A mechanism-based lithium-ion battery life prediction method in a wide temperature range

Abstract

The application belongs to the technical field of power lithium ion batteries, and relates to a mechanism-based wide-temperature-range lithium ion battery life prediction method, which acquires battery product specifications, cell design parameters, measures intrinsic characteristic parameters, electrochemical parameters, kinetic parameters and thermodynamic parameters of electrode materials, and tests rate performance data of the battery under a wide temperature range; an electrochemical model and a thermal model are established to build an electrochemical-thermal coupling model; acceleration aging test data of the battery under different temperatures are acquired; a side reaction model is established to build an electrochemical-thermal-side reaction coupling model; the electrochemical-thermal-side reaction coupling model is verified and aging parameter correction is performed according to the measured data to obtain an electrochemical-thermal-side reaction life prediction model of the lithium ion battery. The application takes into account the internal physical and chemical processes of the battery and external environmental variables, can simulate the law of linear to nonlinear aging attenuation of the battery, and meets the demand of optimizing design of the battery product.

Description

Technical Field

[0001] This invention belongs to the field of power lithium-ion battery technology, specifically relating to a mechanism-based method for predicting the lifespan of wide-temperature-range lithium-ion batteries. Background Technology

[0002] Lithium-ion batteries, as the primary power source for new energy vehicles, account for nearly 40% of the total vehicle cost. Whether parked or in operation, physical, chemical, and electrochemical reactions within the lithium-ion battery cause varying degrees of aging, shortening its lifespan and increasing the purchase cost for consumers. Battery aging is unavoidable. Optimizing battery design through relevant technologies to slow down the aging rate and extend battery life will facilitate the large-scale popularization and commercial development of new energy vehicles, accelerating the achievement of dual-carbon goals.

[0003] In existing technologies, lithium-ion battery life prediction methods include experimental methods and model prediction methods. Experimental methods require extensive testing and disassembly of finished batteries. For example, the lithium-ion battery capacity degradation prediction method based on chemical degradation and mechanical degradation developed in patent CN115047364A requires cyclic testing of multiple batteries and disassembly of batteries with different cycle counts to measure the composition and thickness of the solid electrolyte interface on the negative electrode, as well as the length and depth of surface cracks in the negative electrode material. Then, a mathematical algorithm is used to predict the battery capacity. This method requires extensive experimental testing and battery disassembly, which is not only time-consuming and labor-intensive, but also carries certain risks.

[0004] Model prediction can be mainly divided into three categories: prediction based on empirical data fitting methods, prediction based on data-driven methods, and prediction based on mechanistic models. Empirical fitting methods predict battery life by fitting empirical formulas to limited measured data. This method is relatively simple, but it requires a large amount of test data and has a large prediction error for long life cycles. For example, patents CN114624604A, CN115291131A, CN115407210A, CN113740752A, and CN113125982A are based on empirical fitting methods to predict battery life.

[0005] Data-driven methods collect a large amount of test or actual operation information, and use mathematical statistics to screen out characteristic factors that can be used to characterize battery life, making probabilistic predictions of battery life degradation. The accuracy of this prediction method is greatly affected by the amount of data collected and the data processing method chosen. For example, patents CN114675187A and CN110658460A are based on data-driven methods for life prediction. The first two methods rely on a large number of external battery state parameters (current, voltage, resistance, charge and discharge time) and product test data for life prediction and health status estimation. They obtain a large amount of test data through experiments, fit a regular formula, and predict the battery's health status and service life using an approximate probabilistic statistical method. These prediction methods directly use the number of cycles as the input parameter of the model, and the output parameter is the current capacity loss of the battery. However, they cannot know the internal aging of the battery. For products still in the design stage and without a finished product, these existing prediction methods are also difficult to effectively achieve data collection and life prediction.

[0006] A mechanistic model is a mathematical model describing an object or process based on its internal principles or mechanisms. Mechanism-based lithium-ion battery models take into account both the internal physicochemical processes of the battery and external operating conditions, and are generally simplified or extended from P2D models to meet different R&D needs. Patent CN109446619A discloses a method for optimizing the electrode design parameters of lithium-ion batteries based on electrochemical mechanisms. This method uses energy density and power density as evaluation indicators for battery optimization design, and optimizes the electrode design parameters (including electrode thickness, porosity, active material particle size, compaction density, and areal density, etc.) based on the model's electrochemical performance. However, this method is limited to estimating the battery's electrochemical performance and cannot simulate the battery aging and lifespan degradation process. More importantly, the battery's electrochemical performance changes with battery aging, and the battery aging trend and cycle life exhibit a non-linear coupling effect with the battery design parameters and operating conditions. These problems are difficult to solve with current patented technologies.

[0007] Given the current lack of methods to simultaneously study and predict battery aging processes and lifespan degradation at different temperatures from a mechanistic perspective, it is essential to develop a lifespan prediction method that fully considers both the internal mechanisms and external conditions of the battery, accurately and efficiently predicting battery lifespan under varying temperatures. By analyzing aging characteristics and key factors leading to battery degradation, this method can assist in the offline optimization design of product batteries, saving design costs, shortening the battery's entire lifespan design cycle, and improving battery product design efficiency. Summary of the Invention

[0008] To address the shortcomings of existing technologies, the present invention aims to provide a mechanism-based wide-temperature-range lithium-ion battery life prediction method that can predict the battery life cycle and life decay patterns under specific conditions, as well as the electrochemical performance and internal aging performance of the battery in any cycle process, with less measured data and limited input parameters.

[0009] The technical solution for achieving the objective of this invention is as follows:

[0010] A mechanism-based method for predicting the lifetime of wide-temperature-range lithium-ion batteries includes the following steps:

[0011] Step 1: Obtain battery product specifications and cell design parameters, and measure the intrinsic properties, electrochemical parameters, kinetic parameters, and thermodynamic parameters of the electrode materials;

[0012] Step 2: Obtain battery rate performance test data over a wide temperature range;

[0013] Step 3: Based on the mathematical equations, establish the electrochemical model and the thermal model respectively, build the electrochemical-thermal coupling model, and assign the parameters from Step 1 to the electrochemical-thermal coupling model;

[0014] Step 4: Apply initial conditions, boundary conditions, and operating conditions to the electrochemical-thermal coupling model, divide the computational domain into meshes, and set the solver algorithm;

[0015] Step 5: Obtain accelerated aging test data of the battery at different temperatures;

[0016] Step 6: Establish a side reaction model based on mathematical equations, build an electrochemical-thermal-side reaction coupling model, apply initial conditions, boundary conditions and operating conditions to the electrochemical-thermal-side reaction coupling model, divide the computational domain into grids, and set the solver algorithm;

[0017] Step 7: Validate the electrochemical-thermal-side reaction coupling model and correct the aging parameters based on the measured data to obtain the lithium-ion battery electrochemical-thermal-side reaction lifetime prediction model. Based on the capacity, health status and aging status calculated by the model, evaluate the battery life degradation.

[0018] Furthermore, the battery product specifications mentioned in step one include the battery's nominal capacity, upper and lower limit cutoff voltage, and cutoff current.

[0019] The cell design parameters include: the length and width of the positive electrode, the number of positive electrode sheets, the composition and proportion of the positive electrode material, the composition and proportion of the negative electrode material, the thickness of the positive and negative current collectors, the thickness of the separator, and the separator porosity.

[0020] The intrinsic characteristic parameters refer to the equilibrium potentials of the positive and negative electrode materials under different lithium insertion states.

[0021] The electrochemical parameters include: positive electrode particle radius, negative electrode particle radius, positive electrode material conductivity, negative electrode material conductivity, maximum lithium intercalation concentration of positive electrode material, and maximum lithium intercalation concentration of negative electrode material.

[0022] The kinetic parameters include: reaction rate constant, diffusion coefficient, reaction activation energy, and diffusion activation energy.

[0023] The thermodynamic parameters include: specific volume, entropy coefficient, and heat transfer coefficient.

[0024] Furthermore, the wide-temperature-range rate performance test described in step two involves conducting multi-rate charge-discharge tests on the battery at different temperatures. The physical quantities tested include voltage, current, temperature, and capacity.

[0025] Furthermore, the electrochemical model and the thermal model described in step three are bidirectionally coupled using the Arrhenius equation to describe their interdependence.

[0026] The parameters in the electrochemical model described in step three that are affected by temperature include: electrode solid-phase reaction rate constant, electrode solid-phase diffusion coefficient, electrolyte phase diffusion coefficient, electrolyte phase transport number, electrolyte phase activity dependence coefficient, and electrolyte phase conductivity.

[0027] The mathematical equations of the electrochemical model in step three include solid phase mass conservation, liquid phase mass conservation, solid phase charge conservation, and liquid phase charge conservation. The electrochemical reaction at the electrode / electrolyte interface is described by the Butler-Volmer equation.

[0028] The governing equation for the thermal model described in step three is the energy conservation equation. The thermal gradient between the heat generated and the heat dissipated by the battery causes fluctuations in the battery's internal temperature.

[0029] The heat sources in the thermal model include ohmic heat, electrochemical heat, polarization heat, and mixed heat, where electrochemical heat is reversible and polarization heat is irreversible.

[0030] In step three, the initial lithium intercalation state of the positive and negative electrodes is set according to the actual situation or requirements. Current or voltage boundary conditions are applied to the electrode scale direction, and concentration flux boundary conditions are applied to the electrode particle scale.

[0031] Step three includes, but is not limited to, using the cutoff voltage as the model termination condition.

[0032] Furthermore, the initial conditions in step four include the initial lithium intercalation concentration inside the positive and negative electrodes. Once the maximum lithium intercalation concentration of the electrodes is fixed, the initial conditions of the positive and negative electrodes are represented by the initial lithium intercalation state.

[0033] The boundary conditions include applying current / current density and voltage to the battery.

[0034] The operating conditions refer to the scaling factor and temperature conditions that match actual needs.

[0035] In step four, the calculation results of the electrochemical-thermal coupling model are compared and verified with the charge-discharge data, and the electrochemical-thermal model is calibrated.

[0036] Furthermore, the accelerated aging test method described in step five adopts the national standard battery cycle life test procedure, or the test procedure and test conditions are set by the user according to the actual operating environment and operating conditions of the designed battery.

[0037] The cycle life test procedure is as follows: constant current discharge - 30-minute rest - constant current charging - constant voltage charging - 30-minute rest, and so on for testing. The constant current is 1C, the discharge cutoff voltage is 2.75V, the constant current charging cutoff voltage is 4.2V, and the constant voltage charging cutoff current is C / 20. The above cycle life test method is only for illustrative purposes in this invention, and this invention is not bound by this test procedure.

[0038] Battery cycle aging tests require the measurement of characteristic quantities such as voltage, current, temperature, and charge / discharge capacity.

[0039] Furthermore, in step six, the side reaction model and the thermal model are coupled bidirectionally between temperature and side reactions through the Arrhenius equation;

[0040] The exchange current density of each side reaction in the side reaction model is constrained by the temperature fluctuations in the thermal model.

[0041] The products generated by each side reaction in the side reaction model constrain the heat changes in the thermal model through ohmic heat generation and polarization heat.

[0042] The output physical quantities of the side reaction model include battery capacity, health status, and surface film thickness, among which changes in surface film thickness can be used to assess battery aging.

[0043] The operating conditions refer to the rate conditions, temperature conditions, and cycle conditions that conform to actual needs.

[0044] The side reaction model describes the interfacial parasitic reactions using the cathode Tafel equation. The model also considers the impact of aging factors such as the growth of the solid electrolyte interphase (SEI) and lithium plating at the negative electrode interface on battery life. The SEI and lithium plating byproducts adhere to the surface of the negative electrode particles, leading to a reduction in the liquid phase volume fraction in the electrode and further affecting mass transport in the liquid phase.

[0045] The side reaction model considers the increase in interfacial polarization potential and impedance caused by reduction products adhering to the electrode / electrolyte interface, which in turn affects the ohmic heat and polarization heat of the thermal model. The side reaction model uses the Arrhenius equation to describe the effect of temperature on the exchange current density of each side reaction, thus achieving bidirectional coupling between the side reaction model and the thermal model.

[0046] The method of obtaining the parameters required for constructing the electrochemical-thermal-side reaction model in this invention is not strictly limited. Conventional testing methods in the field can be used, or the above parameters can be obtained through existing theoretical data or literature records.

[0047] The electrochemical model describes the temperature-dependent parameters using the Arrhenius equation: electrode solid-phase diffusion coefficient, reaction rate constant, electrolyte-phase diffusion coefficient, electrolyte-phase conductivity, electrolyte ion transport number, and electrolyte-phase activity dependence coefficient.

[0048] Furthermore, the computational domain grid described in steps four and six is ​​used to discretize the model in a structured or unstructured manner based on the dimensions and geometric configuration of the constructed model.

[0049] In the solvers described in steps four and six, the calculation process of the model is controlled by setting certain termination conditions (upper and lower voltage limits, 80% of nominal capacity). Specifically: in step four, the solver sets the battery voltage to the upper limit voltage or to the lower limit voltage as the termination condition for the charge-discharge calculation of the electrochemical-thermal coupling model; in step six, the solver sets the battery discharge capacity to drop to 80% of the nominal capacity as the termination condition for the calculation of the electrochemical-thermal-side reaction coupling model.

[0050] Furthermore, the physical quantities used in step seven to verify and evaluate the electrochemical-thermal-side reaction coupling model and measured data include battery capacity, health status, and battery aging status (i.e., surface film thickness).

[0051] Cyclic test data are compared with lifetime model calculations to calibrate the lifetime model, and inverse fitting is performed on the unknown parameters in the lifetime model. These unknown parameters include the molar mass and mass density of the by-reaction products, the ionic conductivity within the surface film, and the exchange current density and activation energy of the by-reactions.

[0052] Advantages and beneficial effects of the present invention:

[0053] This invention constructs an electrochemical-thermal model framework based on a P2D (Pseudo-Two-Dimensional Model). Side reactions within the battery simulate the aging process. The thermal model is coupled with the electrochemical and side reaction models via the Arrhenius equation. This allows for prediction of the battery's entire lifespan over a wide temperature range with minimal test data. The developed lifespan model only requires the battery's initial state and external environmental conditions as input to assess the State of Health (SOH) and cycle life of the aged battery throughout its entire lifespan. Compared to existing technologies, this method considers both the battery's internal physicochemical processes and external environmental variables, simulating the linear-to-nonlinear aging degradation of the battery. Furthermore, this predictive model can obtain more detailed aging data, efficiently meeting the needs of battery product optimization design. Attached Figure Description

[0054] Figure 1 This is a simplified geometric model of the battery cell.

[0055] Figure 2 The figures show the functional relationships between the equilibrium potential of the positive and negative electrodes and the lithium intercalation state, where (a) represents the functional relationship between the equilibrium potential of the graphite negative electrode and the lithium intercalation state, and (b) represents the functional relationship between the equilibrium potential of LiNi. 0.6 Co 0.1 Mn 0.3 The functional relationship between the equilibrium potential of the O2 cathode and the lithium intercalation state.

[0056] Figure 3 This represents the functional relationship between the entropy thermal coefficient of the negative electrode and the lithium intercalation state.

[0057] Figure 4 The expansion function is related to the lithium intercalation state of the graphite anode (the expansion function is used to describe the SEI regeneration reaction caused by the expansion and cracking of graphite particles in the anode).

[0058] Figure 5 The simulation results and measured data of the electrochemical performance of 1C discharge at 25℃ and 45℃ are compared. Among them, (a) shows the change of discharge voltage with discharge time, and (b) shows the change of temperature with discharge time.

[0059] Figure 6 This chart compares the simulated state of health (SOH) changes during battery cycling at 25℃ and 45℃ with measured data. The constant current condition was 1C, the discharge cutoff voltage was 2.75V, the charging cutoff potential was 4.3V, and the constant voltage charging cutoff current was C / 20. (a) compares the SOH obtained from the simulated battery cycling at 25℃ with measured data, and (b) compares the SOH obtained from the simulated battery cycling at 45℃ with measured data.

[0060] Figure 7The values ​​represent the battery discharge voltage and temperature changes predicted by the model for any cycle at 25℃ and 45℃. Among them, (a) and (b) represent the battery discharge voltage and temperature changes corresponding to the 700th, 1400th, and 2100th cycles at 25℃, and (c) and (d) represent the battery discharge voltage and temperature changes corresponding to the 700th, 1400th, and 1700th cycles predicted at 45℃.

[0061] Figure 8 The change in the thickness of the thin film (i.e., by-reaction products) on the negative electrode / separator side surface during the battery cycle aging process predicted by the life model, where (a) and (b) correspond to 25°C and 45°C, respectively. Detailed Implementation

[0062] To facilitate understanding of the present invention, the following embodiments are provided. Those skilled in the art should understand that these embodiments are merely illustrative and should not be construed as limiting the scope of the invention.

[0063] It should be noted that, without conflict, the embodiments and features in the embodiments of the present invention may be modified in form and specific quantity.

[0064] This invention constructs an electrochemical-thermal-side reaction lifetime prediction model, which includes a P2D-based electrochemical model, a thermal model, and a side reaction model. Ignoring inconsistencies and non-uniformities within the battery pack, and assuming uniform distribution of concentration, current density, and voltage along the electrode thickness direction, this example simplifies the research object to a one-dimensional cell model, such as... Figure 1 As shown.

[0065] This invention provides a mechanism-based method for predicting the lifetime of wide-temperature-range lithium-ion batteries, with the following specific steps:

[0066] Step 1: Obtain the battery product design parameters, cell geometry, and measure the physicochemical, electrochemical, kinetic, and thermodynamic parameters required for the model.

[0067] The battery in this invention example uses a 78Ah soft-pack lithium-ion battery (LiNi) manufactured by Tianjin Jiewei Power Industry Co., Ltd. 0.6 Co 0.1 Mn 0.3 (O2 / Graphite). The positive electrode in the battery pack measures 505mm × 95.5mm, and there are 23 positive electrode pieces.

[0068] In this example, the diaphragm thickness is 16 μm and the diaphragm porosity is 0.42. A copper current collector with a thickness of 6 μm is used on the negative electrode side, and an aluminum current collector with a thickness of 12 μm is used on the positive electrode side.

[0069] In this invention, the particle size of the positive and negative electrodes is measured using the D50 value obtained by a laser particle size analyzer.

[0070] The remaining parameters mentioned in step one are listed in Table 1.

[0071] In the battery samples used in this invention, both the electrodes and the separator are porous composite materials. Considering that uneven porosity distribution affects the diffusion process of lithium ions in the liquid phase, the liquid phase diffusion coefficient in the model is corrected using porosity and the Bruggeman coefficient.

[0072] D l，eff =ε Bruggeman D l

[0073] Among them, Dl ，eff Let ε be the effective diffusion coefficient of the electrolyte phase within the electrode, and ε be the porosity (i.e., the volume fraction of the electrolyte within the electrode). l is the ion diffusion coefficient in the pure electrolyte. In this example, the Bruggeman coefficients used in the positive electrode, negative electrode, and separator are 2, 1.5, and 2, respectively.

[0074] Table 1 Electrochemical Model Parameters

[0075] Electrode thickness / μm 77 56 Electrode particle size / μm 12.7 4 Porosity ε 0.28 0.21 <![CDATA[Solid-state electrical conductivity σ / S·m -1 > 460 16 <![CDATA[Maximum lithium intercalation concentration c s，max / mol·m -3 > 30507 47653 <![CDATA[Anodic charge transfer coefficient α a > 0.5 0.5 <![CDATA[Cathodic charge transfer coefficient α c > 0.5 0.5 <![CDATA[Reaction rate constant k / m·s -1 > 3.5E-11 2.0E-11 <![CDATA[Reaction activation energy / J·mol -1 > 40000 30000 <![CDATA[solid-phase diffusion coefficient D s / m 2 ·s -1 > 2.2E-14 1E-13 <![CDATA[Diffusion activation energy / J·mol -1 > 65000 45000

[0076] In this invention, an electrolyte is used with LiPF6 as the lithium salt and ethylene carbonate (EC), ethyl methyl carbonate (EMC), and diethyl carbonate (DEC) as solvents. The electrolyte concentration is 1.05 M. Transport parameters in the electrolyte, including diffusion coefficient, conductivity, activity dependence coefficient, and transport number, change with the electrolyte salt concentration. The relationship between transport parameters and electrolyte salt concentration can be expressed by the empirical formula in Table 2 (References: Nyman, A., Behm, M., et al. Electrochemical characterization and modelling of the mass transport phenomenon in LiPF6-EC-EMC electrolyte[J]. Electrochimica Acta, 2008, 53: 6356–6365. and Zavalis, TG, Behm, M., et al. Investigation of short-circuit scenarios in a lithium-ion battery cell[J]. Journal of The Electrochemical Society. 2012, 159: A848–A859).

[0077] Table 2 Electrolyte transport parameters

[0078]

[0079] Furthermore, coin cells with corresponding positive and negative electrode materials were assembled, and the equilibrium potentials under different lithium intercalation states were measured. The functional relationship is as follows: Figure 2 As shown.

[0080] Step two involves conducting charge-discharge performance tests on the test battery at 25℃ and 45℃ using a 1C rate. Typically, battery electrochemical performance is evaluated based on discharge capacity. In this example, the changes in battery voltage and temperature with capacity during discharge are selected as comparative calibration data for the electrochemical-thermal model. Figure 5 As shown.

[0081] Step 3: Based on design parameters, physicochemical properties, and mathematical equations, an electrochemical-thermal coupling model framework is constructed. The electrochemical model adopts the P2D model proposed by Newman et al., which is based on porous electrode theory and concentrated solution theory. This model includes two scales: electrode sheets and electrode particles.

[0082] The electrochemical model employs mass and charge conservation equations to describe the transport processes within the battery and electrode particles, while the Butler-Volmer equations describe the electrochemical reactions at the electrode / electrolyte interface. The governing equations are as follows:

[0083] Conservation of solid matter:

[0084] Solid-state charge conservation:

[0085] Liquid phase mass conservation:

[0086] Liquid phase charge conservation:

[0087] Butler-Volmer equations:

[0088] Interfacial reaction exchange current density:

[0089] Interfacial reaction overpotential: η = φ s -φ l -E eq -Δφ diff

[0090] Among them, c s The value of D represents the lithium-ion concentration within the electrode particles (solid phase), where t represents time. s It is the diffusion coefficient of lithium ions in the solid phase, where r represents the radial direction within the electrode particles.p Let i be the average radius of the electrode particles. s Let σ be the electrode solid-state current density, σ be the electrode solid-state conductivity, and φ be the electrode solid-state conductivity. s c is the solid-state potential of the electrode. l Electrolyte salt concentration, ε is electrode porosity, t + Z is the lithium-ion transference number. + v is the number of charges carried by lithium ions. + r is the number of particles dissociated from 1 mol of electrolyte. p j represents the radius of the electrode particles. n i is the pore wall flux at the electrode / electrolyte interface. l φ is the electrolyte phase current density. l κ is the electrolyte phase potential, and κ is the electrolyte conductivity. Here, R is the activity correlation coefficient, and R is the gas constant, taken as 8.314 J·kg⁻¹. -1 ·K -1 T is the battery temperature, and F is the Faraday constant, taken as 96485 C·mol⁻¹. -1 The exchange current density of the lithium insertion / extraction reaction at j0, α a and α c These are the anodic charge transfer coefficient and the cathodic charge transfer coefficient, respectively (both taken as 0.5 in this example), k is the reaction rate constant, and c... s，max E is the maximum lithium insertion / extraction concentration of the electrode particles, η is the overpotential of the lithium insertion / extraction reaction, and E is the maximum lithium insertion / extraction concentration of the electrode particles. eq It is the equilibrium potential of the lithium insertion / extraction reaction, Δφ diff It is the potential difference generated by lithium ions participating in the lithium insertion / extraction reaction through the thin film on the electrode surface.

[0091] In the electrochemical model, there are no interfacial electrochemical reactions within the membrane, i.e., j n =0.

[0092] The thermal model is based on the law of conservation of energy and considers the influence of the temperature difference between heat generation and heat dissipation within the battery on the battery's overall temperature. In this example, the heat sources in the thermal model include ohmic heat, polarization heat (also known as irreversible heat), and electrochemical heat (also known as reversible heat). The thermal model uses heat conduction to describe the battery's heat dissipation process.

[0093] The thermal model is represented as follows:

[0094] Energy conservation:

[0095] Heat production:

[0096] Heat dissipation:

[0097] Arrhenius empirical formula:

[0098] Where ρ is the battery mass density, C p The isobaric specific heat capacity of the battery (take 1010 J·kg) -1 ·K -1 ), q source Indicates a heat source i For activity (taken as 1 in this example), i i and η i Let represent the local current density and overpotential of the i-th side reaction, respectively. λ is the entropy-thermal coefficient, representing the relationship between equilibrium potential and temperature; λ is the thermal conductivity; and h is the convective heat transfer coefficient (taken as 10 W·m). -2 ·K -1 A is the cross-sectional area of ​​the battery (taken as the area of ​​the positive electrode), T ∞ For ambient temperature, T ref The reference temperature is 298.15K. The parameter value is at the reference temperature. E is the parameter at the corresponding temperature calculated using the Arrhenius formula. a This is the activation energy. The thermal model parameters in this example are shown in Table 3.

[0099] Table 3 Thermal Model Parameters

[0100] <![CDATA[Mass density ρ / kg·m -3 > 2240 1000 4600 8960 2700 <![CDATA[Thermal conductivity λ / W·m -1 ·K -1 > 1.2 0.6 0.6 400 238

[0101] This example considers the relationship between the entropic thermal coefficient of the negative electrode and the lithium intercalation state, such as... Figure 3 As shown.

[0102] In step four, the initial conditions of the electrochemical-thermal model are that the initial lithium intercalation states of the positive and negative electrodes are 0.23 and 0.90, respectively, the operating rate is 1C, the parameterized scanning calculation is performed at temperatures of 25℃ and 45℃, and the discharge cutoff voltage is set to 2.75V.

[0103] Step 5: Perform accelerated aging tests on the battery at 25℃ and 45℃, record the discharge capacity per cycle, and convert it into the battery's current state of health. In this invention, the state of health (SOH) is defined as the ratio of the battery's current maximum discharge capacity to its nominal capacity.

[0104] Step six: The reduction reactions corresponding to SEI and lithium plating are described by the cathode Tafel equation and integrated with the electrochemical-thermal coupling model in step four to build an electrochemical-thermal-side reaction coupling lifetime model.

[0105] The formation of the solid electrolyte interphase (SEI) layer and lithium plating in the negative electrode consumes the recyclable lithium content inside the battery and increases the internal impedance of the battery. Numerous studies have shown that SEI and lithium plating are the main causes of battery aging and lifespan degradation.

[0106] This example primarily considers the impact of SEI and lithium plating on battery life within the side reaction model. SEI encompasses both the SEI growth and SEI regeneration processes. After the battery undergoes formation, an initial SEI layer forms on the surface of the negative electrode particles. During cycling, the products generated from the oxidative decomposition of the electrolyte and its reduction reaction with the electrode active materials continuously promote the growth and thickening of the SEI layer—this is the SEI growth process. During cycling, the extraction and insertion of lithium ions cause cracks or even breakage on the surface of the electrode active particles. The exposed fresh active particles then generate a new SEI layer—this is the SEI regeneration process.

[0107] The lifetime model uses the cathode Tafel equation to describe the SEI growth process, and the equation is as follows:

[0108]

[0109] η SEI =φ s -φ l -E eq，SEI -Δφ diff

[0110] Where i SEI，formation It is the local current density during SEI growth, i0 ，SEI α represents the exchange current density of the SEI growth reaction. c，SE I is the charge transfer coefficient for SEI growth, η SEI E represents the overpotential for SEI growth. eq，SEI This represents the equilibrium potential for SEI growth.

[0111] The SEI regeneration process at the negative electrode under different lithium intercalation states is described by the expansion function. Therefore, the cathode Tafel equation modified by the expansion function can be used to describe the SEI regeneration process.

[0112]

[0113] Among them, i SEI，reformation f is the local current density of SEI regeneration. expansion For the expansion function (e.g.) Figure 4 (As shown).

[0114] The lithium plating process in the side reaction model is also described using the cathode Tafel equation:

[0115]

[0116] η lpl =φ s -φ l -E eq，lpl -Δφ diff

[0117] Among them, i lpl i represents the local current density of the lithium plating reaction. 0，lpl η represents the exchange current density of the lithium plating reaction. lpl This represents the overpotential of the lithium plating reaction.

[0118] In the lifetime model, by-reaction products adhere to the surface of electrode particles and form a surface film. During cycling, lithium ions repeatedly intercalate and deintercalate between the electrolyte and electrode particles. When they pass through the surface film, they encounter some resistance, thus generating a polarization potential Δφ at the film. diff That is: Δφ diff =i*R film Where i is the current density of the lithium insertion / extraction reaction, and R... film The sheet resistance of the thin film on the negative electrode surface.

[0119] In this example, the thin film resistance is directly proportional to the thin film thickness and inversely proportional to the ionic conductivity within the thin film.

[0120]

[0121] In the formula, δ0 is the initial thickness of the thin film formed on the surface of the electrode particles after the formation process, and σ film The ionic conductivity is the conductivity within the surface thin film.

[0122] In the model, the surface film consists of an SEI growth layer, an SEI regeneration layer, and a lithium plating layer. Therefore, the thickness change Δδ of the surface film is the sum of the changes in the three film layers.

[0123]

[0124] In the formula, the subscript j represents the three side reactions of SEI growth, SEI regeneration and lithium plating, M represents the molar mass of each side reaction product, ρ represents the mass density of each side reaction product, and c represents the concentration of each side reaction product. This represents the sum of changes in the volume fraction of each byproduct.

[0125] In the lifetime model, the reaction rate of the side reaction changes with temperature. It is assumed that the exchange current density of the side reaction and temperature satisfy a relationship that can be described by the Arrhenius formula, thereby achieving bidirectional coupling between the side reaction model and the thermal model.

[0126] The parameters in the side reaction model are shown in Table 4.

[0127] Table 4. Side reaction model parameters

[0128] <![CDATA[SEI exchange current density i 0，SEI / A·m -2 > 1.8E-7 <![CDATA[SEI equilibrium potential E eq，SEI / V]]> 0.4 <![CDATA[SEI molar mass M SEI / g·mol -1 > 26 <![CDATA[SEI density ρ SEI / kg·m -3 > 2640 <![CDATA[SEI reaction activation energy E a，SEI / J·mol -1 > 30000 <![CDATA[Lithium plating exchange current density i 0，lpl / A·m -2 > 3.7E-3 <![CDATA[Lithium plating equilibrium potential E eq，lpl / V]]> 0 <![CDATA[Lithium metal molar mass M Li / g·mol -1 > 6.94 <![CDATA[lithium metal density ρ Li / kg·m -3 > 534 <![CDATA[lithium precipitation activation energy E a，Li / J·mol -1 > 50000 <![CDATA[Initial thin film thickness δ0 / nm]]> 5 <![CDATA[Ionic conductivity σ in the surface thin film layer film / S·m -1 > 3.8E-6

[0129] Step 6: In the lifetime model, the initial lithium intercalation states of the positive and negative electrodes are set to 0.23 and 0.90, respectively, and the constant current charge-discharge rate is 1C. The cycle life and health status of the battery at ambient temperatures of 25°C and 45°C are calculated respectively.

[0130] In step seven, the simulated battery health status at different cycle numbers is compared with the measured data, as follows: Figure 6 As shown, the aging parameters and activation energies of the side reactions are obtained by inverse fitting of the side reaction model.

[0131] Preferably, the electrochemical-thermal-side reaction lifetime model finally built in step seven predicts and extracts electrochemical performance data (such as...) in any cycle. Figure 7 (as shown) and changes in aging performance (e.g.) Figure 8 As shown in the figure, this provides a reference for the life assessment and optimization design of battery products.

[0132] Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0133] It should be noted that the terminology used in this application is only for describing specific implementations and is not intended to limit the exemplary embodiments according to this application. The method described in this invention relies on an electrochemical-thermal-side-reaction coupled lifetime model that integrates the nonlinear relationships between electrochemistry, side reactions, and heat. It can predict the health status and cycle life of batteries under wide temperature range conditions, and can predict the lifetime degradation data and battery aging degree for any number of cycles with less experimental data. It can select the optimal design for different battery product requirements, improve product design efficiency, shorten the battery's entire life cycle design cycle, and save design costs.

[0134] The applicant declares that the present invention is illustrated through the above embodiments, but the present invention is not limited to the above detailed methods, that is, it does not mean that the present invention must rely on the above detailed methods to be implemented. Those skilled in the art should understand that any improvements to the present invention, any substitution of any product in the present invention, simplification and extension of the model equations, etc., all fall within the protection scope and disclosure scope of the present invention.

Claims

1. A mechanism-based method for predicting the lifetime of a wide-temperature-range lithium-ion battery, characterized in that, Includes the following steps: Step 1: Obtain battery product specifications and cell design parameters, and measure the intrinsic properties, electrochemical parameters, kinetic parameters, and thermodynamic parameters of the electrode materials; Step 2: Obtain rate performance test data of the battery at different temperatures; Step 3: Based on the mathematical equations, establish the electrochemical model and the thermal model respectively, build an electrochemical-thermal coupling model with bidirectional coupling effect, and assign the parameters from Step 1 to the electrochemical-thermal coupling model; Step 4: Apply initial conditions, boundary conditions, and operating conditions to the electrochemical-thermal coupling model, divide the computational domain into meshes, and set the solver algorithm; Step 5: Obtain accelerated aging test data of the battery at different temperatures; Step 6: Establish a side reaction model based on mathematical equations, build an electrochemical-thermal-side reaction coupling model, apply initial conditions, boundary conditions and operating conditions to the electrochemical-thermal-side reaction coupling model, divide the computational domain into grids, and set the solver algorithm; The side reaction model and the thermal model are bidirectionally coupled through the Arrhenius equation to achieve temperature and side reaction; The exchange current density of each side reaction in the side reaction model is constrained by the temperature fluctuations in the thermal model. The products generated by each side reaction in the side reaction model constrain the heat changes in the thermal model through ohmic heat generation and polarization heat. The output physical quantities of the side reaction model include battery capacity, health status, and surface film thickness. The operating conditions refer to the rate conditions, temperature conditions, and cycle conditions that conform to actual needs. In the side reaction model, the interfacial parasitic reaction is described by the cathode Tafel equation; The SEI growth process is described by the cathode Tafel equation, which is as follows: in It is the local current density during SEI growth. The exchange current density for the SEI growth reaction. is the charge transfer coefficient for SEI growth. F It is Faraday's constant. R The gas constant is... T For battery temperature, This represents the overpotential for SEI growth. φ s This is the solid-state potential of the electrode. φ l This refers to the electrolyte phase potential. This represents the equilibrium potential for SEI growth. φ diff The potential difference generated by lithium ions participating in the lithium insertion / extraction reaction through the thin film on the electrode surface; The SEI regeneration process at the negative electrode under different lithium intercalation states is described by an expansion function, and the SEI regeneration process at the cathode is described by the Tafel equation modified by the expansion function. in, For the local current density of SEI regeneration, It is the expansion function; Step 7: Validate the electrochemical-thermal-side reaction coupling model and correct the aging parameters based on the measured data to obtain the lithium-ion battery electrochemical-thermal-side reaction lifetime prediction model, and evaluate the battery capacity, health status and aging condition based on the model calculation results.

2. The method for predicting the lifespan of a wide-temperature-range lithium-ion battery according to claim 1, characterized in that, The battery product specifications mentioned in step one include the battery's nominal capacity, upper and lower cutoff voltages, and cutoff current. The cell design parameters include: the length and width of the positive electrode, the number of positive electrode sheets, the composition and proportion of the positive electrode material, the composition and proportion of the negative electrode material, the thickness of the positive and negative current collectors, the thickness of the separator, and the separator porosity. The intrinsic characteristic parameters refer to the equilibrium potentials of the positive and negative electrode materials under different lithium intercalation states; The electrochemical parameters include: positive electrode particle radius, negative electrode particle radius, positive electrode material conductivity, negative electrode material conductivity, maximum lithium intercalation concentration of positive electrode material, and maximum lithium intercalation concentration of negative electrode material. The kinetic parameters include: reaction rate constant, diffusion coefficient, reaction activation energy, and diffusion activation energy; The thermodynamic parameters include: specific volume, entropy coefficient, and heat transfer coefficient.

3. The method for predicting the lifespan of a wide-temperature-range lithium-ion battery according to claim 1, characterized in that, The wide-temperature-range rate performance test mentioned in step two involves conducting multi-rate charge and discharge tests on the battery at different temperatures. The physical quantities that need to be measured in the rate performance test include voltage, current, temperature, and capacity.

4. The method for predicting the lifespan of a wide-temperature-range lithium-ion battery according to claim 1, characterized in that, In step three, the temperature change of the battery in the thermal model is correlated with the electrochemical model parameters through the Arrhenius formula. The temperature-affected parameters in the electrochemical model include: electrode solid-phase reaction rate constant, electrode solid-phase ion diffusion coefficient, electrolyte phase ion diffusion coefficient, electrolyte phase ion transport number, electrolyte phase activity dependence coefficient, and electrolyte phase conductivity.

5. The method for predicting the lifespan of a wide-temperature-range lithium-ion battery according to claim 1, characterized in that, The initial conditions in step four include the initial lithium intercalation concentration inside the positive and negative electrodes. Once the maximum lithium intercalation concentration of the electrodes is fixed, the initial conditions of the positive and negative electrodes are represented by the initial lithium intercalation state. The boundary conditions include the applied current / current density and voltage to the battery; The operating conditions refer to the scaling factor and temperature conditions that match actual needs.

6. The method for predicting the lifespan of a wide-temperature-range lithium-ion battery according to claim 1, wherein the accelerated aging test method in step five adopts the national standard battery cycle life test procedure, or the test procedure and test conditions are set by the user according to the actual operating conditions of the designed battery product.

7. The method for predicting the lifespan of a wide-temperature-range lithium-ion battery according to claim 1, characterized in that, The computational domain grid described in steps four and six is ​​used to discretize the model using either a structured or unstructured method, based on the dimensions and geometric configuration of the constructed model.

8. The method for predicting the lifetime of a wide-temperature-range lithium-ion battery according to claim 1, characterized in that, In step four, the solver sets the battery voltage to either the upper limit voltage or the lower limit voltage as the termination condition for the charge-discharge calculation of the electrochemical-thermal coupling model; in step six, the solver sets the battery discharge capacity to drop to 80% of the nominal capacity as the termination condition for the calculation of the electrochemical-thermal-side reaction coupling model.

9. In the wide-temperature-range lithium-ion battery life prediction method according to claim 1, the physical quantities verified by the electrochemical-thermal-side reaction coupling model and measured data in step seven include battery capacity, health status, and aging status.

Images