Data self-generating electrode performance prediction method

Abstract

The invention discloses a data self-generation electrode performance prediction method, which belongs to the technical field of electrochemistry, and comprises the following steps: automatically extracting numerical values conforming to variance distribution of electrode physical property parameters, constructing a geometric model conforming to electrode spatial characteristics, and defining physical property parameters conforming to electrode material properties; all physical fields generated in the electrochemical process are coupled, a finite element method is used for solving and outputting the model, an electrode key performance parameter data set is established, and electrode performance prediction is achieved. According to the method, a mode of analyzing the electrochemical behavior of the electrode by utilizing multiple physical fields has a good coupling interface, and correction of an evaluation mode can be realized by supplementing more physical fields in the later period. The method not only reduces the high cost of a high-sensitivity sensor, but also improves the precision of electrode performance prediction, and greatly optimizes the design process and manufacturing cost of an energy storage device.

Description

Technical Field

[0001] The invention belongs to the field of electrochemical technology, and in particular relates to an electrode performance prediction method with self-generated data. Background Art

[0002] The rapid development of science and technology in modern society has not only greatly facilitated people's lives, but also placed higher and higher requirements on energy storage devices. As the core component of energy storage devices, electrodes determine the performance of the entire device and the safety of its use. At present, researchers have proposed many electrode performance prediction models to guide the design of energy storage devices, but due to the complexity of the electrochemical reaction environment, high-quality battery performance data sets are difficult to collect. Therefore, how to break through the traditional method of evaluating the charge and discharge state of the electrode and achieve high-accuracy prediction of electrode performance is a core issue of concern to the industry today.

[0003] Generally speaking, battery control systems use numerous sensors to obtain parameters such as the internal temperature, pressure, and potential of the electrode to determine the electrode state and predict the electrode behavior. However, this method not only requires additional sensor expenditures, but the sampling time and frequency domain accuracy is often not well matched with the electrochemical reaction process, which limits the development of high-precision electrode performance prediction algorithms.

[0004] In response to the above problems, CN115445971A discloses an electrode screening method, which performs impedance and polarization tests on the electrode-assembled battery, and compares the obtained electrochemical data with the database, thereby realizing the screening and evaluation of the electrode; CN 112560282B discloses a battery electrode design and computer equipment, which introduces statistics and obtains a design scheme with production guidance significance by combining probability distribution with multiple groups of Pareto optimal solution simulations. However, the evaluation of the electrode in the above schemes requires the electrode to undergo actual electrochemical testing. The model screening process is related to specific production lines, processes and batches. The model has poor universality and there are many black box models in the screening process that lack electrochemical explanations in principle. The method of the invention has good adaptability based on the characteristic modeling of electrodes in different systems, and also self-generates data sets through multi-physical fields that conform to actual electrochemical processes, so that the electrode life can be effectively predicted. Summary of the invention

[0005] In view of the problem that the existing electrode performance prediction algorithm is difficult to improve due to the lack of high-quality electrode data sets, the present invention provides a data self-generated electrode performance prediction method. This method not only reduces the high cost of high-sensitivity sensors, but also improves the accuracy of electrode performance prediction, greatly optimizing the design process and manufacturing cost of energy storage devices.

[0006] The present invention is achieved through the following technical solutions:

[0007] A method for predicting electrode performance by self-generating data, comprising:

[0008] Automatically extract values ​​that conform to the variance distribution of electrode physical property parameters, build a geometric model that conforms to the spatial characteristics of the electrode, and define physical property parameters that conform to the properties of the electrode material;

[0009] By coupling the various physical fields occurring in the electrochemical process, the finite element method is used to solve and output the model, and a data set of key performance parameters of the electrode is established to achieve electrode performance prediction.

[0010] Furthermore, the method specifically comprises the following steps:

[0011] Step 1: fitting the distribution function based on the electrode morphology parameters and physical property parameters, and extracting in a form that conforms to the electrode characteristics;

[0012] Step 2: constructing an electrode geometric model using the extracted morphological parameters, wherein the geometric model is consistent with the electrode spatial characteristics; assigning values ​​to the geometric domain using the extracted physical property parameters, wherein the assigned material properties are consistent with the actual electrode material characteristics;

[0013] Step 3: Clearly predict the role and coordination of various physical fields in the electrode electrochemical reaction process, and couple the various physical fields involved in the reaction;

[0014] Step 4: Based on the geometric model of the electrode, solve the multiple physical field equations that have been clarified, set the spatial domain and define the boundary conditions to construct the electrochemical model of the electrode.

[0015] Step 5: Complete the solution of the electrochemical model of the electrode and output the electrode performance parameters. Repeat the process of extracting the electrode space parameters and material parameters. After multiple solutions, the electrode performance parameter data set is constructed to achieve the prediction of electrode performance and life under different working conditions.

[0016] Furthermore, in step 1, the electrode geometric parameters include: electrode particle size, length, through-hole radius, electrode group distribution ratio, roughness, porosity, density and thickness;

[0017] The physical property parameters include: porosity, electrode particle electronic conductivity, electrode particle ionic conductivity, electrochemical kinetic parameters, specific heat capacity, water content, surface tension and hardness.

[0018] Furthermore, in step one, the morphology parameters are obtained by characterization tests, including scanning electron microscopy, atomic force microscopy, transmission electron microscopy, X-ray diffraction spectrum, Rama spectrum, and X-ray photoelectron spectroscopy; the physical property parameters are obtained by experiments, such as AC impedance test, Tafel test, thermogravimetric test, tortuosity test, capacity test, and rate test.

[0019] Furthermore, in step three, the physical field includes electric field, magnetic field, temperature field, stress field, concentration field, flow field, potential field and structural field.

[0020] Furthermore, in step 4, the finite element analysis method is used to solve the electrode model and output the electrode performance parameters.

[0021] Furthermore, in step five, the electrode performance parameters are predicted based on an artificial intelligence algorithm.

[0022] Compared with the prior art, the advantages of the present invention are as follows:

[0023] 1. The present invention solves the problem that the traditional electrode performance prediction algorithm is limited by the quality of the data set and the prediction accuracy of electrode performance parameters is poor. The prediction results show that under the same method, it has better accuracy than the traditional data set;

[0024] 2. The dataset generation method of the present invention is independent of the type of electrode, because it can evaluate different types of electrodes by changing the modeling method, and has better generalization ability;

[0025] 3. The method of analyzing the electrochemical behavior of electrodes using multiple physical fields in the present invention has a good coupling interface, and the evaluation method can be corrected by adding more physical fields in the later stage. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0027] Figure 1 A schematic flow chart of a method for predicting electrode performance by self-generating data according to the present invention;

[0028] Figure 2 It is a schematic diagram of the extraction and fitting of the electrode physical property parameters and spatial parameters of the present invention;

[0029] Figure 3 An electrode multiphysics model constructed for the present invention;

[0030] Figure 4 The figure is a comparison between the theoretical prediction effect of the electrode of the present invention and the actual test result. DETAILED DESCRIPTION

[0031] In order to clearly and completely describe the technical solution and its specific working process of the present invention, the specific implementation methods of the present invention are as follows in conjunction with the accompanying drawings of the specification:

[0032] Example 1

[0033] like Figure 1 and Figure 2 As shown, this embodiment provides a method for predicting electrode performance by self-generating data, and the prediction method includes:

[0034] Automatically extract values ​​that conform to the variance distribution of electrode physical property parameters, build a geometric model that conforms to the spatial characteristics of the electrode, and define physical property parameters that conform to the properties of the electrode material;

[0035] By coupling the various physical fields occurring in the electrochemical process, the finite element method is used to solve and output the model, and a data set of key performance parameters of the electrode is established to achieve electrode performance prediction.

[0036] The specific steps include:

[0037] Step 1: Fitting the distribution function based on the electrode geometric parameters and physical property parameters, and extracting in a form that conforms to the electrode characteristics;

[0038] Step 2: Establish a geometric model using the extracted electrode morphology parameters, wherein the geometric model is consistent with the electrode spatial characteristics; define a geometric domain using the extracted physical property parameters, wherein the material domain is consistent with the material characteristics of the electrode;

[0039] Step 3: Clearly predict the role and coordination of various physical fields in the electrode electrochemical reaction process, couple the various physical fields involved in the reaction, and define the physical field equations based on the constructed geometry and material parameters to complete the establishment of the electrode electrochemical model;

[0040] Step 4: Solve the above electrochemical model and output the electrode performance parameters, repeat the material parameter extraction process to complete the construction of the electrode performance parameter data set, so as to realize the prediction of electrode performance and life under different working conditions.

[0041] In this embodiment, the Si electrode is taken as the prediction object, and the specific steps are as follows:

[0042] S1. Self-construction of electrode geometry model: Si electrode is used as the prediction object, based on the material information of the electrode, including size (diameter 12 mm, thickness 200 um), porosity (13%), density (1.9 g cm -1 ), lattice structure, etc. to fit the corresponding functions, and extract and construct a geometric model that conforms to the spatial characteristics of the electrode;

[0043] S2. Multi-physics coupling of electrochemical processes, such as Figure 3As shown: Based on the huge volume change of Si electrode during the electrochemical process, the various physical fields involved in the electrochemical reaction, including electric field, magnetic field, temperature field, stress field, concentration field, flow field, potential field, and structural field, are coupled to describe the electrochemical process of the electrode;

[0044] S3. Customization of electrode physical property parameters: Fitting function based on material parameters of Si electrode, including conductivity, ion migration coefficient, electrochemical kinetic parameters, relative molecular mass, diffusion coefficient, and ion mobility to define material parameters of combined electrode physical property characteristics;

[0045] S4. Solving the above-constructed geometry, established equations, and defined parameters using the finite element method to construct an electrode electrochemical behavior data set and train a prediction model for the electrode electrochemical performance;

[0046] S5, the electrode to be studied and the lithium sheet are combined into a half-cell, and the secondary electrolyte (1MLiPF 6 In EC:EMC:DEC=1:1:1Vol%), the battery was tested under the condition of 1C full charge and 1C discharge until the capacity of the lithium battery decayed to 80% of the initial capacity, and the number of cycles was recorded. The model output predicted 70 cycles, and the actual electrochemical performance test was 68 cycles.

[0047] Example 2

[0048] This embodiment provides a method for predicting electrode performance by self-generating data, taking a lithium-rich manganese-based positive electrode sheet as a prediction object, and specifically includes the following steps:

[0049] S1. Self-construction of electrode geometry model: Taking lithium-rich manganese-based positive electrode as the prediction object, based on the material information of the electrode, including thickness (300um), tortuosity (0.88), roughness (Ra=1.6), coating density (3mg cm -2 ), specific surface area (4584cm 2 g -1 ), the lattice structure self-constructs a geometric model that conforms to the spatial characteristics of the electrode;

[0050] S2. Multi-physics field coupling of electrochemical process: Based on the working environment of lithium-rich manganese-based electrodes, various physical fields involved in electrochemical reactions are coupled, including electric field, magnetic field, temperature field, stress field, concentration field, flow field, potential field, and structural field, so as to describe the electrochemical process of the electrode;

[0051] S3. Customization of electrode physical properties: Based on the material information of the electrode, including conductivity, ion migration coefficient, electrochemical kinetic parameters, relative molecular mass, diffusion coefficient, and ion mobility, the material parameters that define the physical properties of the combined electrode are defined;

[0052] S4, using the finite element method to solve the above-constructed geometry, established equations, and defined parameters for analysis of the electrochemical behavior of the electrode;

[0053] The predicted results show that the discharge capacity of the electrode at 0.1C, 1C and 3C is 257 mAh g -1 , 223mAh g -1 , 184mAh g -1 The discharge capacities predicted by the classical neural network model for the relevant parameters are 254 mAh g -1 , 241mAh g -1 , 219mAh g -1 Electrochemical data show that the actual capacity of the electrode in actual battery testing is 256 mAh g -1 , 229mAh g -1 , 181mAh g -1 Therefore, it is shown that this method has good accuracy and universality in predicting electrode performance parameters.

[0054] S5. The electrodes to be studied are combined with commercial graphite negative electrodes to form a lithium-ion battery, and the electrolyte is a secondary electrolyte (1M LiPF 6 In EC:EMC:DEC=1:1:1Vol%), the battery was fully charged at 1C and discharged at 1C until the capacity of the lithium battery decayed to 80% of the initial capacity, and the number of cycles was recorded. The model output predicted 300 cycles, and the actual electrochemical performance test was 320 cycles.

[0055] The preferred embodiments of the present invention are described in detail above in conjunction with the accompanying drawings. However, the present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention, a variety of simple modifications can be made to the technical solution of the present invention, and these simple modifications all belong to the protection scope of the present invention.

[0056] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations.

[0057] In addition, various embodiments of the present invention may be arbitrarily combined, and as long as they do not violate the concept of the present invention, they should also be regarded as the contents disclosed by the present invention.

Claims

1. A method for predicting electrode performance by self-generating data, characterized in that: include: Automatically extract values ​​that conform to the variance distribution of electrode physical property parameters, build a geometric model that conforms to the spatial characteristics of the electrode, and define physical property parameters that conform to the properties of the electrode material; By coupling the various physical fields occurring in the electrochemical process, the finite element method is used to solve and output the model, and a data set of key performance parameters of the electrode is established to achieve electrode performance prediction.

2. The electrode performance prediction method of self-generated data according to claim 1, characterized in that: The method specifically comprises the following steps: Step 1: fitting the distribution function based on the electrode morphology parameters and physical property parameters, and extracting in a form that conforms to the electrode characteristics; Step 2: constructing an electrode geometric model using the extracted morphological parameters, wherein the geometric model is consistent with the electrode spatial characteristics; assigning values ​​to the geometric domain using the extracted physical property parameters, wherein the assigned material properties are consistent with the actual electrode material characteristics; Step 3: Clearly predict the role and coordination of various physical fields in the electrode electrochemical reaction process, and couple the various physical fields involved in the reaction; Step 4: Based on the geometric model of the electrode, solve the multiple physical field equations that have been clarified, set the spatial domain and define the boundary conditions to construct the electrochemical model of the electrode. Step 5: Complete the solution of the electrochemical model of the electrode and output the electrode performance parameters. Repeat the process of extracting the electrode space parameters and material parameters. After multiple solutions, the electrode performance parameter data set is constructed to achieve the prediction of electrode performance and life under different working conditions.

3. The electrode performance prediction method of self-generated data according to claim 2, characterized in that: In step 1, the electrode geometric parameters include: electrode particle size, length, through-hole radius, electrode group distribution ratio, roughness, porosity, density and thickness; The physical property parameters include: porosity, electrode particle electronic conductivity, electrode particle ionic conductivity, electrochemical kinetic parameters, specific heat capacity, water content, surface tension and hardness.

4. The electrode performance prediction method of self-generated data according to claim 2, characterized in that: In step one, the morphology parameters are obtained by characterization tests, including scanning electron microscopy, atomic force microscopy, transmission electron microscopy, X-ray diffraction spectrum, Rama spectrum, and X-ray photoelectron spectroscopy; the physical property parameters are obtained by experiments, such as AC impedance test, Tafel test, thermogravimetric test, tortuosity test, capacity test, and rate test.

5. The electrode performance prediction method of self-generated data according to claim 2, characterized in that: In step three, the physical fields include electric field, magnetic field, temperature field, stress field, concentration field, flow field, potential field and structural field.

6. The electrode performance prediction method of self-generated data according to claim 2, characterized in that: In step 4, the finite element analysis method is used to solve the electrode model and output the electrode performance parameters.

7. The electrode performance prediction method of self-generated data according to claim 2, characterized in that: In step five, the electrode performance parameters are predicted based on artificial intelligence algorithms.

Images