Design method and system of solid electrolyte buffer layer

Abstract

The invention provides a method and a system for designing a solid electrolyte buffer layer, which are applied to the field of solid electrolyte buffer layer design. The method comprises the steps that the surface of a solid electrolyte buffer layer is regarded as an infinitesimal circle, and the corresponding relation among the comprehensive internal resistance of the solid electrolyte buffer layer, the thickness of the solid electrolyte buffer layer and the assembling force is obtained based on the relation between the surface area and the assembling force of the infinitesimal circle and the relation between the surface area of the infinitesimal circle and the thickness of the solid electrolyte buffer layer; constructing a regression model by taking the corresponding relation as a constraint function, and solving the regression model by taking the thickness and the assembly force of the solid electrolyte buffer layer as input variables of the regression model, taking the comprehensive internal resistance as output and taking minimization of the comprehensive internal resistance in a value range as a target to obtain the solid electrolyte buffer layer. And the optimal matching design of the comprehensive internal resistance, the solid electrolyte buffer layer thickness and the assembly force is obtained. Coupling design of the thickness, the assembly force and the comprehensive internal resistance of the solid electrolyte buffer layer can be realized, so that the comprehensive internal resistance is minimum.

Description

Technical Field

[0001] The present invention relates to the field of solid electrolyte buffer layer design, and in particular to a solid electrolyte buffer layer design method and system. Background Art

[0002] As a new type of battery technology, solid-state batteries use solid electrolytes instead of traditional liquid electrolytes and have many advantages, including high safety, higher energy density, longer cycle life, faster charging speed and wider temperature adaptability range. These advantages have brought potential revolutionary changes to electric vehicles, mobile devices and other fields. On this basis, reducing the internal resistance of solid-state batteries will not only help improve battery efficiency and extend battery life, but also effectively ensure battery safety.

[0003] The internal resistance of a solid-state battery is mainly composed of two parts: the battery material resistance and the interface contact resistance. The solid electrolyte buffer layer is located between the electrolyte and the electrode, and plays the role of isolating and balancing the charge. Increasing the thickness of the solid electrolyte buffer layer has opposite dual effects on battery performance. First, increasing the thickness of the solid electrolyte buffer layer is beneficial to increasing the electrode contact area, thereby reducing the interface contact resistance; however, the conductivity of the solid electrolyte buffer layer is low, and increasing the thickness of the solid electrolyte will increase the battery material resistance. Generally speaking, the design range of the solid electrolyte buffer layer is between a few microns and several hundred microns. At the same time, the assembly force during battery assembly will affect the interface contact area between the solid electrolyte buffer layer and the electrode, thereby affecting the battery internal resistance. Therefore, the solid electrolyte thickness, assembly force and solid electrolyte internal resistance are mutually influential.

[0004] Prior art, such as Chinese patent CN118412069A discloses a method for determining the optimal preload force for solid-state batteries. This solution obtains the battery interface contact resistance by establishing a three-dimensional electrolyte model. However, this method is complex to model, and many assumptions need to be set during the modeling process. The rationality of the assumptions directly affects the simulation results, and the subjective influence is relatively large. Since it cannot consider the battery contact area, it is impossible to obtain the comprehensive impact of the material resistance caused by the thickness of the buffer layer on the battery internal resistance; Chinese patent CN117199562A discloses a method for improving the cycle life of a battery based on dynamically adjusting the stacking stress. It changes the contact area between the internal pressure of the battery and the electrolyte by applying a dynamically changing pressure to the battery, but does not disclose the relationship between the contact area, pressure and buffer layer thickness. It still needs to rely on experiments to continuously try and cannot directly obtain the optimal thickness of the buffer layer under a specific assembly force. Summary of the invention

[0005] In order to solve the problem that the prior art cannot realize the coupled design of the thickness, assembly force and comprehensive internal resistance of the solid electrolyte buffer layer, the present invention proposes a design method of the solid electrolyte buffer layer, comprising:

[0006] The surface of the solid electrolyte buffer layer is regarded as a microcircle, and based on the relationship between the surface area of ​​the microcircle and the assembly force, and the relationship between the surface area of ​​the microcircle and the thickness of the solid electrolyte buffer layer, the corresponding relationship between the comprehensive internal resistance of the solid electrolyte buffer layer, the thickness of the solid electrolyte buffer layer and the assembly force is obtained;

[0007] A regression model is constructed with the corresponding relationship as a constraint function, the thickness of the solid electrolyte buffer layer and the assembly force as input variables of the regression model, the comprehensive internal resistance as output, and the goal of minimizing the comprehensive internal resistance within a value range. The regression model is solved to obtain the optimal matching design of the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force.

[0008] Optionally, the obtaining of the corresponding relationship between the comprehensive internal resistance of the solid electrolyte buffer layer, the thickness of the solid electrolyte buffer layer, and the assembly force based on the relationship between the surface area of ​​the microelement circle and the assembly force and the relationship between the surface area of ​​the microelement circle and the thickness of the solid electrolyte buffer layer includes:

[0009] Based on the geometric relationship between the surface area and assembly force of the microelement circle and the thickness of the solid electrolyte buffer layer, the coupling relationship between the effective contact area between the solid electrolyte buffer layer and the electrode, the assembly force and the thickness of the solid electrolyte buffer layer is constructed;

[0010] Obtaining the contact resistance between the solid electrolyte buffer layer and the electrode based on the coupling relationship;

[0011] The contact resistance and the material resistance of the solid electrolyte buffer layer are summed to obtain the corresponding relationship among the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force.

[0012] Optionally, the coupling relationship is:

[0013]

[0014] Among them, S 有效 is the effective contact area between the solid electrolyte buffer layer and the electrode; F is the assembly force applied between the electrode and the solid electrolyte buffer layer; E is the elastic modulus of the solid electrolyte buffer layer material; μ is the roughness between the solid electrolyte buffer layer and the electrode; h is the thickness of the solid electrolyte buffer layer; S 理想 is the ideal contact area between the solid electrolyte buffer layer and the electrode, S 理想 =dl, d is the smaller of the solid electrolyte buffer layer and the electrode width, l is the smaller of the solid electrolyte buffer layer and the electrode length.

[0015] Optionally, the contact resistance is:

[0016]

[0017] Among them, R 接触 is the contact resistance, ρ is the conductivity of the solid electrolyte buffer layer, h is the thickness of the solid electrolyte buffer layer, S 有效 is the effective contact area.

[0018] Optionally, the step of constructing the regression model includes:

[0019] The corresponding relationship is used as a constraint function of the regression model; training data is obtained, and the training data includes: the thickness of the solid electrolyte buffer layer, the assembly force, the material property parameters required for solving the comprehensive internal resistance of the solid electrolyte buffer layer, and the corresponding comprehensive internal resistance;

[0020] The thickness of the solid electrolyte buffer layer and the assembly force are taken as input variables, the material property parameters are taken as input quantities, and the comprehensive internal resistance is taken as output. The regression model is iteratively trained until the error between the calculated comprehensive internal resistance and the comprehensive internal resistance in the training data is within a preset range, thereby obtaining a regression model that characterizes the optimal matching relationship between the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force.

[0021] Optionally, the regression model includes a support vector regression model.

[0022] Optionally, the training data is obtained from an extended database, and the steps of constructing the extended database include:

[0023] Acquiring historical data, wherein the historical data includes at least one or more of the following: thickness of the solid electrolyte buffer layer, assembly force, material property parameters required for solving the comprehensive internal resistance of the solid electrolyte buffer layer, and the corresponding comprehensive internal resistance;

[0024] The historical data is used as the original data for fitting to obtain a continuously extended database.

[0025] Optionally, the step of fitting the historical data as original data to obtain a continuously extended database includes:

[0026] Using interpolation method or least square method to fit the historical data to obtain an extended database with continuous data;

[0027] or,

[0028] The regression machine model is trained with one of the solid electrolyte buffer layer thickness, assembly force or comprehensive internal resistance in the historical data as the output variable and the rest as the input variables;

[0029] The trained regression machine model is used to predict the output variable values ​​corresponding to the input variables that are not in the historical database, and the prediction results are added to the historical database to form an extended database.

[0030] Optionally, after acquiring the historical data and before fitting using the historical data as original data, the process further includes:

[0031] The historical data is sequentially subjected to data preprocessing, feature extraction and feature format conversion.

[0032] On the other hand, the present application also provides a design system for a solid electrolyte buffer layer, comprising:

[0033] A relationship acquisition module is used to regard the surface of the solid electrolyte buffer layer as a micro-circle, and based on the relationship between the surface area of ​​the micro-circle and the assembly force and the surface area of ​​the micro-circle and the thickness of the solid electrolyte buffer layer, obtain the corresponding relationship between the comprehensive internal resistance of the solid electrolyte buffer layer, the thickness of the solid electrolyte buffer layer and the assembly force;

[0034] A solution module is used to construct a regression model with the corresponding relationship as a constraint function, with the thickness and assembly force of the solid electrolyte buffer layer as input variables of the regression model, with the comprehensive internal resistance as output, and with the goal of minimizing the comprehensive internal resistance within a value range, to solve the regression model and obtain the optimal matching design of the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force.

[0035] Optionally, the relationship acquisition module obtains the corresponding relationship between the comprehensive internal resistance of the solid electrolyte buffer layer, the thickness of the solid electrolyte buffer layer, and the assembly force based on the relationship between the surface area of ​​the microelement circle and the assembly force and the relationship between the surface area of ​​the microelement circle and the thickness of the solid electrolyte buffer layer, including:

[0036] Based on the geometric relationship between the surface area and assembly force of the microelement circle and the thickness of the solid electrolyte buffer layer, the coupling relationship between the effective contact area between the solid electrolyte buffer layer and the electrode, the assembly force and the thickness of the solid electrolyte buffer layer is constructed;

[0037] Obtaining the contact resistance between the solid electrolyte buffer layer and the electrode based on the coupling relationship;

[0038] The contact resistance and the material resistance of the solid electrolyte buffer layer are summed to obtain the corresponding relationship among the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force.

[0039] Optionally, the coupling relationship in the relationship acquisition module is:

[0040]

[0041] Among them, S 有效is the effective contact area between the solid electrolyte buffer layer and the electrode; F is the assembly force applied between the electrode and the solid electrolyte buffer layer; E is the elastic modulus of the solid electrolyte buffer layer material; μ is the roughness between the solid electrolyte buffer layer and the electrode; h is the thickness of the solid electrolyte buffer layer; S 理想 is the ideal contact area between the solid electrolyte buffer layer and the electrode, S 理想 =dl, d is the smaller of the solid electrolyte buffer layer and the electrode width, l is the smaller of the solid electrolyte buffer layer and the electrode length.

[0042] Optionally, the contact resistance in the relationship acquisition module is:

[0043]

[0044] Among them, R 接触 is the contact resistance, ρ is the conductivity of the solid electrolyte buffer layer, h is the thickness of the solid electrolyte buffer layer, S 有效 is the effective contact area.

[0045] Optionally, the step of constructing the regression model in the solution module includes:

[0046] The corresponding relationship is used as a constraint function of the regression model; training data is obtained, and the training data includes: the thickness of the solid electrolyte buffer layer, the assembly force, the material property parameters required for solving the comprehensive internal resistance of the solid electrolyte buffer layer, and the corresponding comprehensive internal resistance;

[0047] The thickness of the solid electrolyte buffer layer and the assembly force are taken as input variables, the material property parameters are taken as input quantities, and the comprehensive internal resistance is taken as output. The regression model is iteratively trained until the error between the calculated comprehensive internal resistance and the comprehensive internal resistance in the training data is within a preset range, thereby obtaining a regression model that characterizes the optimal matching relationship between the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force.

[0048] Optionally, the regression model in the solution module includes a support vector regression model.

[0049] Optionally, the training data in constructing the regression model in the solution module is obtained from an extended database, and the steps of constructing the extended database include:

[0050] Acquiring historical data, wherein the historical data includes at least one or more of the following: thickness of the solid electrolyte buffer layer, assembly force, material property parameters required for solving the comprehensive internal resistance of the solid electrolyte buffer layer, and the corresponding comprehensive internal resistance;

[0051] The historical data is used as the original data for fitting to obtain a continuously extended database.

[0052] Optionally, in the construction of the extended database in the solution module, the step of fitting the historical data as original data to obtain an extended database with continuous data includes:

[0053] Using interpolation method or least square method to fit the historical data to obtain an extended database with continuous data;

[0054] or,

[0055] The regression machine model is trained with one of the solid electrolyte buffer layer thickness, assembly force or comprehensive internal resistance in the historical data as the output variable and the rest as the input variables;

[0056] The trained regression machine model is used to predict the output variable values ​​corresponding to the input variables that are not in the historical database, and the prediction results are added to the historical database to form an extended database.

[0057] Optionally, after acquiring the historical data and before fitting with the historical data as the original data in the construction of the extended database in the solution module, the following steps are further included:

[0058] The historical data is sequentially subjected to data preprocessing, feature extraction and feature format conversion.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] The present invention provides a design method and system for a solid electrolyte buffer layer, comprising: regarding the surface of the solid electrolyte buffer layer as a micro-circle, obtaining the corresponding relationship between the comprehensive internal resistance of the solid electrolyte buffer layer, the thickness of the solid electrolyte buffer layer and the assembly force based on the relationship between the surface area of ​​the micro-circle and the assembly force and the surface area of ​​the micro-circle and the thickness of the solid electrolyte buffer layer, constructing a regression model with the corresponding relationship as a constraint function, taking the thickness and the assembly force of the solid electrolyte buffer layer as input variables of the regression model, taking the comprehensive internal resistance as output, and taking minimizing the comprehensive internal resistance within a value range as a goal, solving the regression model to obtain an optimal matching design of the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force; the present invention regards the surface of the solid electrolyte buffer layer as a micro-circle, constructs the corresponding relationship between the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force, uses the corresponding relationship as a constraint function of the regression model, and then is constrained by the corresponding relationship when solving the regression model, so as to realize the coupled design of the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A flow chart of the design method of the solid electrolyte buffer layer proposed in the present invention;

[0062] Figure 2A detailed flow chart of the database expansion proposed by the present invention;

[0063] Figure 3 A detailed flow chart of the solid electrolyte buffer layer thickness prediction proposed by the present invention;

[0064] Figure 4 A schematic diagram of the steps of the design method of the solid electrolyte buffer layer proposed in the present invention;

[0065] Figure 5 The present invention proposes Figure 2 Detailed flow chart of step S1 in FIG.

[0066] Figure 6 A schematic diagram of the contact surface between the solid electrolyte buffer layer and the electrode involved in the present invention;

[0067] Figure 7 It is a schematic diagram of the contact between the microelement circle involved in the present invention and the plane when the radius approaches infinity;

[0068] Figure 8 It is a schematic diagram of the contact between the microelement circle and the plane when the radius of the microelement circle involved in the present invention approaches 0;

[0069] Fig. 9 This is a schematic diagram of the steps for constructing the regression model proposed in the present invention;

[0070] Fig.10 A schematic diagram of the steps for constructing the extended database proposed by the present invention;

[0071] Fig.11 A schematic diagram of an optimal decision domain constructed based on the design method of the solid electrolyte buffer layer of the present invention;

[0072] Fig.12 This is a schematic structural diagram of the design system of the solid electrolyte buffer layer proposed in the present invention. DETAILED DESCRIPTION

[0073] The present invention proposes a design method for a solid electrolyte buffer layer, which is based on a machine learning algorithm and the construction of contact resistance to achieve a coupled design that takes into account the thickness of the solid electrolyte buffer layer, the comprehensive internal resistance and the size of the assembly force.

[0074] like Figure 1 As shown, the design method of the solid electrolyte buffer layer mainly includes three parts:

[0075] First, the expansion of the database;

[0076] Second, the acquisition of comprehensive internal resistance;

[0077] Third, prediction of optimal thickness.

[0078] 1. Database expansion:

[0079] Based on the existing database, the features required for subsequent calculations are extracted, and the existing data features are expanded using a machine learning algorithm to obtain extended data. The extended data is added to the existing database to obtain an extended database.

[0080] like Figure 2 The following is a detailed flowchart of database expansion, including:

[0081] (1) Collect existing databases of the comprehensive internal resistance of solid electrolyte buffer layers, the thickness of solid electrolyte buffer layers, and the size of assembly forces, including data sets obtained from laboratory tests, engineering documents, or existing literature.

[0082] (2) Data cleaning and preprocessing: The collected data is cleaned and preprocessed, including removing outliers, filling missing values, standardizing data, etc. to ensure the quality and consistency of the data.

[0083] (3) Feature extraction: Filter useful features from the original data, including the chemical composition of the buffer layer material, coating method, assembly conditions, etc., as well as the correlation between the comprehensive internal resistance of the solid electrolyte buffer layer, the thickness of the solid electrolyte buffer layer and the assembly force.

[0084] (4) Feature conversion: Convert features and use label encoding to convert categorical data into a format that can be processed by machine learning algorithms.

[0085] (5) Model selection: Select a fitting model based on data characteristics and prediction requirements. In this embodiment, a regression machine learning model is preferably used to predict the continuous numerical values ​​of the solid electrolyte buffer layer thickness and assembly force.

[0086] (6) Training model: Use existing historical data to train the selected machine learning model to ensure that the model can accurately predict the solid electrolyte buffer layer thickness and assembly force.

[0087] (7) Model evaluation: Use validation sets and cross-validation techniques to evaluate the performance of regression machine models.

[0088] (8) Database expansion: The newly predicted data on the comprehensive internal resistance of the solid electrolyte buffer layer, the thickness of the solid electrolyte buffer layer, and the size of the assembly force are added to the existing database to enrich and improve the database content.

[0089] Second, the acquisition of comprehensive internal resistance:

[0090] Based on the microelement contact concept, the effective contact area between the solid electrolyte buffer layer and the electrode is obtained, and the contact resistance is calculated based on the effective contact area. The comprehensive internal resistance is the sum of the contact resistance and the material resistance of the solid electrolyte buffer layer.

[0091] 3. Prediction of optimal thickness:

[0092] A regression model is constructed with the comprehensive internal resistance constructed above as a constraint condition. The regression model is trained using data in the extended database to obtain a trained regression model. The trained regression model is used to gradually predict the thickness of the solid electrolyte buffer layer. The coupling relationship between the thickness of the solid electrolyte buffer layer and the assembly force when the minimum comprehensive internal resistance is obtained is used to construct the optimal decision domain.

[0093] like Figure 3 The figure shows a detailed schematic diagram of the optimal thickness prediction process, including:

[0094] (1) Feature engineering establishment: The thickness of the solid electrolyte buffer layer and the assembly force are selected as input variables, and the material properties, especially those related to the comprehensive internal resistance calculation, are used as input quantities.

[0095] (2) Constructing a regression model: The present invention preferably adopts a support vector regression (SVR) model, which takes minimizing the comprehensive internal resistance as the goal, uses the thickness of the solid electrolyte buffer layer and the assembly force as input variables, and uses the material properties related to the calculation of the comprehensive internal resistance as input quantities. The calculation formula of the comprehensive internal resistance is used as a constraint function to establish the SVR model.

[0096] (3) Model training: The dataset is divided into a training set and a test set. The training set is used to train the SVR model, and the test set is used to verify the performance of the SVR model.

[0097] (4) Hyperparameter tuning: During the training process, cross-validation and other techniques are used to adjust the hyperparameters of the SVR model to improve the generalization ability and prediction performance of the SVR model.

[0098] (5) Model application: Using the trained SVR model, the upper and lower limit constraints of the comprehensive internal resistance are input for prediction to obtain the optimal combination of buffer layer thickness and assembly force.

[0099] (6) Optimization and decision-making: Evaluate the trade-off between solid electrolyte buffer layer thickness and assembly force while considering the minimum comprehensive internal resistance.

[0100] (7) Verification and adjustment: Use existing data to verify whether the buffer layer thickness and assembly force predicted by the SVR model meet the requirements of minimizing the comprehensive internal resistance. If so, perform multiple simulations and optimization to update the optimal decision domain.

[0101] Embodiment 1:

[0102] This embodiment provides a design method for a solid electrolyte buffer layer. Figure 4 As shown, the process includes the following steps S1 and S2.

[0103] S1: The surface of the solid electrolyte buffer layer is regarded as a microcircle, and the corresponding relationship between the comprehensive internal resistance of the solid electrolyte buffer layer, the thickness of the solid electrolyte buffer layer and the assembly force is obtained based on the relationship between the surface area of ​​the microcircle and the assembly force, and the relationship between the surface area of ​​the microcircle and the thickness of the solid electrolyte buffer layer.

[0104] In a further preferred embodiment, Figure 5 As shown, the step S1 specifically includes the following steps S11 to S13.

[0105] S11: Based on the geometric relationship between the surface area and assembly force of the microelement circle and the thickness of the solid electrolyte buffer layer, the coupling relationship between the effective contact area between the solid electrolyte buffer layer and the electrode, the assembly force and the thickness of the solid electrolyte buffer layer is constructed.

[0106] The contact surface between the solid electrolyte buffer layer and the electrode is uneven at the microscopic level, so when the solid electrolyte and the electrode are in contact, there will be a situation where they cannot fit completely (such as Figure 6 As shown), the area of ​​contact is called the effective contact area S 有效 .

[0107] If the solid electrolyte buffer layer and the electrode are absolutely smooth, the area of ​​the contact surface is called the ideal contact area S. 理想 .

[0108] In this application, when the solid electrolyte buffer layer contacts the electrode, the contact surface is regarded as a micro-circle contact. When the radius of the micro-circle approaches infinity, such as Figure 7 As shown, the solid electrolyte buffer layer has only one point contact with the electrode, and the effective contact area approaches 0; when the radius of the microelement circle approaches 0, as shown Figure 8 As shown, the solid electrolyte buffer layer and the electrode are in contact with each other through a series of points, which is approximately complete contact. At this time, the effective contact area S 有效 Approaching the ideal contact area S 理想 .

[0109] Based on the above-mentioned microelement circle contact concept, the ideal contact area calculation formula is:

[0110] S 理想 =dl

[0111] Wherein, d is the smaller of the width of the solid electrolyte buffer layer and the electrode, and l is the smaller of the length of the solid electrolyte buffer layer and the electrode.

[0112] The expression of the effective contact area is:

[0113]

[0114] Among them, S 有效 is the effective contact area between the solid electrolyte buffer layer and the electrode; F is the assembly force applied between the electrode and the solid electrolyte buffer layer; E is the elastic modulus of the solid electrolyte buffer layer material; μ is the roughness between the solid electrolyte buffer layer and the electrode; h is the thickness of the solid electrolyte buffer layer; S 理想 is the ideal contact area between the solid electrolyte buffer layer and the electrode.

[0115] It can be seen from the formula of the effective contact area that it characterizes that the effective contact area is directly related to the assembly force applied between the electrode and the solid electrolyte buffer layer, and the thickness of the solid electrolyte buffer layer. Therefore, the effective contact area characterizes the coupling relationship between the effective contact area, the assembly force and the thickness of the solid electrolyte buffer layer.

[0116] S12: Obtaining the contact resistance between the solid electrolyte buffer layer and the electrode based on the coupling relationship.

[0117] In the prior art, the contact resistance between the solid electrolyte buffer layer and the electrode is directly obtained through a large number of experimental measurements, and the internal relationship between the contact resistance and the thickness of the solid electrolyte buffer layer and the assembly force cannot be intuitively obtained, and thus the optimal coupling design of the three cannot be further studied. In this application, due to the determination of the effective contact area, the calculation of the contact resistance becomes possible, and the calculation formula is as follows:

[0118]

[0119] Among them, R 接触 is the contact resistance, ρ is the conductivity of the solid electrolyte buffer layer, h is the thickness of the solid electrolyte buffer layer, S 有效 is the effective contact area between the above-mentioned solid electrolyte buffer layer and the electrode.

[0120] S13: summing the contact resistance and the material resistance of the solid electrolyte buffer layer to obtain a corresponding relationship among the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force.

[0121] The comprehensive internal resistance R of the solid electrolyte buffer layer 缓冲层 It includes two parts. One is the material resistance R of the solid electrolyte buffer layer material itself. 材料 , and the second is the contact resistance R formed by the contact surface between the solid electrolyte buffer layer and the electrode material 接触 , so the comprehensive internal resistance R 缓冲层 It can be expressed as:

[0122]

[0123] It can be seen that the comprehensive internal resistance characterizes the corresponding relationship among the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force.

[0124] Table 1 is a comparison between the comprehensive internal resistance calculated by the method of the present application and the comprehensive internal resistance of the actual test. It can be seen that the error between the comprehensive internal resistance calculated by the method of the present application and the comprehensive internal resistance of the actual test is within 3%, which proves that the method in the present application can accurately calculate the comprehensive internal resistance of the solid electrolyte buffer layer.

[0125] Table 1: Comparison of the comprehensive internal resistance calculated by this application method and the comprehensive internal resistance actually tested

[0126]

[0127] S2: A regression model is constructed with the corresponding relationship as a constraint function, the thickness of the solid electrolyte buffer layer and the assembly force as input variables of the regression model, the comprehensive internal resistance as output, and the goal is to minimize the comprehensive internal resistance within a value range. The regression model is solved to obtain the optimal matching design of the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force.

[0128] Since the corresponding relationship describes the coupling relationship between the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force, the regression model with the corresponding relationship as the constraint function can comprehensively consider the mutual constraint relationship between the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force. Taking the thickness of the solid electrolyte buffer layer and the assembly force as the input variables of the regression model, and taking the comprehensive internal resistance minimized within the value range as the goal, the regression model is solved to obtain the corresponding solid electrolyte buffer layer thickness and assembly force when the comprehensive internal resistance is minimized. Then, a certain solid electrolyte buffer layer thickness corresponds to a unique assembly force and comprehensive internal resistance, and different thicknesses correspond to different assembly forces and comprehensive internal resistances. The optimal matching relationship between the three can form the optimal decision domain.

[0129] In a further preferred embodiment, the regression model includes one or more of the following models: linear regression, ridge regression, decision tree regression, random forest regression, gradient boosting regression, support vector regression, XGBoost regression, LightGBM regression; the present application preferably uses a support vector regression (SVR) model, such as Fig. 9 As shown, the steps of constructing the regression model include steps 1 to 3:

[0130] Step 1: Using the corresponding relationship as a constraint function of the regression model;

[0131] Step 2: Obtain training data, wherein the training data includes: thickness of the solid electrolyte buffer layer, assembly force, material property parameters required for solving the comprehensive internal resistance of the solid electrolyte buffer layer, and the corresponding comprehensive internal resistance.

[0132] In a further preferred embodiment, the training data is obtained from an extended database, such as Fig.10 As shown, the steps of constructing the extended database include step 21 and step 22:

[0133] Step 21: Acquire historical data, wherein the historical data includes at least one or more of the following: thickness of the solid electrolyte buffer layer, assembly force, material property parameters required for solving the comprehensive internal resistance of the solid electrolyte buffer layer, and the corresponding comprehensive internal resistance.

[0134] Step 22: Use the historical data as original data to perform fitting to obtain a continuous extended database.

[0135] In a further preferred solution, in step 21, a database of existing solid electrolyte buffer layer comprehensive internal resistance, solid electrolyte buffer layer thickness and assembly force magnitude may be obtained from laboratory tests, engineering documents or existing literature.

[0136] The collected data are preprocessed, including cleaning, removing outliers, filling missing values, standardizing data, and other steps to ensure the quality and consistency of the data.

[0137] Features that may be useful in the future are screened out from the preprocessed data, such as the chemical composition of the buffer layer material, coating method, assembly conditions, etc., as well as the correlation features between the buffer layer internal resistance, buffer layer thickness and assembly force.

[0138] Convert features into a format that can be used later. For example, label encoding can be used to convert categorical data into a format that can be processed by machine learning algorithms.

[0139] In a further preferred solution, in step 22, fitting the historical data as original data to obtain a continuously extended database includes:

[0140] The historical data are fitted by using an interpolation method or a least square method to obtain a continuous extended database, and continuous data are obtained based on discrete and limited historical data to achieve database expansion.

[0141] or,

[0142] The regression machine model is trained with one of the solid electrolyte buffer layer thickness, assembly force or comprehensive internal resistance in the historical data as the output variable and the rest as the input variables;

[0143] The trained regression machine model is used to predict the output variable values ​​corresponding to the input variables that are not in the historical database, and the prediction results are added to the historical database to form an extended database.

[0144] Step 3: Taking the thickness of the solid electrolyte buffer layer and the assembly force as input variables, the material property parameters as input quantities, and the comprehensive internal resistance as output, the regression model is iteratively trained until the error between the calculated comprehensive internal resistance and the comprehensive internal resistance in the training data is within a preset range, thereby obtaining a regression model that characterizes the optimal matching relationship between the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer, and the assembly force.

[0145] After the training is completed, the existing data can be used to verify and optimize the regression model to determine the optimal parameters.

[0146] In the application process, the feasible range of the comprehensive internal resistance of the buffer layer is taken as a constraint condition, and the trained regression model is used to predict the thickness of the solid electrolyte buffer layer. Then, the solid electrolyte buffer layer thickness and assembly force corresponding to the minimum comprehensive internal resistance can be obtained. Different assembly forces and buffer layer thicknesses correspond to different comprehensive internal resistances. Therefore, the coupled optimal design of the three can form the optimal decision domain (such as Fig.11 As shown in Figure 1 ). The combination of assembly force and the thickness of the poly solid electrolyte buffer layer that falls within the optimal decision domain can achieve a lower comprehensive internal resistance. Therefore, this figure can be used to guide the design of the thickness and applied pressure of the solid electrolyte buffer layer, and help designers of solid-state lithium batteries determine the thickness and applied pressure of the solid electrolyte buffer layer to obtain a lower comprehensive internal resistance of the battery, and obtain a solid-state battery with high safety, higher energy density, and longer cycle life.

[0147] Embodiment 2:

[0148] The present invention based on the same inventive concept also provides a design system for a solid electrolyte buffer layer, such as Fig.12 As shown, including:

[0149] A relationship acquisition module is used to regard the surface of the solid electrolyte buffer layer as a micro-circle, and based on the relationship between the surface area of ​​the micro-circle and the assembly force and the surface area of ​​the micro-circle and the thickness of the solid electrolyte buffer layer, obtain the corresponding relationship between the comprehensive internal resistance of the solid electrolyte buffer layer, the thickness of the solid electrolyte buffer layer and the assembly force;

[0150] A solution module is used to construct a regression model with the corresponding relationship as a constraint function, with the thickness and assembly force of the solid electrolyte buffer layer as input variables of the regression model, with the comprehensive internal resistance as output, and with the goal of minimizing the comprehensive internal resistance within a value range, to solve the regression model and obtain the optimal matching design of the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force.

[0151] In a further preferred embodiment, the relationship acquisition module obtains the corresponding relationship between the comprehensive internal resistance of the solid electrolyte buffer layer, the thickness of the solid electrolyte buffer layer and the assembly force based on the relationship between the surface area of ​​the microelement circle and the assembly force and the relationship between the surface area of ​​the microelement circle and the thickness of the solid electrolyte buffer layer, including:

[0152] Based on the geometric relationship between the surface area and assembly force of the microelement circle and the thickness of the solid electrolyte buffer layer, the coupling relationship between the effective contact area between the solid electrolyte buffer layer and the electrode, the assembly force and the thickness of the solid electrolyte buffer layer is constructed;

[0153] Obtaining the contact resistance between the solid electrolyte buffer layer and the electrode based on the coupling relationship;

[0154] The contact resistance and the material resistance of the solid electrolyte buffer layer are summed to obtain the corresponding relationship among the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force.

[0155] In a further preferred solution, the coupling relationship in the relationship acquisition module is:

[0156]

[0157] Among them, S 有效 is the effective contact area between the solid electrolyte buffer layer and the electrode; F is the assembly force applied between the electrode and the solid electrolyte buffer layer; E is the elastic modulus of the solid electrolyte buffer layer material; μ is the roughness between the solid electrolyte buffer layer and the electrode; h is the thickness of the solid electrolyte buffer layer; S 理想 is the ideal contact area between the solid electrolyte buffer layer and the electrode, S 理想 =dl, d is the smaller of the solid electrolyte buffer layer and the electrode width, l is the smaller of the solid electrolyte buffer layer and the electrode length.

[0158] In a further preferred solution, the contact resistance in the relationship acquisition module is:

[0159]

[0160] Among them, R 接触 is the contact resistance, ρ is the conductivity of the solid electrolyte buffer layer, h is the thickness of the solid electrolyte buffer layer, S 有效 is the effective contact area.

[0161] In a further preferred embodiment, the step of constructing the regression model in the solution module includes:

[0162] The corresponding relationship is used as a constraint function of the regression model; training data is obtained, and the training data includes: the thickness of the solid electrolyte buffer layer, the assembly force, the material property parameters required for solving the comprehensive internal resistance of the solid electrolyte buffer layer, and the corresponding comprehensive internal resistance;

[0163] The thickness of the solid electrolyte buffer layer and the assembly force are taken as input variables, the material property parameters are taken as input quantities, and the comprehensive internal resistance is taken as output. The regression model is iteratively trained until the error between the calculated comprehensive internal resistance and the comprehensive internal resistance in the training data is within a preset range, thereby obtaining a regression model that characterizes the optimal matching relationship between the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force.

[0164] In a further preferred solution, the regression model in the solution module includes a support vector regression model.

[0165] In a further preferred embodiment, the training data in the construction of the regression model in the solution module is obtained from an extended database, and the steps of constructing the extended database include:

[0166] Acquiring historical data, wherein the historical data includes at least one or more of the following: thickness of the solid electrolyte buffer layer, assembly force, material property parameters required for solving the comprehensive internal resistance of the solid electrolyte buffer layer, and the corresponding comprehensive internal resistance;

[0167] The historical data is used as the original data for fitting to obtain a continuously extended database.

[0168] In a further preferred solution, the step of obtaining a continuous extended database by fitting the historical data as original data in the construction of the extended database in the solution module includes:

[0169] Using interpolation method or least square method to fit the historical data to obtain an extended database with continuous data;

[0170] or,

[0171] The regression machine model is trained with one of the solid electrolyte buffer layer thickness, assembly force or comprehensive internal resistance in the historical data as the output variable and the rest as the input variables;

[0172] The trained regression machine model is used to predict the output variable values ​​corresponding to the input variables that are not in the historical database, and the prediction results are added to the historical database to form an extended database.

[0173] In a further preferred solution, after acquiring the historical data and before fitting with the historical data as the original data in the construction of the extended database in the solution module, the following steps are further included:

[0174] The historical data is sequentially subjected to data preprocessing, feature extraction and feature format conversion.

[0175] The above are merely embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are included in the scope of the claims of the present invention to be approved.

Claims

1. A method for designing a solid electrolyte buffer layer, characterized in that: include: The surface of the solid electrolyte buffer layer is regarded as a microcircle, and based on the relationship between the surface area of ​​the microcircle and the assembly force, and the relationship between the surface area of ​​the microcircle and the thickness of the solid electrolyte buffer layer, the corresponding relationship between the comprehensive internal resistance of the solid electrolyte buffer layer, the thickness of the solid electrolyte buffer layer and the assembly force is obtained; A regression model is constructed with the corresponding relationship as a constraint function, the thickness of the solid electrolyte buffer layer and the assembly force as input variables of the regression model, the comprehensive internal resistance as output, and the goal of minimizing the comprehensive internal resistance within a value range. The regression model is solved to obtain the optimal matching design of the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force.

2. The method for designing a solid electrolyte buffer layer according to claim 1, characterized in that: The method of obtaining the corresponding relationship between the comprehensive internal resistance of the solid electrolyte buffer layer, the thickness of the solid electrolyte buffer layer and the assembly force based on the relationship between the surface area of ​​the microelement circle and the assembly force and the relationship between the surface area of ​​the microelement circle and the thickness of the solid electrolyte buffer layer includes: Based on the geometric relationship between the surface area and assembly force of the microelement circle and the thickness of the solid electrolyte buffer layer, the coupling relationship between the effective contact area between the solid electrolyte buffer layer and the electrode, the assembly force and the thickness of the solid electrolyte buffer layer is constructed; Obtaining the contact resistance between the solid electrolyte buffer layer and the electrode based on the coupling relationship; The contact resistance and the material resistance of the solid electrolyte buffer layer are summed to obtain the corresponding relationship among the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force.

3. The method for designing a solid electrolyte buffer layer according to claim 2, characterized in that: The coupling relationship is: Among them, S 有效 is the effective contact area between the solid electrolyte buffer layer and the electrode; F is the assembly force applied between the electrode and the solid electrolyte buffer layer; E is the elastic modulus of the solid electrolyte buffer layer material; μ is the roughness between the solid electrolyte buffer layer and the electrode; h is the thickness of the solid electrolyte buffer layer; S 理想 is the ideal contact area between the solid electrolyte buffer layer and the electrode, S 理想 =dl, d is the smaller of the solid electrolyte buffer layer and the electrode width, l is the smaller of the solid electrolyte buffer layer and the electrode length.

4. The method for designing a solid electrolyte buffer layer according to claim 2 or 3, characterized in that: The contact resistance is: Among them, R 接触 is the contact resistance, ρ is the conductivity of the solid electrolyte buffer layer, h is the thickness of the solid electrolyte buffer layer, S 有效 is the effective contact area.

5. The method for designing a solid electrolyte buffer layer according to claim 1, characterized in that: The steps of constructing the regression model include: The corresponding relationship is used as a constraint function of the regression model; training data is obtained, and the training data includes: the thickness of the solid electrolyte buffer layer, the assembly force, the material property parameters required for solving the comprehensive internal resistance of the solid electrolyte buffer layer, and the corresponding comprehensive internal resistance; The thickness of the solid electrolyte buffer layer and the assembly force are taken as input variables, the material property parameters are taken as input quantities, and the comprehensive internal resistance is taken as output. The regression model is iteratively trained until the error between the calculated comprehensive internal resistance and the comprehensive internal resistance in the training data is within a preset range, thereby obtaining a regression model that characterizes the optimal matching relationship between the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force.

6. The method for designing a solid electrolyte buffer layer according to claim 1 or 5, characterized in that: The regression model includes a support vector regression model.

7. The method for designing a solid electrolyte buffer layer according to claim 6, characterized in that: The training data is obtained from an extended database, and the steps of constructing the extended database include: Acquiring historical data, wherein the historical data includes at least one or more of the following: thickness of the solid electrolyte buffer layer, assembly force, material property parameters required for solving the comprehensive internal resistance of the solid electrolyte buffer layer, and the corresponding comprehensive internal resistance; The historical data is used as the original data for fitting to obtain a continuously extended database.

8. The method for designing a solid electrolyte buffer layer according to claim 7, characterized in that: The method of fitting the historical data as original data to obtain a continuous extended database includes: Using interpolation method or least square method to fit the historical data to obtain an extended database with continuous data; or, The regression machine model is trained with one of the solid electrolyte buffer layer thickness, assembly force or comprehensive internal resistance in the historical data as the output variable and the rest as the input variables; The trained regression machine model is used to predict the output variable values ​​corresponding to the input variables that are not in the historical database, and the prediction results are added to the historical database to form an extended database.

9. The method for designing a solid electrolyte buffer layer according to claim 7, characterized in that: After acquiring the historical data and before fitting with the historical data as the original data, the following steps are also included: The historical data is sequentially subjected to data preprocessing, feature extraction and feature format conversion.

10. A design system for a solid electrolyte buffer layer, characterized in that: include: A relationship acquisition module is used to regard the surface of the solid electrolyte buffer layer as a micro-circle, and based on the relationship between the surface area of ​​the micro-circle and the assembly force and the surface area of ​​the micro-circle and the thickness of the solid electrolyte buffer layer, obtain the corresponding relationship between the comprehensive internal resistance of the solid electrolyte buffer layer, the thickness of the solid electrolyte buffer layer and the assembly force; A solution module is used to construct a regression model with the corresponding relationship as a constraint function, with the thickness and assembly force of the solid electrolyte buffer layer as input variables of the regression model, with the comprehensive internal resistance as output, and with the goal of minimizing the comprehensive internal resistance within a value range, to solve the regression model and obtain the optimal matching design of the comprehensive internal resistance, the thickness of the solid electrolyte buffer layer and the assembly force.

Images