Synchronous multi-objective optimization method for electrode and structure parameters of high-capacity lithium ion battery

Abstract

The invention provides a synchronous multi-objective optimization method and system for electrode and structure parameters of a high-capacity lithium ion battery, and aims to solve the technical defects that step-by-step optimization efficiency of electrode and structure design parameters is low, a coupling effect is not considered, and electrochemical performance and thermal safety performance cannot be balanced synchronously in the prior art. According to the method, an electrode-current collector-battery cell multi-scale structure-activity model coupling an electrode electrochemical model, a current collector electric model and a battery cell thermal model is constructed, so that correlation between electrode / structure design parameters and specific energy-maximum temperature-maximum temperature difference performance indexes is realized; in combination with a multi-objective optimization framework and an agent model method, a to-be-evaluated parameter sample is screened through multiple strategies and iterative optimization is carried out, and finally a non-dominated solution set of equilibrium specific energy, maximum temperature and maximum temperature difference is output. According to the method, the optimal design scheme can be quickly obtained under limited computing resources, and the requirements of electric vehicles, energy storage power stations and other scenes on high energy density and high safety of the high-capacity lithium ion battery are met.

Description

Technical Field

[0001] This invention relates to the field of lithium-ion battery design optimization technology, specifically to a method and system for simultaneous multi-objective optimization of electrode and structural parameters of a high-capacity lithium-ion battery. Background Technology

[0002] To meet the high energy density demands of applications such as electric vehicles and energy storage power stations, the development of lithium-ion batteries has shown a trend of continuously increasing cell capacity. This increase in capacity presents new challenges to the design of lithium-ion battery cells: on the one hand, it requires optimizing electrode design parameters to improve ion and electron transport within the battery, thereby further increasing specific energy; on the other hand, it also requires optimizing structural design parameters to improve internal heat transfer and avoid thermal safety issues caused by longer heat conduction paths and greater heat generation during charging and discharging in larger capacity cells. Traditional lithium-ion battery electrode and structural design optimization are done in two steps: first optimizing the electrode design, then optimizing the structural design. This method is inefficient and does not consider the coupling effect of electrode and structural design on lithium-ion battery performance, resulting in suboptimal design outcomes.

[0003] Patent document CN109446619A discloses a method for optimizing electrode design parameters of a lithium-ion battery. The method involves selecting the battery to be optimized, measuring the practically measurable electrode design parameters, and establishing the electrochemical parameters of the lithium-ion battery based on the measured and estimated parameters. The thermally coupled model was used, and the estimated parameters were adjusted through experiments. The optimized electrode design parameters were obtained using two optimization methods, with the optimization objectives being maximizing energy density and maximizing the product of energy density and power density. However, this patent document is limited to optimizing electrode design parameters, and the optimization objectives are limited to energy density and power density. It cannot simultaneously optimize structural design parameters, nor can it consider the thermal performance of the battery. Furthermore, the local optimization algorithm it employs is highly dependent on the initial values ​​of the electrode design parameters, making it difficult to guarantee the global optimality of the final solution.

[0004] Patent document CN108595840A discloses a modeling method, system, and electrode tab size optimization method for lithium-ion batteries. The modeling method establishes a one-dimensional electrochemical model and a three-dimensional thermal model of the lithium-ion battery based on its appearance parameters, electrode design parameters, electrode kinetic parameters, and electrode material thermophysical properties, and then couples the two models to obtain a coupled model. After obtaining the coupled model, constant current discharge experiments of lithium-ion batteries with different appearance parameters can be simulated by changing the appearance parameters of the coupled model. Thermal analysis is then performed on the lithium-ion batteries with these appearance parameters based on the experimental results. However, this patent document only addresses structural design parameters, and the optimization objective only involves thermal performance, failing to consider the influence of electrode design parameters on electrochemical performance. Furthermore, since no optimization algorithm is incorporated, the resulting design may not be optimal.

[0005] Therefore, the market needs a method and system for simultaneous multi-objective optimization of electrode and structural parameters of large-capacity lithium-ion batteries that can achieve a synergistic balance between electrochemical performance and thermal safety performance, and improve the design efficiency and global optimality of large-capacity lithium-ion batteries. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the purpose of this invention is to provide a method and system for simultaneous multi-objective optimization of electrode and structural parameters in high-capacity lithium-ion batteries.

[0007] A method for simultaneous multi-objective optimization of electrode and structural parameters of a high-capacity lithium-ion battery according to the present invention includes: Step S1: Determine the high-capacity lithium-ion cell to be optimized, and obtain the electrode design parameters, structural design parameters, electrochemical parameters, and thermal parameters, thereby constructing a multi-scale structure-property model of the electrode-current collector-cell.

[0008] Step S2: Determine the adjustable range of electrode design parameters and structural design parameters, randomly select design parameter samples within the adjustable range, and input the design parameter samples into the electrode-current collector-cell multi-scale structure-property model to obtain the simulation results of specific energy-maximum temperature-maximum temperature difference performance indicators, and then filter the current non-dominated solution set from the design parameter samples; Step S3: Construct a structure-property surrogate model using the design parameter samples and simulation results of specific energy-maximum temperature-maximum temperature difference performance indicators, and construct a Pareto front surrogate model using the current non-dominated solution set; Step S4: Based on the structure-property surrogacy model and the Pareto frontier surrogacy model, screen the design parameter samples to be evaluated; Step S5: Calculate the specific energy, maximum temperature, and maximum temperature difference performance index values ​​corresponding to the design parameter sample to be evaluated based on the electrode-current collector-cell multi-scale structure-property model; Step S6: Update the design parameter samples, the simulation results of the specific energy-maximum temperature-maximum temperature difference performance index and the current non-dominated solution set, and determine whether the maximum number of simulations of the electrode-current collector-cell multi-scale structure-property model has been reached. If yes, output the final non-dominated solution set and execute step S7; otherwise, return to step S3. Step S7: Plot the Pareto front of the final non-dominated solution set, select the region within the Pareto front that meets the design requirements for specific energy, maximum temperature, and maximum temperature difference, and extract the optimized electrode design parameters and structural design parameters for that region. Preferably, the electrode design parameters include positive electrode thickness, positive electrode porosity, negative electrode porosity, and NP ratio; The structural design parameters include current collector thickness, current collector aspect ratio, and number of electrode pairs; The electrochemical parameters include solid-phase diffusion coefficient, solid-phase diffusion activation energy, reaction rate constant, reaction activation energy, initial solid-phase lithium concentration, and Bruggeman index. The thermal parameters include material density, specific heat capacity at constant pressure, thermal conductivity, and equivalent resistance. The electrode-current collector-cell multi-scale structure-property model is coupled with multiple sub-models describing battery characteristics at different scales, including the electrode electrochemical model, the current collector hydroelectric model, and the cell thermal model. The methods for randomly sampling the design parameters include Monte Carlo sampling, orthogonal sampling, and Latin hypercube sampling. The specific energy-maximum temperature-maximum temperature difference performance indicators include: Specific energy: ; Maximum temperature: ; Maximum temperature difference: ; in, For time, , , For three-dimensional spatial coordinate components, For voltage, This represents the average temperature of the battery cell. For temperature, This refers to the total mass of the battery cells. Rated current, This represents the total discharge time.

[0009] Preferably, the electrode electrochemical model includes: Solid-state charge conservation: ; Liquid phase charge conservation: ; Solid-phase mass conservation: ; Liquid phase mass conservation: ; Electrode dynamics: ; Electrochemical heat generation rate: ; in, The distance to the negative electrode current collector-negative electrode interface. The distance to the center of the particle. For solid-state potential, The liquid phase potential, Electrolyte concentration, This represents the concentration of lithium in the solid phase. This represents the average temperature of the battery cell. The electrochemical heat generation rate; This represents the volume fraction of the solid phase. For solid-state conductivity, The volume current density, It is the liquid volume fraction. Liquid phase conductivity, The transference number of ions. For activity, Let be the solid-phase diffusion coefficient. Where is the liquid phase diffusion coefficient, Where is the particle radius, The reaction rate constant is... For the maximum solid-phase lithium concentration, The surface lithium concentration, The anode exchange coefficient, The cathode exchange coefficient, This is an overpotential. This is the electrode equilibrium potential. For solid-phase Bruggeman index, The Bruggeman index for liquid phase. Let be the ideal gas constant. For Faraday constant, subscript Indicate the scope of the equation; The current collector hydroelectric model includes: Solid-state charge conservation: ; Heat generation rate of the current collector: ; in, The solid-phase potential of the current collector. Rated current, For the number of pole pieces, The cross-sectional area of ​​the collector is... For the current collector thickness, Heat generation rate of the current collector; subscript Indicate the scope of the equation; The cell thermal model includes: Electrode heat conservation equation: ; Cell heat conservation equation: ; in, The density of the pole ears, The specific heat capacity at constant pressure of the electrode tab. For the thermal conductivity of the electrode, The equivalent resistance is calculated based on the tab size. The volume of the pole ears, For cell density, For the specific heat capacity of the battery cell at constant voltage, Thermal conductivity in the planar direction of the battery cell Thermal conductivity in the thickness direction of the battery cell. The rate at which the battery cell generates heat.

[0010] Preferably, in the electrode electrochemical model, the domain of solid phase charge conservation and electrode kinetic equation is the negative electrode and positive electrode; the domain of liquid phase charge conservation and liquid phase mass conservation equation is the negative electrode, membrane, and positive electrode; the domain of solid phase mass conservation equation is the negative electrode particles and positive electrode particles; and the relationship between solid phase diffusion coefficient, reaction rate constant and temperature is described by the Arrhenius formula. In the current collector model, the domain of application of the equation is the negative electrode current collector and the positive electrode current collector; In the cell thermal model, the domain of the tab heat conservation equation is the negative and positive tabs, while the domain of the cell heat conservation equation is the cell region. The coupling method of the electrode electrochemical model, the current collector electroelectric model, and the cell thermal model includes: the electrode electrochemical model receives the average cell temperature calculated from the cell thermal model and returns the electrochemical heat generation rate; the cell thermal model receives the electrochemical heat generation rate from the electrode electrochemical model and the current collector heat generation rate from the current collector electroelectric model, and returns the average cell temperature to the electrode electrochemical model.

[0011] Preferably, the cell region is a multiphase mixture of electrodes, a separator, a current collector, and an electrolyte, and its density, specific heat capacity at constant pressure, thermal conductivity, and heat generation rate are calculated using the following superposition formula: Cell density: ; Cell specific heat capacity: ; Thermal conductivity of the cell in the planar direction: ; Thermal conductivity in the thickness direction of the battery cell: ; Cell heat generation rate: ; in, For the first j Density of each region For the first j The specific heat capacity under constant pressure in each region For the first j Thermal conductivity of each region For the first j The thickness of each region, For the first j Solid density of each region For the first j Solid-state isobaric specific heat capacity of each region For the first j Solid-phase thermal conductivity of each region The electrolyte density, The specific heat capacity of the electrolyte. The value represents the thermal conductivity of the electrolyte. The subscripts nc, ne, sp, pe, and pc represent the negative electrode current collector, negative electrode, diaphragm, positive electrode, and positive electrode current collector, respectively.

[0012] Preferably, the structure-property surrogate model includes a specific energy inverse number structure-property surrogate model, a maximum temperature structure-property surrogate model, and a maximum temperature difference structure-property surrogate model. The inputs are electrode design parameters and structural design parameters, and the outputs are specific energy inverse number, maximum temperature, and maximum temperature difference, respectively. The model structure adopts a cubic radial basis function interpolation model. The Pareto front surrogate model takes the maximum temperature and maximum temperature difference as inputs and outputs the inverse of the specific energy. The model structure uses a linear radial basis function interpolation model.

[0013] Preferably, step S4 includes screening design parameter samples to be evaluated based on a maximum-minimum distance strategy, screening design parameter samples to be evaluated based on an approximate multi-objective optimization strategy, screening design parameter samples to be evaluated based on an approximate single-objective optimization strategy, screening design parameter samples to be evaluated based on a local perturbation strategy, and screening design parameter samples to be evaluated based on a random sampling strategy.

[0014] Preferably, the maximum-minimum distance strategy includes: first, constructing the following sub-optimization problem based on the Pareto front agent model,

[0015] in, For the Pareto frontier agent model, This represents the minimum value of the maximum temperature in the current non-dominated solution set. This represents the maximum value of the maximum temperature in the current non-dominated solution set. This represents the minimum of the maximum temperature difference in the current non-dominated solution set. The maximum temperature difference in the current non-dominated solution set is the maximum value, and a single-objective genetic algorithm is used to solve it, thereby obtaining the maximum-minimum distance solution. Subsequently, based on the structure-function proxy model, the following sub-optimization problems were constructed.

[0016] in, , , These are, respectively, the structure-property surrogate model based on specific energy opposite number, the structure-property surrogate model based on maximum temperature, and the structure-property surrogate model based on maximum temperature difference. To design variable vectors, The thickness of the positive electrode. Porosity is the positive electrode porosity. The negative porosity, For NP ratio, For the number of pole pieces, For the current collector thickness, The aspect ratio of the current collector is defined, and the subscripts min and max represent the lower and upper bounds of the adjustable range, respectively. A multi-objective genetic algorithm is used to solve the problem, thereby obtaining a set of non-dominated solutions. Finally, solutions that satisfy the given conditions are selected from the set of non-dominated solutions. At most one solution that does not appear in the design parameter sample is added to the design parameter sample to be evaluated. The approximate multi-objective optimization strategy includes constructing the following sub-optimization problems based on the structure-efficacy surrogate model.

[0017] A multi-objective genetic algorithm is used to solve the problem, thereby obtaining the non-dominated solution set. At most 7 non-dominated solutions that do not appear in the design parameter sample are randomly selected from the non-dominated solution set and added to the design parameter sample to be evaluated. The approximate single-objective optimization strategy includes constructing the following three sub-optimization problems based on the structure-function surrogate model.

[0018] Each solution is solved using a single-objective genetic algorithm to obtain three optimal solutions. At most three solutions that do not appear in the design parameter sample are selected and added to the design parameter sample to be evaluated. The local perturbation strategy includes: First, randomly selecting no more than 1000 candidate design parameter samples within an adjustable range; second, based on the structure-performance surrogate model, evaluating the approximate values ​​of the specific energy inverse, maximum temperature, and maximum temperature difference for each candidate design parameter sample, and calculating the minimum Euclidean distance from each candidate design parameter sample to the design parameter sample; further, using the min-max method to normalize the specific energy inverse, maximum temperature, maximum temperature difference, and minimum Euclidean distance for each candidate design parameter sample to obtain normalized specific energy inverse, normalized maximum temperature, normalized maximum temperature difference, and normalized minimum Euclidean distance; finally, using the weighted sum of the average of the normalized specific energy inverse, normalized maximum temperature, and normalized maximum temperature difference and the normalized minimum Euclidean distance as the comprehensive score for each candidate design parameter sample, and selecting the candidate design parameter sample with the lowest comprehensive score to add to the design parameter sample to be evaluated.

[0019] The random sampling strategy includes: First, randomly selecting no more than 1000 candidate design parameter samples within an adjustable range; second, based on the structure-performance surrogate model, evaluating the approximate values ​​of the specific energy inverse, maximum temperature, and maximum temperature difference for each candidate design parameter sample, and calculating the minimum Euclidean distance from each candidate design parameter sample to the design parameter sample; further, using the min-max method to normalize the specific energy inverse, maximum temperature, maximum temperature difference, and minimum Euclidean distance for each candidate design parameter sample to obtain normalized specific energy inverse, normalized maximum temperature, normalized maximum temperature difference, and normalized minimum Euclidean distance; finally, using the weighted sum of the average of the normalized specific energy inverse, normalized maximum temperature, and normalized maximum temperature difference and the normalized minimum Euclidean distance as the comprehensive score for each candidate design parameter sample, and selecting the candidate design parameter sample with the lowest comprehensive score to add to the design parameter sample to be evaluated.

[0020] A high-capacity lithium-ion battery electrode and structural parameter synchronous multi-objective optimization system provided by the present invention includes: Module M1: Identify the high-capacity lithium-ion cell to be optimized, and obtain electrode design parameters, structural design parameters, electrochemical parameters, and thermal parameters, thereby constructing a multi-scale structure-property model of electrode-current collector-cell.

[0021] Module M2: Determines the adjustable range of electrode design parameters and structural design parameters, randomly selects design parameter samples within the adjustable range, and inputs the design parameter samples into the electrode-current collector-cell multi-scale structure-property model to obtain the simulation results of specific energy-maximum temperature-maximum temperature difference performance indicators, and then filters the current non-dominated solution set from the design parameter samples; Module M3: Constructs a structure-property surrogate model using design parameter samples and simulation results of specific energy-maximum temperature-maximum temperature difference performance indicators, and constructs a Pareto front surrogate model using the current non-dominated solution set; Module M4: Based on the structure-property surrogate model and the Pareto frontier surrogate model, screen the design parameter samples to be evaluated; Module M5: Calculates the specific energy, maximum temperature, and maximum temperature difference performance index values ​​corresponding to the design parameter samples to be evaluated based on the electrode-current collector-cell multi-scale structure-property model; Module M6: Updates the design parameter samples, the simulation results of the specific energy-maximum temperature-maximum temperature difference performance index and the current non-dominated solution set, and determines whether the maximum number of simulations of the electrode-current collector-cell multi-scale structure-property model has been reached. If yes, it outputs the final non-dominated solution set and then triggers module M7; otherwise, it triggers module M3. Module M7: Draw the Pareto front of the final non-dominated solution set, select the region in the Pareto front that meets the design requirements for specific energy, maximum temperature, and maximum temperature difference, and extract the optimization scheme of electrode design parameters and structural design parameters for this region.

[0022] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention solves the problems of low-dimensional electrochemical-thermal coupling models being unable to simultaneously optimize structural design parameters and high-dimensional electrochemical-thermal coupling models being difficult to solve by adopting a multi-scale modeling strategy that combines coupled electrode electrochemical models, current collector electroelectric models, and cell thermal models. It achieves the effect of simulating battery specific energy-maximum temperature-maximum temperature difference performance indicators under different electrode design parameters and structural design parameters, while reducing the computational resource requirements.

[0023] 2. This invention solves the problem of numerous design objectives and difficulty in balancing them in the design of large-capacity lithium-ion batteries by adopting a design parameter optimization strategy based on a multi-objective optimization framework. It achieves the effect of simultaneously obtaining multiple optimal solutions with different specific energies, maximum temperatures, and maximum temperature difference levels.

[0024] 3. This invention solves the problem of high computational resource requirements in traditional multi-objective optimization, which requires the use of complex models to evaluate a large number of design parameter samples, by adopting a method based on a surrogate model for screening design parameters to be evaluated. It achieves the effect of quickly obtaining multi-objective optimization solutions under limited computational resources. Attached Figure Description

[0025] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a schematic diagram of the working method of the present invention; Figure 2 This describes the composition and coupling mechanism of the multi-scale structure-property model of electrode-current collector-cell in this invention; Figure 3 This is a schematic diagram of the process for screening design parameters to be evaluated based on the structure-property surrogacy model and the Pareto frontier surrogacy model in this invention; Figure 4 This is a schematic diagram of the final non-dominated solution set in this invention. Detailed Implementation

[0026] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0027] This invention constructs a multi-scale structure-property model of electrode-current collector-cell by coupling an electrode electrochemical model, a current collector current-cell electrochemical model, and a cell thermal model. This model achieves the simultaneous correlation of electrode design parameters, structural design parameters, and specific energy-maximum temperature-maximum temperature difference performance indicators with relatively low computational complexity. By combining a multi-objective optimization framework with a surrogate model method, a simultaneous multi-objective optimization method for electrode design parameters and structural design parameters based on a surrogate model is constructed. This method can obtain a design optimization scheme that balances specific energy-maximum temperature-maximum temperature difference performance indicators under limited computational resources.

[0028] Example 1 This invention provides a method for simultaneous multi-objective optimization of electrode and structural parameters in a high-capacity lithium-ion battery, such as... Figure 1 As shown, it includes: Step S1: Identify the high-capacity lithium-ion cell to be optimized, and obtain the electrode design parameters, structural design parameters, electrochemical parameters, and thermal parameters to construct a multi-scale structure-property model of the electrode-current collector-cell. The electrode design parameters and structural design parameters are obtained through direct measurement, while the electrochemical parameters and thermal parameters are obtained through indirect correction. Step S1 includes the following steps: Step S1.1: Obtain the electrode design parameters of the lithium-ion cell to be optimized by direct measurement method, including but not limited to: positive electrode thickness, positive electrode porosity, negative electrode porosity, and NP ratio.

[0029] Step S1.2: Obtain the structural design parameters of the lithium-ion cell to be optimized by direct measurement, including but not limited to: current collector thickness, current collector aspect ratio, and number of electrode pairs.

[0030] Step S1.3: Construct a multi-scale structure-property model framework for the electrode-current collector-cell, such as... Figure 2As shown, the electrode-current collector-cell multi-scale structure-property model couples three sub-models describing battery characteristics at different scales, including an electrode electrochemical model, a current collector electroelectric model, and a cell thermal model. The electrode electrochemical model is characterized by including the following equations: (1) Solid-state charge conservation:

[0031] (2) Conservation of charge in the liquid phase:

[0032] (3) Solid phase mass conservation:

[0033] (4) Conservation of liquid phase mass:

[0034] (5) Electrode dynamics:

[0035] (6) Electrochemical heat generation rate:

[0036] in, For time, The distance to the negative electrode current collector-negative electrode interface. The distance to the center of the particle. For solid-state potential, The liquid phase potential, Electrolyte concentration, This represents the concentration of lithium in the solid phase. This represents the average temperature of the battery cell. The electrochemical heat generation rate; This represents the volume fraction of the solid phase. For solid-state conductivity, The volume current density, It is the liquid volume fraction. Liquid phase conductivity, The transference number of ions. For activity, Let be the solid-phase diffusion coefficient. Where is the liquid phase diffusion coefficient, Where is the particle radius, The reaction rate constant is... For the maximum solid-phase lithium concentration, The surface lithium concentration, The anode exchange coefficient, The cathode exchange coefficient, This is an overpotential. This is the electrode equilibrium potential. For solid-phase Bruggeman index, The Bruggeman index for liquid phase. Let be the ideal gas constant. For Faraday constant, subscript The domain of application of the equations is indicated. In the electrode electrochemical model, the domains of application for solid-phase charge conservation and electrode kinetics equations are the negative and positive electrodes; the domains of application for liquid-phase charge conservation and liquid-phase mass conservation equations are the negative electrode, the membrane, and the positive electrode; and the domains of application for solid-phase mass conservation equations are the negative electrode particles and the positive electrode particles. The relationship between the solid-phase diffusion coefficient, the reaction rate constant, and temperature is described using the Arrhenius equation.

[0037] The current collector hydroelectric model includes the following equations: (1) Solid-state charge conservation:

[0038] (2) Heat generation rate of the current collector:

[0039] in, The solid-phase potential of the current collector. Rated current, For the number of pole pieces, The cross-sectional area of ​​the collector is... For the current collector thickness, Heat generation rate of the current collector; subscript This indicates the domain of application of the equation. In the current collector model, the domain of application of the equation is the negative current collector and the positive current collector.

[0040] The cell thermal model includes the following equations: (1) Electrode heat conservation equation:

[0041] (2) Cell heat conservation equation:

[0042] in, For temperature, , , For three-dimensional spatial coordinate components, The density of the pole ears, The specific heat capacity at constant pressure of the electrode tab. For the thermal conductivity of the electrode, The equivalent resistance is calculated based on the tab size. The volume of the pole ears, For cell density, For the specific heat capacity of the battery cell at constant voltage, Thermal conductivity in the planar direction of the battery cell Thermal conductivity in the thickness direction of the battery cell. Let be the heat generation rate of the battery cell. The domain of the electrode heat conservation equation is the negative and positive electrode tabs, while the domain of the battery cell heat conservation equation is the entire battery cell region.

[0043] The cell region is a multiphase mixture of electrodes, separator, current collector, and electrolyte. Its density, specific heat capacity at constant pressure, thermal conductivity, and heat generation rate are calculated using the following superposition formula: (1) Cell density:

[0044] (2) Specific heat capacity of the battery cell:

[0045] (3) Thermal conductivity of the cell in the planar direction:

[0046] (4) Thermal conductivity in the thickness direction of the battery cell:

[0047] (5) Cell heat generation rate:

[0048] in, For the first j Density of each region For the first j The specific heat capacity under constant pressure in each region For the first j Thermal conductivity of each region For the first j The thickness of each region, For the first j Solid density of each region For the first j Solid-state isobaric specific heat capacity of each region For the first j Solid-phase thermal conductivity of each region The electrolyte density, The specific heat capacity of the electrolyte. The value represents the thermal conductivity of the electrolyte; the subscripts nc, ne, sp, pe, and pc represent the negative electrode current collector, negative electrode, diaphragm, positive electrode, and positive electrode current collector, respectively.

[0049] The electrode electrochemical model, the current collector electroelectric model, and the cell thermal model are coupled in the following manner: the electrode electrochemical model receives the average cell temperature calculated from the cell thermal model and returns the electrochemical heat generation rate; the cell thermal model receives the electrochemical heat generation rate from the electrode electrochemical model and the current collector heat generation rate from the current collector electroelectric model, and returns the average cell temperature to the electrode electrochemical model.

[0050] Step S1.4: Conduct discharge tests on lithium-ion cells under different ambient temperatures and different rates. The specific test conditions can be set as follows: ambient temperature includes 20℃, 25℃, and 30℃, and rate includes 0.5 times, 1 times, and 1.5 times the rated rate. Different test conditions are combined using a fully factorial design. The specific steps are as follows: (1) Place the lithium-ion cell to be optimized in a constant temperature chamber, set the temperature of the constant temperature chamber to 25℃, and charge the cell to full charge in constant current and constant voltage mode, and then let it stand for 2 hours; (2) Set the temperature of the constant temperature chamber to the specified temperature, and discharge the cell to the cutoff voltage at a constant current of the specified rate, and record the process time, current, voltage, and temperature; (3) Repeat the above steps until all test conditions are completed.

[0051] Step 1.5: Calibrate electrochemical and thermal parameters to ensure that the voltage and temperature response curves of the electrode-current collector-cell multi-scale structure-property model simulation at different ambient temperatures and rate of change are consistent with experimental results. Electrochemical parameters include: solid-phase diffusion coefficient, solid-phase diffusion activation energy, reaction rate constant, reaction activation energy, initial solid-phase lithium concentration, Bruggeman index, etc. Thermal parameters include: material density, specific heat capacity at constant pressure, thermal conductivity, and equivalent resistance.

[0052] Step S2: Determine the adjustable range of electrode design parameters and structural design parameters. Within the adjustable range, randomly sample design parameters and input these samples into the electrode-current collector-cell multi-scale structure-property model to obtain simulation results of specific energy, maximum temperature, and maximum temperature difference performance indicators. Then, filter the current non-dominated solution set from the sample design parameters. Step S2 includes the following steps: Step S2.1: Determine the adjustable range of electrode design parameters and structural design parameters, including: (1) Adjustable range of positive electrode thickness: ; (2) Adjustable range of positive electrode porosity: ; (3) Adjustable range of negative electrode porosity: ; (4) Adjustable range of electrode logarithms: ; (5) Adjustable range of current collector thickness:

[0053] (6) Adjustable range of current collector aspect ratio: .

[0054] Step S2.2: Randomly sample design parameters within the adjustable range. Methods for randomly sampling design parameters include, but are not limited to: Monte Carlo sampling, orthogonal sampling, Latin hypercube sampling, etc.

[0055] Step S2.3: Input the design parameter samples into the electrode-current collector-cell multi-scale structure-property model, simulate the voltage and temperature response curves under different design parameters, and then obtain the simulation results of the specific energy-maximum temperature-maximum temperature difference performance index based on the simulated voltage and temperature response curves. The specific energy-maximum temperature-maximum temperature difference performance index should include the following indicators: (1) Specific energy:

[0056] (2) Maximum temperature:

[0057] (3) Maximum temperature difference:

[0058] in, This refers to the total mass of the battery cells. For voltage, This represents the total discharge time.

[0059] Step S2.4: Based on the simulation results of the specific energy-maximum temperature-maximum temperature difference performance index, filter the current non-dominated solution set.

[0060] Step S3: Construct surrogate models based on the design parameter samples, simulation results of multidimensional performance indicators, and the current non-dominated solution set. The surrogate models include structure-performance surrogate models and Pareto front surrogate models. Specifically, the structure-performance surrogate models include specific energy inverse number structure-performance surrogate models, maximum temperature structure-performance surrogate models, and maximum temperature difference structure-performance surrogate models. The inputs are electrode design parameters and structural design parameters, and the outputs are specific energy inverse number, maximum temperature, and maximum temperature difference performance indicators, respectively, which can be mathematically represented as:

[0061] in, , , These are, respectively, the structure-property surrogate model based on specific energy opposite number, the structure-property surrogate model based on maximum temperature, and the structure-property surrogate model based on maximum temperature difference. , , These are approximate values ​​for the specific energy inverse, maximum temperature, and maximum temperature difference, respectively. To design variable vectors, The thickness of the positive electrode. Porosity is the positive electrode porosity. The negative porosity, For NP ratio, For the number of pole pieces, For the current collector thickness, Aspect ratio of the current collector. A cubic radial basis function interpolation model is selected as the model structure of this surrogate model, and a structure-property surrogate model is constructed using design parameter samples and simulation results of specific energy-maximum temperature-maximum temperature difference performance indicators.

[0062] The Pareto frontier surrogate model takes the maximum temperature and maximum temperature difference as inputs and outputs the inverse of the specific energy, which can be mathematically represented as:

[0063] in, For the Pareto frontier agent model, , , These represent the approximate value of the inverse of the specific energy at the Pareto front, the maximum temperature at the Pareto front, and the maximum temperature difference at the Pareto front, respectively. Let the specific energy on the Pareto front be a function of the maximum temperature and the maximum temperature difference on the Pareto front. A linear radial basis function interpolation model is selected as the model structure for this surrogate model, and the specific energy, maximum temperature, and maximum temperature difference in the current non-dominated solution set are used to construct the Pareto front surrogate model.

[0064] Step S4: Based on the structure-property surrogacy model and the Pareto frontier surrogacy model, screen the design parameter samples to be evaluated. For example... Figure 3 As shown, step S4 includes screening design parameter samples based on a target value strategy, a near-multi-objective optimization strategy, a near-single-objective optimization strategy, a local perturbation strategy, and a random sampling strategy. Specifically: The maximum-minimum distance strategy is used to select samples of design parameters to be evaluated. Specifically, the maximum-minimum distance strategy involves: first, constructing the following sub-optimization problem based on the Pareto front surrogate model;

[0065] in, This represents the minimum value of the maximum temperature in the current non-dominated solution set. This represents the maximum value of the maximum temperature in the current non-dominated solution set. This represents the minimum of the maximum temperature difference in the current non-dominated solution set. The maximum temperature difference in the current non-dominated solution set is the maximum value, and a single-objective genetic algorithm is used to solve it, thereby obtaining the maximum-minimum distance solution. Subsequently, based on the structure-performance proxy model, the following sub-optimization problems were constructed:

[0066] A multi-objective genetic algorithm is then used to solve the problem, thereby obtaining a set of non-dominated solutions. Finally, solutions satisfying the given conditions are selected from the set of non-dominated solutions. At most one solution that does not appear in the design parameter sample is added to the design parameter sample to be evaluated.

[0067] A sample of design parameters to be evaluated is selected based on an approximate multi-objective optimization strategy. Specifically, the approximate multi-objective optimization strategy involves constructing the following sub-optimization problems based on a structure-performance surrogate model:

[0068] A multi-objective genetic algorithm is used to solve the problem, thereby obtaining the set of non-dominated solutions. At most seven non-dominated solutions that do not appear in the design parameter sample are randomly selected from the set of non-dominated solutions and added to the design parameter sample to be evaluated.

[0069] The sample of design parameters to be evaluated is selected based on an approximate single-objective optimization strategy. Specifically, the approximate single-objective optimization strategy involves constructing the following three sub-optimization problems based on a structure-performance surrogate model:

[0070] Each solution was obtained using a single-objective genetic algorithm, resulting in three optimal solutions. At most three solutions that did not appear in the design parameter sample were selected and added to the design parameter sample to be evaluated.

[0071] The selection of design parameter samples for evaluation is based on a local perturbation strategy. Specifically, the local perturbation strategy involves: first, applying random perturbation centered on the design parameter sample to generate a set of no more than 1000 candidate design parameter samples for each sample; second, evaluating the approximate values ​​of specific energy inverse number, maximum temperature, and maximum temperature difference for each candidate design parameter sample based on a structure-performance surrogate model; further, calculating the minimum Euclidean distance from each candidate design parameter sample to the design parameter sample; further, normalizing the specific energy inverse number, maximum temperature, maximum temperature difference, and minimum Euclidean distance for each candidate design parameter sample using the min-max method to obtain normalized specific energy inverse number, normalized maximum temperature, normalized maximum temperature difference, and normalized minimum Euclidean distance; and finally, using the weighted sum of the average of the normalized specific energy inverse number, normalized maximum temperature, and normalized maximum temperature difference with the normalized minimum Euclidean distance as the comprehensive score for each candidate design parameter sample.

[0072] in, For the first i The overall score of the candidate design parameter samples To normalize the scoring weights, , , These are the normalized specific energy inverse, normalized maximum temperature, normalized maximum temperature difference, and normalized minimum Euclidean distance, respectively. Further, for each group of candidate design parameter samples, only the candidate design parameter sample with the lowest comprehensive score is retained, and other samples are deleted. Further, based on the approximate values ​​of the specific energy inverse, maximum temperature, and maximum temperature difference of the retained candidate design parameter samples, the dominance relationships between candidate design parameter samples and design parameter samples, as well as among candidate design parameter samples themselves, are determined, and candidate design parameter samples not dominated by design parameter samples or other candidate design parameter samples are retained. Finally, the retained candidate design parameter samples are added to the design parameter samples to be evaluated.

[0073] A random sampling strategy was used to screen the design parameter samples to be evaluated. Specifically, the random sampling strategy involved: First, randomly selecting no more than 1000 candidate design parameter samples within an adjustable range. Methods for random sampling included, but were not limited to, Monte Carlo sampling, orthogonal sampling, and Latin hypercube sampling. Second, based on a structure-performance surrogate model, the approximate values ​​of the specific energy inverse number, maximum temperature, and maximum temperature difference for each candidate design parameter sample were evaluated. Further, the minimum Euclidean distance from each candidate design parameter sample to the design parameter sample was calculated. Further, the min-max method was used to normalize the specific energy inverse number, maximum temperature, maximum temperature difference, and minimum Euclidean distance for each candidate design parameter sample. Finally, the weighted sum of the normalized average of the specific energy inverse number, normalized maximum temperature, and normalized maximum temperature difference, and the normalized minimum Euclidean distance was used as the comprehensive score for each candidate design parameter sample, and the candidate design parameter sample with the lowest comprehensive score was added to the set of design parameter samples to be evaluated.

[0074] Step S5: Calculate the specific energy, maximum temperature, and maximum temperature difference performance index values ​​corresponding to the design parameter sample to be evaluated based on the electrode-current collector-cell multi-scale structure-property model. Step S6: Update the design parameter samples, the simulation results of the specific energy-maximum temperature-maximum temperature difference performance index, and the current non-dominated solution set. Determine if the maximum number of simulations for the electrode-current collector-cell multi-scale structure-property model has been reached. If yes, output the final non-dominated solution set and proceed to step S7; otherwise, return to step S3. A schematic diagram of the final non-dominated solution set is shown below. Figure 4 As shown.

[0075] Step S7: Draw the Pareto front of the final non-dominated solution set, select the region in the Pareto front that meets the design requirements of specific energy, maximum temperature, and maximum temperature difference, and extract the optimization scheme of electrode design parameters and structural design parameters in this region.

[0076] Example 2 This invention also provides a multi-objective optimization system for the synchronous optimization of electrode and structural parameters of a large-capacity lithium-ion battery. The multi-objective optimization system for the synchronous optimization of electrode and structural parameters of a large-capacity lithium-ion battery can be implemented by executing the process steps of the multi-objective optimization method for the synchronous optimization of electrode and structural parameters of a large-capacity lithium-ion battery. That is, those skilled in the art can understand the multi-objective optimization method for the synchronous optimization of electrode and structural parameters of a large-capacity lithium-ion battery as a preferred embodiment of the multi-objective optimization system for the synchronous optimization of electrode and structural parameters of a large-capacity lithium-ion battery.

[0077] A high-capacity lithium-ion battery electrode and structural parameter synchronous multi-objective optimization system provided by the present invention includes: Module M1: Identifies the high-capacity lithium-ion cell to be optimized and obtains electrode design parameters, structural design parameters, electrochemical parameters, and thermal parameters to construct a multi-scale structure-property model of the electrode-current collector-cell. Electrode design parameters include positive electrode thickness, positive electrode porosity, negative electrode porosity, and NP ratio. Structural design parameters include current collector thickness, current collector aspect ratio, and number of electrode pairs. Electrochemical parameters include solid-phase diffusion coefficient, solid-phase diffusion activation energy, reaction rate constant, reaction activation energy, initial solid-phase lithium concentration, and Bruggeman index. Thermal parameters include material density, specific heat capacity at constant pressure, thermal conductivity, and equivalent resistance. The multi-scale structure-property model of the electrode-current collector-cell couples multiple sub-models describing battery characteristics at different scales, including an electrode electrochemical model, a current collector electroelectric model, and a cell thermal model. The electrode electrochemical model includes: Solid-state charge conservation: .

[0078] Liquid phase charge conservation: .

[0079] Solid-phase mass conservation: .

[0080] Liquid phase mass conservation: .

[0081] Electrode dynamics: .

[0082] Electrochemical heat generation rate: .

[0083] in, For time, The distance to the negative electrode current collector-negative electrode interface. The distance to the center of the particle. For solid-state potential, The liquid phase potential, Electrolyte concentration, This represents the concentration of lithium in the solid phase. This represents the average temperature of the battery cell. The electrochemical heat generation rate is denoted as . This represents the volume fraction of the solid phase. For solid-state conductivity, The volume current density, It is the liquid volume fraction. Liquid phase conductivity, The transference number of ions. For activity, Let be the solid-phase diffusion coefficient. Where is the liquid phase diffusion coefficient, Where is the particle radius, The reaction rate constant is... For the maximum solid-phase lithium concentration, The surface lithium concentration, The anode exchange coefficient, The cathode exchange coefficient, This is an overpotential. This is the electrode equilibrium potential. For solid-phase Bruggeman index, The Bruggeman index for liquid phase. Let be the ideal gas constant. For Faraday constant, subscript This indicates the scope of the equation.

[0084] The current collector hydroelectric model includes: Solid-state charge conservation: .

[0085] Heat generation rate of the current collector: .

[0086] in, The solid-phase potential of the current collector. Rated current, For the number of pole pieces, The cross-sectional area of ​​the collector is... For the current collector thickness, The heat generation rate of the current collector. (Subscript) This indicates the scope of the equation.

[0087] The cell thermal model includes: Electrode heat conservation equation: .

[0088] Cell heat conservation equation: .

[0089] in, For temperature, , , For three-dimensional spatial coordinate components, The density of the pole ears, The specific heat capacity at constant pressure of the electrode tab. For the thermal conductivity of the electrode, The equivalent resistance is calculated based on the tab size. The volume of the pole ears, For cell density, For the specific heat capacity of the battery cell at constant voltage, Thermal conductivity in the planar direction of the battery cell Thermal conductivity in the thickness direction of the battery cell. The cell heat generation rate is defined as follows: In the electrode electrochemical model, the domains of application of the solid phase charge conservation and electrode kinetic equations are the negative and positive electrodes; the domains of application of the liquid phase charge conservation and liquid phase mass conservation equations are the negative electrode, separator, and positive electrode; and the domain of application of the solid phase mass conservation equation is the negative electrode particles and positive electrode particles. The relationship between the solid phase diffusion coefficient, reaction rate constant, and temperature is described using the Arrhenius equation. In the current collector model, the domains of application of the equations are the negative and positive electrode current collectors. In the cell thermal model, the domains of application of the tab heat conservation equation are the negative and positive electrode tabs, and the domain of application of the cell heat conservation equation is the cell region. The cell region is a multiphase mixture of electrodes, separator, current collector, and electrolyte. Its density, specific heat capacity at constant pressure, thermal conductivity, and heat generation rate are calculated using the following superposition formula: Cell density: .

[0090] Cell specific heat capacity: .

[0091] Thermal conductivity of the cell in the planar direction: .

[0092] Thermal conductivity in the thickness direction of the battery cell: .

[0093] Cell heat generation rate: .

[0094] in, For the first j Density of each region For the first j The specific heat capacity under constant pressure in each region For the first j Thermal conductivity of each region For the first j The thickness of each region, For the first j Solid density of each region For the first j Solid-state isobaric specific heat capacity of each region For the first j Solid-phase thermal conductivity of each region The electrolyte density, The specific heat capacity of the electrolyte. This represents the thermal conductivity of the electrolyte. The subscripts nc, ne, sp, pe, and pc represent the negative electrode current collector, negative electrode, diaphragm, positive electrode, and positive electrode current collector, respectively.

[0095] The coupling mechanism of the electrode electrochemical model, the current collector electroelectric model, and the cell thermal model includes: the electrode electrochemical model receives the average cell temperature calculated from the cell thermal model and returns the electrochemical heat generation rate. The cell thermal model receives the electrochemical heat generation rate from the electrode electrochemical model and the current collector heat generation rate from the current collector electroelectric model, and returns the average cell temperature to the electrode electrochemical model.

[0096] Module M2: Determines the adjustable range of electrode and structural design parameters. Within this range, it randomly samples design parameters and inputs these samples into a multi-scale structure-property model of the electrode-current collector-cell. This yields simulation results for specific energy, maximum temperature, and maximum temperature difference performance indicators. The current non-dominated solution set is then selected from the design parameter samples. Methods for randomly sampling design parameters include Monte Carlo sampling, orthogonal sampling, and Latin hypercube sampling. The specific energy, maximum temperature, and maximum temperature difference performance indicators include: Specific energy: .

[0097] Maximum temperature: .

[0098] Maximum temperature difference: .

[0099] in, This refers to the total mass of the battery cells. For voltage, This represents the total discharge time.

[0100] Module M3: Constructs surrogate models based on design parameter samples, simulation results of specific energy-maximum temperature-maximum temperature difference performance indicators, and the current non-dominated solution set. The surrogate models include a structure-property surrogate model and a Pareto front surrogate model. The structure-property surrogate model uses a cubic radial basis function interpolation model, and calculates the radial basis function coefficients using the design parameter samples and simulation results of specific energy-maximum temperature-maximum temperature difference performance indicators. The Pareto front surrogate model uses a linear radial basis function interpolation model, and calculates the radial basis function coefficients using the current non-dominated solution set.

[0101] Module M4: Based on the structure-efficacy surrogate model and the Pareto front surrogate model, it filters the sample of design parameters to be evaluated. Module M4 includes filtering the sample of design parameters to be evaluated based on the maximum-minimum distance strategy, the approximate multi-objective optimization strategy, the approximate single-objective optimization strategy, the local perturbation strategy, and the random sampling strategy. The maximum-minimum distance strategy includes: firstly, based on the Pareto front surrogate model, constructing the following sub-optimization problems...

[0102] A single-objective genetic algorithm is used to solve the problem, thereby obtaining the maximum-minimum distance solution. Subsequently, based on the structure-function proxy model, the following sub-optimization problems were constructed.

[0103] A multi-objective genetic algorithm is used to solve the problem, thereby obtaining a set of non-dominated solutions. Finally, a solution that satisfies the given conditions is selected from the set of non-dominated solutions. At most one solution that does not appear in the design parameter sample is added to the design parameter sample to be evaluated.

[0104] Approximate multi-objective optimization strategies include constructing the following sub-optimization problems based on structure-performance surrogate models:

[0105] A multi-objective genetic algorithm is used to solve the problem, thereby obtaining the non-dominated solution set. At most 7 non-dominated solutions that do not appear in the design parameter sample are randomly selected from the non-dominated solution set and added to the design parameter sample to be evaluated.

[0106] The approximate single-objective optimization strategy includes constructing the following three sub-optimization problems based on the structure-performance surrogate model.

[0107] Each solution is obtained by using a single-objective genetic algorithm. At most three solutions that do not appear in the design parameter sample are selected and added to the design parameter sample to be evaluated.

[0108] The local perturbation strategy includes: first, randomly perturbing the design parameter samples to generate a set of no more than 1000 candidate design parameter samples for each sample; second, based on the structure-performance surrogate model, evaluating the approximate values ​​of the specific energy inverse, maximum temperature, and maximum temperature difference for each candidate design parameter sample; then, calculating the minimum Euclidean distance from each candidate design parameter sample to the design parameter sample; and using the min-max method to normalize the specific energy inverse, maximum temperature, maximum temperature difference, and minimum Euclidean distance for each candidate design parameter sample to obtain normalized specific energy inverse, normalized maximum temperature, normalized maximum temperature difference, and normalized minimum Euclidean distance. Further, using the weighted sum of the average of the normalized specific energy inverse, normalized maximum temperature, and normalized maximum temperature difference with the normalized minimum Euclidean distance as the comprehensive score for each candidate design parameter sample, as shown in the following formula:

[0109] in, For the first i The overall score of the candidate design parameter samples To normalize the scoring weights, , , These are the normalized inverse specific energy, normalized maximum temperature, normalized maximum temperature difference, and normalized minimum Euclidean distance, respectively. Next, for each group of candidate design parameter samples, only one candidate design parameter sample with the lowest comprehensive score is retained, and other samples are deleted. Based on the approximate values ​​of the inverse specific energy, maximum temperature, and maximum temperature difference of the retained candidate design parameter samples, the dominance relationship between candidate design parameter samples and design parameter samples, as well as among candidate design parameter samples, is determined. Candidate design parameter samples that are not dominated by design parameter samples or other candidate design parameter samples are retained. Finally, the retained candidate design parameter samples are added to the design parameter samples to be evaluated.

[0110] The random sampling strategy includes first randomly selecting no more than 1000 candidate design parameter samples within an adjustable range; second, based on the structure-performance surrogate model, evaluating the approximate values ​​of specific energy inverse, maximum temperature, and maximum temperature difference for each candidate design parameter sample, calculating the minimum Euclidean distance from each candidate design parameter sample to the design parameter sample, and normalizing the specific energy inverse, maximum temperature, maximum temperature difference, and minimum Euclidean distance for each candidate design parameter sample using the min-max method; finally, using the weighted sum of the average of the normalized specific energy inverse, normalized maximum temperature, and normalized maximum temperature difference with the normalized minimum Euclidean distance as the comprehensive score for each candidate design parameter sample, and selecting the candidate design parameter sample with the lowest comprehensive score to add to the design parameter sample to be evaluated.

[0111] Module M5: Calculates the specific energy, maximum temperature, and maximum temperature difference performance index values ​​corresponding to the design parameter samples to be evaluated based on the multi-scale structure-property model of electrode-current collector-cell.

[0112] Module M6: Updates the design parameter samples, simulation results of specific energy-maximum temperature-maximum temperature difference performance indicators, and the current non-dominated solution set. It then determines whether the maximum number of simulations for the electrode-current collector-cell multi-scale structure-property model has been reached. If so, it outputs the final non-dominated solution set and triggers module M7. If not, it triggers module M3.

[0113] Module M7: Draw the Pareto front of the final non-dominated solution set, select the region in the Pareto front that meets the design requirements for specific energy, maximum temperature, and maximum temperature difference, and extract the optimization scheme of electrode design parameters and structural design parameters for this region.

[0114] Those skilled in the art will understand that, besides implementing the system and its various devices, modules, and units provided by this invention in the form of purely computer-readable program code, the same functions can be achieved entirely through logical programming of the method steps, making the system and its various devices, modules, and units of this invention function in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, the system and its various devices, modules, and units provided by this invention can be considered as a hardware component, and the devices, modules, and units included therein for implementing various functions can also be considered as structures within the hardware component; alternatively, the devices, modules, and units for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0115] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the features in Embodiments 1 and 2 of this application can be arbitrarily combined with each other.

Claims

1. A method for simultaneous multi-objective optimization of electrode and structural parameters in a high-capacity lithium-ion battery, characterized in that, include: Step S1: Determine the high-capacity lithium-ion cell to be optimized, and obtain the electrode design parameters, structural design parameters, electrochemical parameters, and thermal parameters, thereby constructing a multi-scale structure-property model of the electrode-current collector-cell. Step S2: Determine the adjustable range of electrode design parameters and structural design parameters, randomly select design parameter samples within the adjustable range, and input the design parameter samples into the electrode-current collector-cell multi-scale structure-property model to obtain the simulation results of specific energy-maximum temperature-maximum temperature difference performance indicators, and then filter the current non-dominated solution set from the design parameter samples; Step S3: Construct a structure-property surrogate model using the design parameter samples and simulation results of specific energy-maximum temperature-maximum temperature difference performance indicators, and construct a Pareto front surrogate model using the current non-dominated solution set; Step S4: Based on the structure-property surrogacy model and the Pareto frontier surrogacy model, screen the design parameter samples to be evaluated; Step S5: Calculate the specific energy, maximum temperature, and maximum temperature difference performance index values ​​corresponding to the design parameter sample to be evaluated based on the electrode-current collector-cell multi-scale structure-property model; Step S6: Update the design parameter samples, the simulation results of the specific energy-maximum temperature-maximum temperature difference performance index and the current non-dominated solution set, and determine whether the maximum number of simulations of the electrode-current collector-cell multi-scale structure-property model has been reached. If yes, output the final non-dominated solution set and execute step S7; otherwise, return to step S3. Step S7: Draw the Pareto front of the final non-dominated solution set, select the region in the Pareto front that meets the design requirements of specific energy, maximum temperature, and maximum temperature difference, and extract the optimization scheme of electrode design parameters and structural design parameters in this region.

2. The method for simultaneous multi-objective optimization of electrode and structural parameters of a high-capacity lithium-ion battery according to claim 1, characterized in that, The electrode design parameters include positive electrode thickness, positive electrode porosity, negative electrode porosity, and NP ratio. The structural design parameters include current collector thickness, current collector aspect ratio, and number of electrode pairs; The electrochemical parameters include solid-phase diffusion coefficient, solid-phase diffusion activation energy, reaction rate constant, reaction activation energy, initial solid-phase lithium concentration, and Bruggeman index. The thermal parameters include material density, specific heat capacity at constant pressure, thermal conductivity, and equivalent resistance. The electrode-current collector-cell multi-scale structure-property model is coupled with multiple sub-models describing battery characteristics at different scales, including the electrode electrochemical model, the current collector hydroelectric model, and the cell thermal model. The methods for randomly sampling the design parameters include Monte Carlo sampling, orthogonal sampling, and Latin hypercube sampling. The specific energy-maximum temperature-maximum temperature difference performance indicators include: Specific energy: ; Maximum temperature: ; Maximum temperature difference: ; in, For time, , , For three-dimensional spatial coordinate components, For voltage, This represents the average temperature of the battery cell. For temperature, This refers to the total mass of the battery cells. Rated current, This represents the total discharge time.

3. The method for simultaneous multi-objective optimization of electrode and structural parameters of a large-capacity lithium-ion battery according to claim 2, characterized in that, The electrode electrochemical model includes: Solid-state charge conservation: ; Liquid phase charge conservation: ; Solid-phase mass conservation: ; Liquid phase mass conservation: ; Electrode dynamics: ； Electrochemical heat generation rate: ； in, The distance to the negative electrode current collector-negative electrode interface. The distance to the center of the particle. For solid-state potential, The liquid phase potential, Electrolyte concentration, This represents the concentration of lithium in the solid phase. This represents the average temperature of the battery cell. The electrochemical heat generation rate; This represents the volume fraction of the solid phase. For solid-state conductivity, The volume current density, It is the liquid volume fraction. Liquid phase conductivity, The transference number of ions. For activity, The solid-phase diffusion coefficient is... Where is the liquid phase diffusion coefficient, Where is the particle radius, The reaction rate constant is... For the maximum solid-phase lithium concentration, The surface lithium concentration, The anode exchange coefficient, The cathode exchange coefficient, This is an overpotential. This is the electrode equilibrium potential. For solid-phase Bruggeman index, The Bruggeman index for liquid phase. Let be the ideal gas constant. For Faraday constant, subscript Indicate the scope of the equation; The current collector hydroelectric model includes: Solid-state charge conservation: ; Heat generation rate of the current collector: ; Among them, the dependent variable The solid-phase potential of the current collector. Rated current, For the number of pole pieces, The cross-sectional area of ​​the collector is... For the current collector thickness, Heat generation rate of the current collector; subscript Indicate the scope of the equation; The cell thermal model includes: Electrode heat conservation equation: ; Cell heat conservation equation: ; in, The density of the pole ears, The specific heat capacity at constant pressure of the electrode tab. For the thermal conductivity of the electrode, The equivalent resistance is calculated based on the tab size. The volume of the pole ears, For cell density, For the specific heat capacity of the battery cell at constant voltage, Thermal conductivity in the planar direction of the battery cell Thermal conductivity in the thickness direction of the battery cell. The rate at which the battery cell generates heat.

4. The method for simultaneous multi-objective optimization of electrode and structural parameters of a high-capacity lithium-ion battery according to claim 3, characterized in that, In the electrode electrochemical model, the domains of solid phase charge conservation and electrode kinetic equations are the negative and positive electrodes, the domains of liquid phase charge conservation and liquid phase mass conservation equations are the negative electrode, the membrane, and the positive electrode, and the domains of solid phase mass conservation equations are the negative electrode particles and the positive electrode particles. The relationship between solid phase diffusion coefficient, reaction rate constant and temperature is described by the Arrhenius formula. In the current collector model, the domain of application of the equation is the negative electrode current collector and the positive electrode current collector; In the cell thermal model, the domain of the tab heat conservation equation is the negative and positive tabs, while the domain of the cell heat conservation equation is the cell region. The coupling method of the electrode electrochemical model, the current collector hydroelectric model, and the cell thermal model includes: the electrode electrochemical model receives the average cell temperature calculated from the cell thermal model and returns the electrochemical heat generation rate; The cell thermal model receives the electrochemical heat generation rate from the electrode electrochemical model and the current collector heat generation rate from the current collector electrochemical model, and returns the average cell temperature to the electrode electrochemical model.

5. The method for simultaneous multi-objective optimization of electrode and structural parameters of a high-capacity lithium-ion battery according to claim 4, characterized in that, The cell region is a multiphase mixture of electrodes, separator, current collector, and electrolyte. Its density, specific heat capacity at constant pressure, thermal conductivity, and heat generation rate are calculated using the following superposition formula: Cell density: ; Cell specific heat capacity: ; Thermal conductivity of the cell in the planar direction: ; Thermal conductivity in the thickness direction of the battery cell: ; Cell heat generation rate: ; in, For the first j Density of each region For the first j The specific heat capacity under constant pressure in each region For the first j Thermal conductivity of each region For the first j The thickness of each region, For the first j Solid density of each region For the first j Solid-state isobaric specific heat capacity of each region For the first j Solid-phase thermal conductivity of each region The electrolyte density, The specific heat capacity of the electrolyte. The value represents the thermal conductivity of the electrolyte; the subscripts nc, ne, sp, pe, and pc represent the negative electrode current collector, negative electrode, diaphragm, positive electrode, and positive electrode current collector, respectively.

6. The method for simultaneous multi-objective optimization of electrode and structural parameters of a high-capacity lithium-ion battery according to claim 1, characterized in that, The structure-property surrogate model includes the specific energy inverse number structure-property surrogate model, the maximum temperature structure-property surrogate model, and the maximum temperature difference structure-property surrogate model. The inputs are electrode design parameters and structural design parameters, and the outputs are specific energy inverse number, maximum temperature, and maximum temperature difference, respectively. The model structure adopts a cubic radial basis function interpolation model. The Pareto front surrogate model takes the maximum temperature and maximum temperature difference as inputs and outputs the inverse of the specific energy. The model structure uses a linear radial basis function interpolation model.

7. The method for simultaneous multi-objective optimization of electrode and structural parameters of a high-capacity lithium-ion battery according to claim 1, characterized in that, Step S4 includes screening design parameter samples to be evaluated based on the maximum-minimum distance strategy, screening design parameter samples to be evaluated based on the approximate multi-objective optimization strategy, screening design parameter samples to be evaluated based on the approximate single-objective optimization strategy, screening design parameter samples to be evaluated based on the local perturbation strategy, and screening design parameter samples to be evaluated based on the random sampling strategy.

8. The method for simultaneous multi-objective optimization of electrode and structural parameters of a high-capacity lithium-ion battery according to claim 7, characterized in that, The maximum-minimum distance strategy includes: first, based on the Pareto front agent model, constructing the following sub-optimization problem. in, For the Pareto frontier agent model, This represents the minimum value of the maximum temperature in the current non-dominated solution set. This represents the maximum value of the maximum temperature in the current non-dominated solution set. This represents the minimum of the maximum temperature difference in the current non-dominated solution set. The maximum temperature difference in the current non-dominated solution set is the maximum value, and a single-objective genetic algorithm is used to solve it, thereby obtaining the maximum-minimum distance solution. Subsequently, based on the structure-function proxy model, the following sub-optimization problems were constructed. in, , , These are, respectively, the structure-property surrogate model based on specific energy opposite number, the structure-property surrogate model based on maximum temperature, and the structure-property surrogate model based on maximum temperature difference. To design variable vectors, The thickness of the positive electrode. Porosity is the positive electrode porosity. The negative porosity, For NP ratio, For the number of pole pieces, For the current collector thickness, The aspect ratio of the current collector is defined, and the subscripts min and max represent the lower and upper bounds of the adjustable range, respectively. A multi-objective genetic algorithm is used to solve the problem, thereby obtaining a set of non-dominated solutions. Finally, solutions that satisfy the given conditions are selected from the set of non-dominated solutions. At most one solution that does not appear in the design parameter sample is added to the design parameter sample to be evaluated. The approximate multi-objective optimization strategy includes: constructing the following sub-optimization problems based on the structure-efficacy surrogate model. A multi-objective genetic algorithm is used to solve the problem, thereby obtaining the non-dominated solution set. At most 7 non-dominated solutions that do not appear in the design parameter sample are randomly selected from the non-dominated solution set and added to the design parameter sample to be evaluated. The approximate single-objective optimization strategy includes: constructing the following three sub-optimization problems based on the structure-function surrogate model. Each solution is solved using a single-objective genetic algorithm to obtain three optimal solutions. At most three solutions that do not appear in the design parameter sample are selected and added to the design parameter sample to be evaluated. The local perturbation strategy includes: First, randomly perturbing the design parameter sample to generate a set of no more than 1000 candidate design parameter samples for each design parameter sample; second, evaluating the approximate values ​​of the specific energy inverse, maximum temperature, and maximum temperature difference for each candidate design parameter sample based on a structure-performance surrogate model, and calculating the minimum Euclidean distance from each candidate design parameter sample to the design parameter sample; further, normalizing the specific energy inverse, maximum temperature, maximum temperature difference, and minimum Euclidean distance for each candidate design parameter sample using the min-max method to obtain normalized specific energy inverse, normalized maximum temperature, and normalized maximum temperature difference. The difference and normalized minimum Euclidean distance are further used as the weighted sum of the average of the normalized specific energy inverse, normalized maximum temperature, normalized maximum temperature difference and normalized minimum Euclidean distance as the comprehensive score of each candidate design parameter sample. The candidate design parameter sample with the lowest comprehensive score is retained for each group of candidate design parameter samples. Furthermore, the dominance relationship between candidate design parameter samples and design parameter samples, as well as among candidate design parameter samples, is determined. Candidate design parameter samples that are not dominated by design parameter samples or other candidate design parameter samples are retained. Finally, the retained candidate design parameter samples are added to the design parameter samples to be evaluated. The random sampling strategy includes: First, randomly selecting no more than 1000 candidate design parameter samples within an adjustable range; second, based on the structure-performance surrogate model, evaluating the approximate values ​​of the specific energy inverse, maximum temperature, and maximum temperature difference for each candidate design parameter sample, and calculating the minimum Euclidean distance from each candidate design parameter sample to the design parameter sample; further, using the min-max method to normalize the specific energy inverse, maximum temperature, maximum temperature difference, and minimum Euclidean distance for each candidate design parameter sample to obtain normalized specific energy inverse, normalized maximum temperature, normalized maximum temperature difference, and normalized minimum Euclidean distance; finally, using the weighted sum of the average of the normalized specific energy inverse, normalized maximum temperature, and normalized maximum temperature difference and the normalized minimum Euclidean distance as the comprehensive score for each candidate design parameter sample, and selecting the candidate design parameter sample with the lowest comprehensive score to add to the design parameter sample to be evaluated.

9. A multi-objective optimization system for the electrode and structural parameters of a high-capacity lithium-ion battery, characterized in that, include: Module M1: Identify the high-capacity lithium-ion cell to be optimized, and obtain electrode design parameters, structural design parameters, electrochemical parameters, and thermal parameters, thereby constructing a multi-scale structure-property model of electrode-current collector-cell. Module M2: Determines the adjustable range of electrode design parameters and structural design parameters, randomly selects design parameter samples within the adjustable range, and inputs the design parameter samples into the electrode-current collector-cell multi-scale structure-property model to obtain the simulation results of specific energy-maximum temperature-maximum temperature difference performance indicators, and then filters the current non-dominated solution set from the design parameter samples; Module M3: Constructs a structure-property surrogate model using design parameter samples and simulation results of specific energy-maximum temperature-maximum temperature difference performance indicators, and constructs a Pareto front surrogate model using the current non-dominated solution set; Module M4: Based on the structure-performance surrogate model and the Pareto frontier surrogate model, screen the design parameter samples to be evaluated; Module M5: Calculates the specific energy, maximum temperature, and maximum temperature difference performance index values ​​corresponding to the design parameter samples to be evaluated based on the electrode-current collector-cell multi-scale structure-property model; Module M6: Updates the design parameter samples, the simulation results of the specific energy-maximum temperature-maximum temperature difference performance index and the current non-dominated solution set, and determines whether the maximum number of simulations of the electrode-current collector-cell multi-scale structure-property model has been reached. If yes, it outputs the final non-dominated solution set and then triggers module M7; otherwise, it triggers module M3. Module M7: Draw the Pareto front of the final non-dominated solution set, select the region in the Pareto front that meets the design requirements for specific energy, maximum temperature, and maximum temperature difference, and extract the optimization scheme of electrode design parameters and structural design parameters for this region.

Images