are polarimetric curves; the first sample set includes multiple first sample data, which take historical second time points, one or more first time points, and feature point data on the upper half curve of the first time points as input data, and feature point data on the upper half curve of the second time points as labels; the second sample set includes multiple second sample data; the second sample data take historical second time points, one or more first time points, and feature point data on the lower half curve of the first time points as input data, and feature point data on the lower half curve of the second time points as labels.
[0031] Step 102: Train the fusion model using the first sample set and the second sample set respectively to obtain the fusion model trained on the first sample set and the fusion model trained on the second sample set, which are used as the upper half curve prediction model and the lower half curve prediction model respectively.
[0032] Step 103: Use the upper half curve prediction model and the lower half curve prediction model to predict the cyclic voltammetry curve.
[0033] By implementing steps 101-103 above, we can accurately predict changes in cyclic voltammetry curves using efficient data-driven methods, thereby enabling fuel cell life assessment and optimization.
[0034] In another exemplary embodiment, step 101 above can be replaced by steps 201-203.
[0035] Step 201: Collect lifetime operation data of proton exchange membrane fuel cell (PEMFC) and collect the cyclic voltammetry curve of PEMFC at regular intervals.
[0036] Step 202 involves processing the feature point data of the collected cyclic voltammetry curves. Different sampling densities are used for different regions to ensure that the error of the cyclic voltammetry curves before and after sampling is within a controllable range, while reducing the data volume by approximately 70% and ensuring that the processed data fits the original data well. Specific results are as follows: Figure 2 As shown.
[0037] Step 203 involves "block processing" of the feature point data processed in step 202, that is, dividing the cyclic voltammetry curve into upper and lower parts, as shown in the figure. Figure 3As shown, for the upper half of the curve, 0-0.4 represents the hydrogen desorption region, 0.4-0.7 represents the electric double layer region, and 0.7-1 represents the platinum oxidation region; for the lower half of the curve, 1-0.7 represents the platinum reduction region, 0.7-0.4 represents the electric double layer region, and 0.4-0 represents the hydrogen adsorption region. This method transforms the cyclic voltammetry curve into a "quasi-polarization curve," avoiding the unpredictable situation caused by multiple currents corresponding to the same voltage.
[0038] In another exemplary embodiment, two base models, CatBoost and RF (Random Forest), are used, and the meta-model is a stacking fusion model of the LightGBM model.
[0039] In another exemplary embodiment, step 102 above employs cross-validation. For example, the sample set is divided into 5 parts, and each time four parts are used as the training set and one part as the validation set to train the base model (CatBoost+RF), as follows: Figure 4 As shown, after the base model is trained, its parameters are fixed, and its output is used as the input to the meta-model. The meta-model is then trained using the entire sample set. Step 102 can be replaced by steps 301-307 as follows.
[0040] Step 301: Divide the target sample set into multiple sample subsets; the target sample set is either the first sample set or the second sample set.
[0041] Step 302: Use one sample subset from the multiple sample subsets as the validation set, and use the sample subsets other than the validation set as the training set.
[0042] Step 303: Use the training set to train the two base models in the fusion model respectively to obtain two trained base models.
[0043] Step 304: Validate the two trained base models using the validation set.
[0044] Step 305: If the verification fails, replace another sample subset in the multiple sample subsets as the verification set, and use the sample subset other than the verification set as the training set, then return to the step of "training two base models using the training set to obtain two trained base models".
[0045] Step 306: When the verification is successful, output the two trained base models as two well-trained base models.
[0046] Step 307: Fix the parameters of the two trained base models, and use the target sample set to train the meta-model in the fusion model to obtain the fusion model trained on the target sample set.
[0047] The Catboost model consists of T decision trees, each with a value of f. t The final output is obtained by summing the predicted values:
[0048]
[0049] in, The initial prediction result for the Catboost model is f. t (x m x is the predicted value of the t-th tree in the Catboost model. m The input data is the m-th target sample data, and η is the learning rate, which controls the contribution of each tree. In this embodiment, it is 0.05. The target sample data is either the first sample data or the second sample data.
[0050] RF model: The predictions of this model are the average of the predictions from a set of T' trees.
[0051]
[0052] in, For the initial prediction results of the RF model, f t' (x m ) represents the predicted value of the t'th tree in the RF model.
[0053] The outputs of the two base models above serve as the input features of the meta-model:
[0054]
[0055] The output of the final meta-model is the final predicted value:
[0056]
[0057] in, The fusion prediction results of the lightGBM model. This represents the fusion prediction result of the t-th tree in the lightGBM model.
[0058] During the training phase, both the base model and the meta-model use MSE as the loss function for parameter optimization.
[0059]
[0060] Where MSE is the mean squared error loss function, N is the number of feature point data on the target curve at the second time point, and y i The label for the i-th feature point data on the target curve at the second time point. The prediction result of the i-th feature point data on the target curve at the second time point output by the target model, wherein the target model is any one of the two base models and the meta-model, and the target curve is the upper half curve or the lower half curve.
[0061] In another exemplary embodiment, during prediction, predictions are made using the upper half curve prediction model and the lower half curve prediction model respectively, and the prediction results are finally combined into a complete lifetime cyclic voltammetry curve.
[0062] The final predicted value is obtained using R. 2 Evaluate the prediction accuracy of the trained fusion model:
[0063]
[0064] Among them, R 2 For R 2 Evaluation results, y i,j Let i be the label of the i-th feature point data on the target curve at the second time point in the j-th target sample data. This represents the prediction result of the i-th feature point data on the target curve at the second time point output by the meta-model. The average value of the label of the i-th feature point data on the target curve at the second time point in each target sample data is given by N, where N represents the number of feature point data on the target curve at the second time point, and M represents the number of target sample data. The target sample data is either the first sample data or the second sample data. When the target sample data is the first sample data, the target curve is the upper half of the curve. When the target sample data is the second sample data, the target curve is the lower half of the curve.
[0065] The final results indicate good accuracy in predicting cyclic voltammetry curves during PEMFC lifetime testing. When the test set proportion is 30%, the upper half of the curve data shows good accuracy (R). 2 The lower half of the curve data is 98.02%, R. 2 It is 99.40%, with an average R 2 The result was 98.7%. Figure 5 As shown.
[0066] In the subsequent prediction process, the predicted future time points and the feature point data of the known cyclic voltammetry curves are input into the trained fusion model. First, the trained base model makes the prediction. Then, the outputs of all base models are used as the inputs of the meta-model. Finally, the meta-model makes the final prediction based on the outputs of the base models.
[0067] Based on the same inventive concept, this application also provides a cyclic voltammetry curve prediction device for implementing the cyclic voltammetry curve prediction method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the cyclic voltammetry curve prediction device provided below can be found in the limitations of the cyclic voltammetry curve prediction method described above, and will not be repeated here.
[0068] In one exemplary embodiment, a cyclic voltammetry curve prediction device is provided, comprising:
[0069] A sample set construction module is used to construct a first sample set and a second sample set based on historical cyclic voltammetry curves. The cyclic voltammetry curve includes an upper half curve and a lower half curve, both of which are polarimetric curves. The first sample set includes multiple first sample data, which take historical second time points, one or more first time points, and feature point data on the upper half curve of the first time points as input data, and feature point data on the upper half curve of the second time points as labels. The second sample set includes multiple second sample data, which take historical second time points, one or more first time points, and feature point data on the lower half curve of the first time points as input data, and feature point data on the lower half curve of the second time points as labels.
[0070] The training module is used to train the fusion model using the first sample set and the second sample set respectively, to obtain the fusion model trained on the first sample set and the fusion model trained on the second sample set, which are used as the upper half curve prediction model and the lower half curve prediction model respectively.
[0071] The prediction module is used to predict cyclic voltammetry curves using the upper half curve prediction model and the lower half curve prediction model.
[0072] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 6As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media to run. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a cyclic voltammetry curve prediction method.
[0073] Those skilled in the art will understand that Figure 6 The structures shown are merely block diagrams of some structures related to the present application and do not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements. In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.
[0074] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.
[0075] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.
[0076] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.
[0077] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).
[0078] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.
[0079] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0080] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.
Claims
1. A method for predicting cyclic voltammetry curves, characterized in that, include: Based on historical cyclic voltammetry curves, a first sample set and a second sample set are constructed. The cyclic voltammetry curves include an upper half and a lower half, both of which are polarimetric curves. The first sample set includes multiple first sample data points, which are input data based on historical second time points, one or more first time points, and feature point data on the upper half of the first time point's curve, with the feature point data on the upper half of the second time point serving as the label. The second sample set includes multiple second sample data points, which are input data based on historical second time points, one or more first time points, and feature point data on the lower half of the first time point's curve, with the feature point data on the lower half of the second time point serving as the label. The fusion model is trained using the first sample set and the second sample set respectively, and the fusion model trained with the first sample set and the fusion model trained with the second sample set are used as the upper half curve prediction model and the lower half curve prediction model respectively. Cyclic voltammetry curves are predicted using the upper half curve prediction model and the lower half curve prediction model.
2. The cyclic voltammetry curve prediction method according to claim 1, characterized in that, The fusion model includes two base models and one meta-model, and both base models are connected to the meta-model. Both base models are used to predict feature point data on the upper or lower half of the curve at the second time point based on the input data, and obtain two initial prediction results; The meta-model is used to fuse the two initial prediction results to obtain a fused prediction result.
3. The cyclic voltammetry curve prediction method according to claim 2, characterized in that, The two base models are the CatBoost model and the RF model, and the meta-model is the lightGBM model.
4. The cyclic voltammetry curve prediction method according to claim 2, characterized in that, The fusion model is trained using the first sample set and the second sample set respectively, resulting in a fusion model trained on the first sample set and a fusion model trained on the second sample set. These are used as the upper half curve prediction model and the lower half curve prediction model, respectively. Specifically, this includes: The target sample set is divided into multiple sample subsets; the target sample set is either a first sample set or a second sample set. Use one subset of samples from multiple sample subsets as the validation set, and use the other subsets of samples as the training set. The two base models in the fusion model are trained using the training set to obtain two trained base models. The two trained base models were validated using the validation set. If the verification fails, replace another sample subset in the multiple sample subsets as the verification set, use the sample subset other than the verification set as the training set, and return to the step of "training two base models using the training set to obtain two trained base models". When the verification passes, output the two trained base models as two well-trained base models; By fixing the parameters of two trained base models, the meta-model in the fusion model is trained using the target sample set to obtain the fusion model trained on the target sample set.
5. The cyclic voltammetry curve prediction method according to claim 4, characterized in that, The loss function used when training the two base models and the loss function used when training the meta-model are both mean squared error loss functions; The mean squared error loss function is: Where MSE is the mean squared error loss function, N is the number of feature point data on the target curve at the second time point, and y i The label for the i-th feature point data on the target curve at the second time point. The prediction result of the i-th feature point data on the target curve at the second time point output by the target model, wherein the target model is any one of the two base models and the meta-model, and the target curve is the upper half curve or the lower half curve.
6. The cyclic voltammetry curve prediction method according to claim 4, characterized in that, With the parameters of two pre-trained base models fixed, the meta-model in the fusion model is trained using the target sample set to obtain the fusion model trained on the target sample set. This process further includes: Using R 2 Evaluate the prediction accuracy of the fusion model trained on the target sample set; R 2 The calculation formula is: Among them, R 2 For R 2 Evaluation results, y i,j Let i be the label of the i-th feature point data on the target curve at the second time point in the j-th target sample data. This represents the prediction result of the i-th feature point data on the target curve at the second time point output by the meta-model. The average value of the label of the i-th feature point data on the target curve at the second time point in each target sample data is given by N, where N represents the number of feature point data on the target curve at the second time point, and M represents the number of target sample data. The target sample data is either the first sample data or the second sample data. When the target sample data is the first sample data, the target curve is the upper half of the curve. When the target sample data is the second sample data, the target curve is the lower half of the curve.
7. A cyclic voltammetry curve prediction device, characterized in that, The cyclic voltammetry curve prediction device applies the cyclic voltammetry curve prediction method according to any one of claims 1-6, and the cyclic voltammetry curve prediction device comprises: A sample set construction module is used to construct a first sample set and a second sample set based on historical cyclic voltammetry curves. The cyclic voltammetry curves include an upper half and a lower half, both of which are polarimetric curves. The first sample set includes multiple first sample data points, which take historical second time points, one or more first time points, and feature point data on the upper half of the first time point as input data, and feature point data on the upper half of the second time point as labels. The second sample set includes multiple second sample data points, which take historical second time points, one or more first time points, and feature point data on the lower half of the first time point as input data, and feature point data on the lower half of the second time point as labels. The training module is used to train the fusion model using the first sample set and the second sample set respectively, to obtain the fusion model trained on the first sample set and the fusion model trained on the second sample set, which are used as the upper half curve prediction model and the lower half curve prediction model respectively. The prediction module is used to predict cyclic voltammetry curves using the upper half curve prediction model and the lower half curve prediction model.
8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the cyclic current-ampere curve prediction method according to any one of claims 1-6.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the cyclic current-ampere curve prediction method according to any one of claims 1-6.
10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the cyclic current-ampere curve prediction method according to any one of claims 1-6.