Method for identifying thermal runaway reaction kinetic parameters of high-nickel ternary battery
By using DSC testing and particle swarm optimization algorithms, the problem of fitting multiple exothermic peaks in high-nickel ternary batteries was solved, enabling efficient identification of kinetic parameters and reducing the risk of thermal runaway.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BEIJING JIAOTONG UNIV
- Filing Date
- 2026-01-14
- Publication Date
- 2026-05-05
AI Technical Summary
Existing technologies have slow convergence speed and are prone to getting trapped in local optima when fitting the DSC heat flow curves of multiple exothermic peaks in high-nickel ternary batteries. They are also difficult to accurately identify kinetic parameters, resulting in poor fitting performance of thermal runaway models.
The heat flow curves of single-component and multi-component battery samples were obtained by DSC testing. The activation energy and forward factor were identified by combining the Kissinger equation. The particle swarm optimization algorithm was applied to fit parameters such as reaction order, reaction order and enthalpy to decouple the thermal contribution of thermal runaway reaction.
It improves the convergence speed and accuracy of kinetic parameter identification, effectively decouples the thermal contributions between battery materials, and reduces the risk of thermal runaway in high-nickel ternary batteries.
Smart Images
Figure CN121983157A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of lithium-ion battery safety, specifically relating to a method for identifying the kinetic parameters of thermal runaway reaction in high-nickel ternary batteries based on DSC multi-exothermic peak heat flux curves. Background Technology
[0002] High-nickel ternary (NCM) cathode materials are widely used in electric vehicles and energy storage batteries due to their excellent electrochemical performance and long cycle life. However, while effectively improving battery energy density, high-nickel ternary cathode materials also lead to decreased battery thermal stability and increased risk of thermal runaway. Therefore, studying the thermal runaway mechanism of high-nickel ternary batteries and establishing accurate mathematical models to predict battery thermal runaway behavior are crucial foundations for advancing their safety design.
[0003] Existing research indicates that the thermal runaway process of lithium-ion batteries typically involves multiple consecutive exothermic stages. Typical reaction pathways include: SEI film decomposition, side reactions between the negative electrode and electrolyte, separator melting and structural damage, thermal decomposition of the positive electrode material, electrolyte decomposition reaction, binder pyrolysis, and electrolyte combustion. Chemical kinetic analysis based on the Arrhenius equation can obtain the kinetic parameters of these reactions (such as activation energy and frequency factor), thereby establishing a thermal runaway model to predict the battery's thermal runaway behavior. Current research methods involve DSC (Digital Subtraction Angiography) of battery components to obtain heat flow curves at different temperature rise rates, establishing a kinetic model, and combining the Kissinger method and genetic algorithms to identify kinetic parameters. However, the DSC heat flow curves of high-nickel ternary cathode materials often exhibit multiple overlapping exothermic peaks. Genetic algorithms suffer from slow convergence and a tendency to get trapped in local optima when fitting DSC heat flow curves with multiple exothermic peaks, resulting in poor fitting performance. Summary of the Invention
[0004] To address the shortcomings of existing technical solutions, the present invention aims to provide a method for identifying the kinetic parameters of thermal runaway reaction in high-nickel ternary batteries based on DSC multi-exothermic peak heat flow curves. This method solves the problems of slow convergence speed and easy getting trapped in local optima when traditional genetic algorithms fit multi-exothermic peak DSC heat flow curves of high-nickel ternary batteries. The method specifically includes the following steps: S1. Disassemble the battery in the glove box, obtain the positive and negative electrode active materials of the battery, prepare single-component and multi-component samples of the battery for DSC testing, and obtain the heat flow curves of the battery materials at different temperature rise rates. S2. Establish a chemical reaction kinetic model and apply the Kissinger equation to identify the activation energy. and forward factor Kinetic parameters of multiple exothermic peaks in two heat source samples; S3. Applying particle swarm optimization algorithm to fit and identify the reaction order.m Reaction order n and enthalpy Other kinetic parameters, such as the thermal contribution of multiple exothermic reactions in thermal runaway, are quantitatively decomposed by fitting the heat flow curves of the main heat sources.
[0005] Furthermore, the specific steps of step S1 are as follows: S11. Disassemble the fully charged battery in the glove box, obtain the positive and negative electrode materials and separator of the battery, and grind them. S12. Prepare three single-component crucible samples ("negative electrode active material", "positive electrode active material", and "electrolyte") and four multi-component crucible samples ("negative electrode active material + electrolyte", "positive electrode active material + electrolyte", "negative electrode active material + positive electrode active material", and "negative electrode active material + positive electrode active material + electrolyte") in a glove box. The proportions of each component in the multi-component samples are equal to the mass ratio of the galvanic cell. Place the seven samples in... Heat flux curves were obtained by DSC testing at the heating rate. Samples with ≥4 exothermic peaks and calorific value ranking among the top two of all samples were selected as multi-peak primary heat source samples. DSC tests were performed at different heating rates to obtain heat flux curves at different heating rates.
[0006] Furthermore, the specific steps of step S2 are as follows: S21. Establish chemical reaction kinetics and heat generation models to describe the exothermic behavior of battery materials, i.e. ; ; ; ; In the formula, For chemical reaction rate, The normalized concentration of reactant x Forward factor, For activation energy, The molar gas constant, Thermodynamic temperature It is a concentration function. The reaction order is... The reaction is exothermic. Mass of reactants For reaction enthalpy; S22. Based on the DSC multi-peak heat flux curves of the main heat sources at different temperature rise rates, the activation energies of multiple reactions are identified using the Kissinger equation. and forward factor A, i.e. ; In the formula, The heating rate for DSC testing. Let be the peak temperature of the exothermic peak, and u be the number of DSC tests performed at different heating rates, derived from... The activation energy of the exothermic reaction can be calculated from the slope of the fitted straight line. ,Depend on The forward factor is obtained by calculating the intercept of the fitted line. A .
[0007] Furthermore, step S3 applies a particle swarm optimization algorithm to the reaction order of other kinetic parameters. m Reaction order n and enthalpy Optimization fitting is performed, and the fitness function is the root mean square error of the DSC heat flux curves under different temperature rise rates, i.e. ; Preferably, high-nickel ternary batteries refer to ternary batteries with a nickel content of more than 50%, such as NCM532, NCM622, and NCM811.
[0008] Compared with the prior art, the advantages of this invention are: 1. The method can effectively identify the thermal runaway reaction kinetic parameters of high-nickel ternary batteries, providing a parameter basis for thermal runaway modeling of high-nickel ternary batteries; 2. The optimization algorithm converges faster than the traditional genetic algorithm, and while being simple to implement, it also has strong global search capabilities; 3. It achieves the decoupling of thermal contributions between different reactions of battery materials during thermal runaway, which can help improve the research on high heat-generating reactions between materials and reduce the risk of thermal runaway in high-nickel ternary batteries. Attached Figure Description
[0009] Figure 1 A flowchart of a method for identifying the kinetic parameters of thermal runaway reaction in a high-nickel ternary battery based on DSC multi-exothermic peak heat flux curves, as described in this invention; Figure 2 Seven material components of lithium-ion batteries were sampled. Heat flux curves at heating rates; Figure 3 For lithium-ion battery samples "negative electrode + electrolyte" and "positive electrode + negative electrode", respectively... Heat flux curves at heating rates; Figure 4 To fit the Kissinger method The result; Figure 5 for Fitting results of particle swarm optimization algorithm for heat flux curves of "negative electrode + electrolyte" sample at heating rate; Figure 6 for Fitting results of particle swarm optimization algorithm for heat flux curves of "positive electrode + negative electrode" samples under heating rate; Figure 7 To investigate the "negative electrode + electrolyte" of the sample Decoupling results of the thermal contribution of chemical reactions under heating rate; Figure 8 To test the "positive electrode + negative electrode" of the sample Decoupling results of the thermal contribution of chemical reactions at different heating rates. Detailed Implementation
[0010] To make the objectives, advantages, and features of the present invention more apparent, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be emphasized that the above drawings and the following description are merely exemplary and not intended to limit the scope of the present invention or its application.
[0011] like Figure 1 As shown, a method for identifying the kinetic parameters of thermal runaway reaction in high-nickel ternary batteries based on DSC multi-exothermic peak heat flux curves includes the following steps: S1. Disassemble the battery in a glove box, obtain the positive and negative electrode active materials, prepare single-component and multi-component battery samples, and perform DSC tests to obtain the heat flow curves of the battery materials at different temperature rise rates.
[0012] S11. Disassemble the fully charged battery in the glove box, obtain the positive and negative electrode materials of the battery, and grind them.
[0013] S12. Prepare three single-component crucible samples ("negative electrode active material", "positive electrode active material", and "electrolyte") and four multi-component crucible samples ("negative electrode active material + electrolyte", "positive electrode active material + electrolyte", "negative electrode active material + positive electrode active material", and "negative electrode active material + positive electrode active material + electrolyte") in a glove box. The proportions of each component in the multi-component samples are equal to the mass ratio of the galvanic cell. Place the seven samples in... Heat flux profiles were obtained by performing DSC tests at different heating rates. Additionally, samples from the main heat sources with multiple peaks were analyzed separately. DSC tests were performed at different heating rates to obtain heat flow curves.
[0014] S2. Establish a reaction kinetic model and apply the Kissinger equation to identify the kinetic parameters of the multiple exothermic peaks in the main heat source sample: activation energy. and forward factor .
[0015] S21. Establish a chemical reaction kinetic model (Arrhenius equation) and a heat generation model to describe the exothermic behavior of battery materials, i.e. ; ; ; ; In the formula, For chemical reaction rate, The normalized concentration of reactant x Forward factor, For activation energy, The reaction order is... Mass of reactants This is the enthalpy of the reaction.
[0016] S22. Based on the DSC multi-peak heat flux curves of the main heat sources at different temperature rise rates, the activation energies of multiple reactions are identified using the Kissinger equation. and forward factor ,Right now ; In the formula, The heating rate for DSC testing. Let be the peak temperature of the exothermic reaction, and u be the number of DSC tests performed at different heating rates. Therefore, the activation energy of the exothermic reaction... can be The slope of the fitted line is calculated, and the forward factor can be obtained from... The intercept of the fitted line in the image is calculated.
[0017] S3. Apply particle swarm optimization algorithm to fit and identify other kinetic parameters: reaction order. Reaction order and enthalpy This decouples the thermal contribution of multiple exothermic reactions in the main heat source sample of thermal runaway.
[0018] The following specific embodiment uses a commercial 50Ah NCM622 / graphite pouch cell as an example for illustration. This method is also applicable to the identification of kinetic parameters of other high-nickel ternary batteries with multiple exothermic peak DSC heat flow curves.
[0019] (1) First, the battery was cycled three times using a charge-discharge device to determine its capacity, and batteries with good consistency were selected for backup. The fully charged battery was disassembled in a glove box to obtain the positive and negative electrode materials and grind them. Three single-component crucible samples and four multi-component crucible samples were prepared as shown in Table 1. The mass ratio of each component in the multi-component crucible sample was the same as the proportion of each component in the original battery. The DSC test was set to the temperature rise rate. The temperature range is 50°C to 500°C, and the gas flow rate is... In order to identify the dynamic parameters, subsequent steps were taken... and DSC tests were performed on the negative electrode active material + electrolyte (negative electrode + electrolyte) and the positive electrode active material + negative electrode active material (positive electrode + negative electrode) of the sample at the following four heating rates. These two groups of samples are the two main heat sources in the battery thermal runaway process.
[0020] To eliminate random errors, two sets of all crucible samples were prepared for testing, with the material ratios and experimental settings being exactly the same.
[0021] Table 1 DSC Test Samples ; (2) The DSC test results of the seven groups of samples showed good consistency between the two groups. Figure 2 Table 2 shows the calorific value of each sample as one typical result. Figure 2 The heat flux and calorific value in Table 2 were normalized using the mass of the negative electrode active material, which was 1.6 mg.
[0022] Table 2 Calorific value of the sample at the heating rate ;
[0023] The heating behavior of the "positive electrode + negative electrode + electrolyte" sample before 200°C is similar to that of the "negative electrode + electrolyte" sample. However, after 200°C, the positions of its three exothermic peaks are close to those of the "positive electrode + negative electrode" sample. This indicates that before 200°C, the battery's heating mainly originates from the "negative electrode-electrolyte reaction," while after 200°C, the heating source is the "positive electrode-negative electrode reaction." Furthermore, the heat generation of the "negative electrode + electrolyte" and "positive electrode + negative electrode" samples is the highest among all samples, with [specific values missing]. and This indicates that the "negative electrode-electrolyte reaction" and the "positive electrode-negative electrode reaction" are the main heat sources for battery thermal runaway. The five peaks in the "negative electrode + electrolyte" sample are numbered R1~R5, corresponding to reactions Q1~Q5, and the four peaks in the "positive electrode + negative electrode" sample are numbered R6~R9, corresponding to reactions Q6~Q9.
[0024] (3) Establish chemical reaction kinetic models for the exothermic reactions in the "negative electrode + electrolyte" and "positive electrode + negative electrode" samples and identify kinetic parameters to provide a data basis for thermal runaway modeling. The heat generation of chemical reactions is often described by chemical reaction kinetic equations (Arrhenius equations), i.e. ; ; ; ; In the formula, For chemical reaction rate, The normalized concentration of reactant x Forward factor, For activation energy, The reaction order is... Mass of reactants This is the enthalpy of the reaction.
[0025] Each exothermic reaction in the kinetic equation requires Five kinetic parameters describe the process. It is difficult to obtain all the kinetic parameters of a reaction in a multi-reaction coupled DSC curve. Among the five parameters, the forward factor and activation energy can be determined using the Kissinger method. DSC tests at different heating rates show that the peak temperature of the exothermic peak and the heating rate satisfy the Kissinger equation, i.e. ; In the formula, The heating rate for DSC testing. Let be the peak temperature of the exothermic reaction, and u be the number of DSC tests performed at different heating rates. Therefore, the activation energy of the exothermic reaction... can be The slope of the fitted line is calculated, and the forward factor can be obtained from... The intercept of the fitted line in the image is calculated.
[0026] Other kinetic parameters: reaction order Reaction order and enthalpy Other nonlinear optimization algorithms were used. Considering the large number of parameters in this application, Particle Swarm Optimization (PSO) is simpler to implement than other optimization algorithms while possessing strong global search capabilities. The fitness of the generated solution is evaluated using the root mean square error formula, i.e. ; (4) Figure 3The DSC test results of the "negative electrode + electrolyte" and "positive electrode + negative electrode" samples at different temperature rise rates are shown in Table 3. The peak temperatures of the exothermic peaks in the four DSC curves are listed in Table 3.
[0027] Table 3. Peak temperatures of the exothermic peaks for samples "negative electrode + electrolyte" and "positive electrode + negative electrode" at four heating rates. Peak temperature of the exothermic peak ;
[0028] Among them, the "positive electrode + negative electrode" sample is At a heating rate of , the exothermic peak R7 exhibits a significant thermal hysteresis effect, causing a significant shift in the position and shape of the exothermic peaks R7, R8, and R9. Therefore, this curve is only used for identifying the parameters of reaction Q6. Figure 4 Showing about From the scatter plot and the best linear fit line, we can see the exothermic peaks R1 to R9. and The fit exhibits a good linear relationship at different heating rates, and the forward factors of the nine reactions can be calculated from the slope and intercept of the fitted line. and activation energy .
[0029] Table 4. Dynamic parameters identified by Kissinger method and particle swarm optimization algorithm ; Table 4 lists the identification results of all parameters under the Kissinger method and the particle swarm optimization algorithm. Figure 5 and Figure 6 The DSC test results of the "negative electrode + electrolyte" and "positive electrode + negative electrode" samples under different heating rates simulated by the model are shown and compared with the actual experimental results. The curves of different heating rates intersect each other on the vertical axis. In order to observe and compare, Figure 7 and Figure 8 Showing based on The DSC test results of the two sets of samples at different heating rates were decoupled from the thermal contribution of each reaction. It can be seen that the simulated DSC curves fit the experimental results well at different heating rates. The root mean square errors (RMSE) of the two sets of samples are 0.1075 and 0.3001, respectively. This indicates that the model and the identified parameters can well reflect the exothermic mechanism of the two sets of samples, and can effectively quantify and decouple the heat generation of all exothermic reactions.
[0030] The above embodiments have provided a detailed description of the technical solution of the present invention. Obviously, the present invention is not limited to the described embodiments. Based on the embodiments of the present invention, those skilled in the art can make various modifications, but any modifications that are equivalent to or similar to the present invention fall within the scope of protection of the present invention.
[0031] The contents not described in detail in this specification are existing technologies known to those skilled in the art.
Claims
1. A method for identifying the kinetic parameters of thermal runaway reaction in a high-nickel ternary battery, characterized in that, Includes the following steps: S1. Disassemble the battery in the glove box, obtain the positive and negative electrode active materials of the battery, prepare single-component and multi-component samples of the battery for DSC testing, and obtain the heat flow curves of the battery materials at different temperature rise rates. S2. Establish a chemical reaction kinetic model and apply the Kissinger equation to identify the activation energy. and forward factor Kinetic parameters of multiple exothermic peaks in two heat source samples; S3. Applying particle swarm optimization algorithm to fit and identify the reaction order. m Reaction order n and enthalpy Other kinetic parameters, such as the thermal contribution of multiple exothermic reactions in thermal runaway, are quantitatively decomposed by fitting the heat flow curves of the main heat sources.
2. The method for identifying the kinetic parameters of thermal runaway reaction in a high-nickel ternary battery according to claim 1, characterized in that, The specific steps of step S1 are as follows: S11. Disassemble the fully charged battery in the glove box, obtain the positive and negative electrode materials and separator of the battery, and grind them. S12. Prepare three single-component crucible samples ("negative electrode active material", "positive electrode active material", and "electrolyte") and four multi-component crucible samples ("negative electrode active material + electrolyte", "positive electrode active material + electrolyte", "negative electrode active material + positive electrode active material", and "negative electrode active material + positive electrode active material + electrolyte") in a glove box. The proportions of each component in the multi-component samples are the same as the mass ratio of the galvanic cell. Place the seven samples in... Heat flux curves were obtained by DSC testing at the heating rate. Samples with ≥4 exothermic peaks and calorific value ranking among the top two of all samples were selected as multi-peak primary heat source samples. DSC tests were performed at different heating rates to obtain heat flux curves at different heating rates.
3. The method for identifying the kinetic parameters of thermal runaway reaction in a high-nickel ternary battery according to claim 1, characterized in that, The specific steps of step S2 are as follows: S21. Establish chemical reaction kinetics and heat generation models to describe the exothermic behavior of battery materials, i.e. ; ; ; ; In the formula, For chemical reaction rate, The normalized concentration of reactant x Forward factor, For activation energy, The molar gas constant, Thermodynamic temperature It is a concentration function. The reaction order is... The reaction is exothermic. Mass of reactants For reaction enthalpy; S22. Based on the DSC multi-peak heat flux curves of the main heat sources at different temperature rise rates, the activation energies of multiple reactions are identified using the Kissinger equation. and forward factor A, i.e. ; In the formula, The heating rate for DSC testing. Let be the peak temperature of the exothermic peak, and u be the number of DSC tests performed at different heating rates, derived from... The activation energy of the exothermic reaction can be calculated from the slope of the fitted straight line. ,Depend on The forward factor is obtained by calculating the intercept of the fitted line. A .
4. The method for identifying the kinetic parameters of thermal runaway reaction in a high-nickel ternary battery according to claim 1, characterized in that, Step S3 applies the particle swarm optimization algorithm to the reaction order of other kinetic parameters. m Reaction order n and enthalpy Optimization fitting is performed, and the fitness function is the root mean square error of the DSC heat flux curves under different temperature rise rates, i.e. 。 5. The method for identifying the kinetic parameters of thermal runaway reaction in a high-nickel ternary battery according to claim 1, characterized in that, High-nickel ternary batteries refer to ternary batteries such as NCM532, NCM622, and NCM811, which have a nickel content of more than 50%.