A method for predicting the service life of lithium-ion batteries based on electrochemical model

Abstract

The present invention discloses a method for predicting the capacity decay of a lithium-ion battery based on the coupling of chemical degradation and mechanical degradation, and relates to the technical field of lithium-ion battery life prediction. The method tests the composition and thickness of the solid electrolyte interface of the negative electrode plate with different cycle times, tests the length and depth of the surface cracks of the negative electrode material, and then predicts the use capacity of the lithium-ion battery through an electrochemical model and a mathematical algorithm, including analyzing and predicting the formation and growth of the solid electrolyte interface on the surface of the negative electrode material, and predicting the life of the lithium-ion battery in stages throughout its life cycle.

Description

Technical Field

[0001] The present invention belongs to the technical field of lithium ion battery life prediction, and in particular relates to a method for predicting the service life of a lithium ion battery based on an electrochemical model. Background Art

[0002] Lithium-ion batteries are considered the most promising alternative and green energy storage technology for their light weight, non-toxicity, and use in portable electronic devices, hybrid electric vehicles, plug-in hybrid electric vehicles, and other power facilities, with high storage capacity and energy density compared to conventional batteries.

[0003] However, lithium-ion batteries will be affected by many external factors in actual operation and often fail to achieve the lifespan we expect. However, premature replacement will cause economic losses, and failure to replace in time will have a great impact on system stability. Therefore, if you accurately understand the battery life status, you will undoubtedly be able to find a balance between economy and reliability.

[0004] At present, the common method to determine the cycle life of a battery is to perform a charge and discharge cycle experiment on the battery. A major disadvantage of this method is that the test time is relatively long, sometimes even taking several months, consuming huge manpower and material resources and being inefficient. Therefore, it is very important to predict the battery life in a shorter period of time.

[0005] There is a strong demand for cycle life prediction of commercial lithium-ion batteries in many applications, including life prediction of electric vehicle batteries and life prediction of energy storage system cells. Currently, the methods for battery cycle life prediction are mainly divided into two categories: mechanism-based methods and computer technology-based methods. The mechanism-based method mainly analyzes and explores the failure mechanism of lithium iron phosphate power batteries during the cycle process through constant current cycles, operating cycle and storage experiments, but the current evaluation method system for cycle life of lithium iron phosphate batteries of different systems is not perfect. The computer technology-based method mainly uses fitting and regression analysis methods to establish an empirical model that can predict the cycle life of lithium-ion batteries. However, due to the small amount of experimental data and the incomplete database, there are still certain errors in the prediction.

[0006] More and more researchers are applying machine learning techniques to predict the remaining service life of batteries using data collected in laboratories and online operating environments. For example, the invention patent with publication number CN105445671A discloses a lithium-ion battery life prediction method based on unscented particle filtering. By using a double exponential capacity decay model as a lithium-ion battery capacity degradation model, the state transfer equation and measurement equation of the lithium-ion battery capacity are obtained, and then the distribution of the initial value of the state variable of the model is obtained by combining the life decay data of other batteries, and then the life of the battery whose life needs to be predicted is predicted. Usually, these works make predictions after accumulating at least 25% of the operating data, or use special measurement methods to make predictions at the beginning of use, but due to the typical nonlinear degradation process (capacity degradation in early cycles is negligible) The data set used so far is relatively small (spanning a limited life range), so it is extremely challenging to use this method to accurately predict the cycle life.

[0007] The battery capacity decay during electrochemical cycling is due to a variety of degradation mechanisms, such as irreversible lithium loss due to the formation of solid electrolyte interphase (SEI), particle cracking, particle segregation, etc. Among these degradation mechanisms, chemical degradation caused by the reduction of solvent by lithium, which can lead to the formation of SEI, is widely considered to be the main cause of active lithium loss. In the prior art, the invention patent with publication number CN104849670A discloses a method for qualitatively judging the quality of battery life according to the ratio of battery current value to rated capacity, and proposes to obtain characteristic parameters of battery life through short-time testing to achieve rapid prediction of battery life, while the invention patent with publication number CN106908737A discloses a lithium-ion battery life prediction method based on battery electrochemical reaction mechanism simulation, through parameter measurement, establishment of electrochemical reaction mechanism model, model coupling calculation and other steps, according to the change of battery capacity during simulation calculation, mathematical fitting is performed to predict the service life of the battery, and the invention patent with publication number CN106383324A discloses a lithium-ion battery life prediction method based on capacity decay mechanism decomposition analysis, and establishes a multi-decay mode decomposition model through different decay mechanisms and predicts decay trend. However, these disclosed patent solutions all have the problem of short battery life prediction time, and fail to predict the entire life cycle of long-life lithium-ion batteries, and the life decay of lithium-ion batteries is a stage nonlinear process, and these disclosed patent solutions fail to predict and analyze the life decay of each stage. Summary of the invention

[0008] The purpose of the present invention is to address the deficiencies of the prior art and propose a method for predicting the capacity decay of lithium-ion batteries based on the coupling of chemical degradation and mechanical degradation, including analyzing and predicting the formation and growth of the solid electrolyte interface on the surface of the negative electrode material, and predicting the life of the lithium-ion battery in stages throughout its entire life cycle.

[0009] The present invention provides a method for predicting the service life of a lithium-ion battery based on an electrochemical model, comprising the following steps:

[0010] Step 1: Place several new lithium-ion batteries in a constant temperature and humidity chamber with an air-conditioning system and leave them for 20 to 60 minutes;

[0011] Step 2, charging each lithium-ion battery in step 1 at a constant current of 0.5C to 2C until the voltage reaches 3.65V, and then charging at a constant voltage until the cut-off current of each lithium-ion battery drops to 0.05C, and then stopping, and leaving for 20 minutes, and starting the air conditioning system to cool the battery during the charging process, so that the battery temperature is controlled within the range of 20 to 30°C;

[0012] Step 3, discharge each lithium-ion battery in step 2 to 2.4V at 0.5C, start the air conditioning device to cool the battery during the discharge process, so that the battery temperature is controlled within the range of 20-30°C, and let it stand for 20-60 minutes after the discharge is completed;

[0013] Step 4, numbering several new lithium-ion batteries in the order of 1, 2, 3..., n, where n is equal to the total number of lithium-ion batteries, and performing a cycle test on all lithium-ion batteries according to the method of steps 1 to 3 above for 50 to 300 cycles, wherein the number of cycle tests of all lithium-ion batteries is not exactly the same, and recording the voltage, current and capacity data of each lithium-ion battery in each cycle according to the numbering, and then disassembling the lithium-ion batteries that have completed the cycle test in a nitrogen environment to obtain the electrode plates and clean them for standby use;

[0014] Step 5, the composition and thickness of the solid electrolyte interface on the negative electrode sheet of the lithium-ion battery with different test cycle numbers, as well as the length and depth of the surface cracks of the test negative electrode material, and then the lithium-ion battery usage capacity is predicted through electrochemical models and mathematical algorithms.

[0015] Preferably, the lithium-ion battery includes one or more of a lithium iron ion battery, a ternary lithium-ion battery, a lithium manganate battery and a lithium titanate battery.

[0016] Furthermore, there are at least 6 lithium-ion batteries, and all the lithium-ion batteries are divided into at least 6 groups according to their numbers. In step 4, all the lithium-ion batteries are cycled and tested 50 to 300 times according to the methods of steps 1 to 3 above. Specifically, the lithium-ion batteries in the 6 groups are tested according to the number of cycles of 50, 100, 150, 200, 250, and 300.

[0017] More preferably, the electrode plates disassembled in step 4 are soaked in unused electrolyte for 1 to 2 hours, then soaked in alcohol for 30 minutes, and repeated 2 to 3 times.

[0018] Specifically, in step 5, the electrochemical model couples the chemical degradation of the interface between the electrolyte and the negative electrode material with the mechanical degradation of the negative electrode material.

[0019] More specifically, in step 5, the electrochemical model divides the active lithium loss of the lithium-ion battery into lithium loss during SEI formation in the first cycle (which is related to the SEI film formed on the entire surface of the carbon negative electrode active particles), lithium loss caused by SEI formation due to crack expansion on the newly formed surface in subsequent cycles, lithium loss caused by the continuous growth of the thickness of the SEI formed in the first cycle, and lithium loss caused by the continuous growth of the thickness of the new SEI formed on the surface due to crack expansion.

[0020] Furthermore, the solid electrolyte interface A on the negative electrode initial =4πR 2 +8πR 2 l cr ρ cr a 0 =4πR 2 (1+2l cr ρ cr a 0 ), the lithium loss caused by the continuous growth of SEI thickness formed in the first cycle Where F is the Faraday constant, ρ SEI is the density of the formed SEI film, M SEI is the molecular weight of the compounds that make up the SEI; is the thickness of the SEI film formed on the surface of the electrode plate, Among them, Q 0 initial is the active lithium capacity before the formation of SEI film, The battery efficiency of forming SEI film in the first cycle;

[0021] Right now Among them, l cr is the crack length, ρ cris the number of cracks per unit area of ​​the particle, a 0 is the depth of the initial crack, n SEI is the number of lithium atoms lost per SEI molecule formed;

[0022] If we simplify the above representation, we can make For B 0 ,but

[0023] The lithium loss caused by SEI formation on the newly formed surface during the subsequent cycles Where dQ is the differential of the battery capacity, dN is the differential of the number of cycles, and dA is cr is the differential of the negative electrode area, F is the Faraday constant, ρ SEI is the density of the formed SEI film, M SEI is the molecular weight of the compounds that make up the SEI; is the thickness of the SEI film formed on the surface of the electrode plate, The calculation method is consistent with the above.

[0024] The existing SEI will slowly grow thicker and cause capacity loss. The life test and model assume that the thickness of the SEI film is related to the number of cycles and is positively correlated with the square root of the number of cycles. Therefore, if the accurate measurement formula is not obtained, it is assumed to be L SEI =K th N 1 / 2 , where K th is a temperature-dependent constant. It is assumed that the SEI on the crack surface exposed after each discharge cycle quickly grows to the final thickness of the cycle. During the first charge and discharge cycle, it is assumed that all surfaces are covered with a thickness of L. SEI 0 The SEI film grows thicker from the second cycle. The initial exposure area of ​​the spherical particles after the first cycle is the sum of the spherical area and the existing crack area. For commercial batteries, the radius of electrode particles is about 5 microns. Generally only a few nanometers, so but The area is A initial =4πR 2 (1+2ρ cr a 0 l cr ), that is, the lithium loss caused by the continuous growth of SEI thickness formed in the first cycle is

[0025]

[0026] Among them, dL SEI is the thickness of the SEI film, K th is the thickness growth constant of the SEI film, and N is the number of battery cycles.

[0027] The surface area of ​​the electrode of a lithium-ion battery increases after each charge and discharge cycle. Once a new area is exposed to the electrolyte, a new SEI will form in this area during the subsequent lithiation process. The SEI formed in this crack area will increase in thickness in subsequent cycles. For the Nth cycle, the crack area opened in the first cycle will experience N-1 cycles, so the area will have a thickness growth associated with the N-1 cycle. Similarly, in cycle N, the crack area opened in the second cycle will experience N-2 cycles, so the thickness growth is related to the N-2 cycles. For N cycles, the crack accumulation experiment needs to be calculated from the second cycle. For N cycles, the capacity loss caused by the thickening of the cracked SEI film can be expressed as i is a positive integer; and the crack accumulation experiment needs to be calculated from the second cycle, then the lithium loss caused by the continuous growth of the thickness of the new SEI formed on the surface due to crack extension is:

[0028]

[0029] Where F is the Faraday constant, ρ cr is the number of cracks per unit area of ​​the particle, m is the fracture mechanics factor of the carbon anode, b is the stress constant 1.12, k is the Paris formula constant, σ θ.max is the maximum diffusion stress at the crack, n SEI is the number of lithium atoms lost per SEI molecule to form, ρ SEI is the density of the formed SEI film, M SEI is the molecular weight of the compounds that make up the SEI, is the thickness of the SEI film formed on the crack surface, K th is the thickness growth constant of the SEI film.

[0030] Furthermore, crack propagation and SEI film growth and thickening are processes that are closely related to the temperature of the battery, so the constant k in the Paris formula and the constant K in the SEI film growth formula are th The relationship with temperature follows the Arrhenius formula:

[0031]

[0032]

[0033] The k in the model 0 and k 0th As the fitting parameter, the activation energy of crack growth E a1 and the activation energy E of SEI film growth a2 It can be found by searching relevant literature data that the higher the temperature, the thinner the SEI film is. By substituting m, ρ cr , l cr and a 0 The corresponding k 0 and k th .

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] The technical solution of the present invention analyzes the formation and growth of the solid electrolyte interface on the surface of the negative electrode material, establishes special life prediction models for different negative electrode materials, and has higher prediction accuracy; it predicts the life of the lithium-ion battery over its entire life cycle to solve the current problems of small life prediction data volume and short prediction time, and analyzes different attenuation stages separately to solve the current problem of difficulty in predicting the nonlinear attenuation of the life of lithium-ion batteries. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The foregoing and following detailed description of the present invention will become more apparent when read in conjunction with the following drawings, in which:

[0037] Figure 1 Schematic diagram of lithium-ion battery service life prediction based on electrochemical model. DETAILED DESCRIPTION

[0038] The following is a further explanation of the technical solution for achieving the purpose of the present invention through several specific embodiments. It should be noted that the technical solution claimed for protection by the present invention includes but is not limited to the following embodiments.

[0039] As a specific implementation of the present invention, this embodiment provides a method for predicting the service life of a lithium-ion battery based on an electrochemical model, comprising the following steps:

[0040] Step 1, placing a number of brand new lithium-ion batteries in a constant temperature and humidity chamber with an air-conditioning system, and leaving them to stand for 20 to 60 minutes, wherein the lithium-ion batteries include one or more of lithium iron phosphate batteries, ternary lithium-ion batteries, lithium manganese oxide and lithium titanate batteries, and there are at least 6 lithium-ion batteries, and all lithium-ion batteries can be divided into at least 6 groups.

[0041] Step 2, charging each lithium-ion battery in step 1 at a constant current of 0.5C to 2C until the voltage reaches 3.65V, and then charging at a constant voltage until the cut-off current of each lithium-ion battery drops to 0.05C, and then leaving it for 20 minutes. During the charging process, the air conditioning system is started to cool the battery so that the battery temperature is controlled within the range of 20 to 30°C.

[0042] Step 3, discharge each lithium-ion battery in step 2 to 2.4V at 0.5C, start the air conditioning device to cool the battery during the discharge process, so that the battery temperature is controlled within the range of 20 to 30°C, and let it stand for 20 to 60 minutes after the discharge is completed.

[0043] Step 4, numbering several new lithium-ion batteries in the order of 1, 2, 3..., n, where n is equal to the total number of lithium-ion batteries, and performing a cycle test on all lithium-ion batteries 50 to 300 times according to the method of steps 1 to 3 above. Specifically, the lithium-ion batteries in the six groups are tested according to the number of cycles of 50, 100, 150, 200, 250, and 300, respectively, that is, the number of cycle tests of all lithium-ion batteries is not exactly the same, and the voltage, current, and capacity data of each cycle of each lithium-ion battery are recorded separately according to the numbering, and then the lithium-ion batteries that have completed the cycle test are disassembled in a nitrogen environment to obtain electrode plates and clean them for use. Preferably, the disassembled electrode plates are soaked in unused electrolyte for 1 to 2 hours, then soaked in alcohol for 30 minutes, and repeated 2 to 3 times to complete the cleaning process.

[0044] Step 5, the composition and thickness of the solid electrolyte interface on the negative electrode sheet of the lithium ion battery with different test cycle numbers, and the length and depth of the surface cracks of the test negative electrode material, and then predict the use capacity of the lithium ion battery through an electrochemical model and a mathematical algorithm, and the electrochemical model couples the chemical degradation of the interface between the electrolyte and the negative electrode material with the mechanical degradation of the negative electrode material. The electrochemical model divides the active lithium loss of the lithium ion battery into lithium loss during SEI formation in the first cycle (this is related to the SEI film formed on the entire surface of the carbon negative electrode active particles), lithium loss caused by SEI formation due to crack expansion on the newly formed surface in subsequent cycles, lithium loss caused by the continuous growth of the thickness of the SEI formed in the first cycle, and lithium loss caused by the continuous growth of the thickness of the new SEI formed on the surface due to crack expansion.

[0045] Specifically, the solid electrolyte interface A on the negative electrode sheet initial =4πR 2 +8πR 2 l cr ρ cr a 0 =4πR 2(1+2l cr ρ cr a 0 ), the lithium loss caused by the continuous growth of SEI thickness formed in the first cycle Where F is the Faraday constant, ρ SEI is the density of the formed SEI film, M SEI is the molecular weight of the compounds that make up the SEI; is the thickness of the SEI film formed on the surface of the electrode plate, Among them, Q 0 initial is the active lithium capacity before the formation of SEI film, The battery efficiency of forming SEI film in the first cycle;

[0046] Right now Among them, l cr is the crack length, ρ cr is the number of cracks per unit area of ​​the particle, a 0 is the depth of the initial crack, n SEI is the number of lithium atoms lost by each SEI molecule; if the above expression is simplified, we can make For B 0 ,but

[0047] The lithium loss caused by SEI formation on the newly formed surface during the subsequent cycles Where dQ is the differential of the battery capacity, dN is the differential of the number of cycles, and dA is cr is the differential of the negative electrode area, F is the Faraday constant, ρ SEI is the density of the formed SEI film, M SEI is the molecular weight of the compounds that make up the SEI; is the thickness of the SEI film formed on the surface of the electrode plate, The calculation method is the same as above; specifically, it is assumed that under constant current conditions, the mass transfer in spherical particles satisfies Fick's mass transfer law. The mass transfer of active lithium in electrode particles is The initial and boundary conditions are C(r,0)=C 0 , 0≤r≤R; C(0,t)=finite, t≥0, where i is the current on the surface of the electrode particles.

[0048] So the solution formula under constant current conditions is

[0049]

[0050] Where x is r / R, τ = tD / R 2 ,λn is tan(λ n )=λ n The positive root of (n=1, 2, 3…), the concentration c(r, t)>0 and cannot exceed the limiting concentration of the available active sites of the negative electrode material.

[0051] For a thin film electrode, the specific surface area is a s , thickness L, geometric area A, battery current I cell , then the current density on the surface of the electrode particles is The specific surface area of ​​the porous electrode is s and the radius R of the active spherical particle and the porosity ε of the solid phase 1a About s =3ε 1a / R, so The maximum stress on the electrode surface is the stress in the tangential direction of the particle surface at the beginning of discharge, which can be quantified as

[0052]

[0053] Where E is Young's modulus, v is the Poisson's ratio of the electrode material, Ω is the partial molar volume of the carbon material, and D is the mass transfer factor in the solid phase.

[0054] For deep discharge at low discharge rates, this is a long time process, and the charge and discharge cycle is longer than the characteristic time R of the mass transfer and diffusion of lithium ions in the negative electrode material. 2 / D is much larger, so the exponential term in the stress formula is equal to 0, and the summation term can be ignored. θ The magnitude of increases monotonically to an asymptote and is maximum on the particle surface, so we call the long-term asymptotic value of the surface tangential stress σ θ，max .

[0055] The stress formula is applied to the generation and expansion of surface cracks on the negative electrode carbon material particles. Since we study the surface of micro-nano particles in isolation, the influence of surface mechanics (such as surface tension) is not considered here, and the radial component of stress is zero, so the maximum value of the surface stress in the tangential direction is

[0056] For a cycle of stress, the electrode surface is subject to tensile stress during most of the discharge process, while there is only a small compressive stress during the charging process. The Paris law is used to study the expansion of a crack with a length of a as the charge and discharge cycle N increases. Where k and m are constants, and the type I stress intensity factor K between the full load state and the charging state corresponding to the steady-state constant current discharge is I difference Therefore, only the crack propagation caused by tensile stress during the discharge process needs to be studied.

[0057] For very small cracks on the particle surface, the crack length a is much smaller than R, and the constant b = 1.12, which can be simplified to:

[0058]

[0059]

[0060]

[0061] In (1, a 0 ) to (N,a) we can get:

[0062]

[0063] During the discharge process, as the crack expands, the area of ​​the newly exposed surface is 2l cr (da), where l cr is the length of the crack along the surface direction. The number and length of the cracks in the surface direction are considered constant (each cycle extends along the radial direction of the particle). Therefore, ρ is defined cr is the crack density on the surface of the negative electrode particles (the number of cracks per unit particle surface area).

[0064] The number of cracks is: N cr =4πR 2 ρ cr ,

[0065] The total area of ​​the crack is: A cr =8πR 2 ρ cr al cr ,

[0066] Then, the crack area change in each discharge cycle is:

[0067]

[0068] Assume that the newly exposed surface forms Li upon contact with the electrolyte during the next discharge cycle. 2 CO 3 , thickness is L SEI 0 , then the capacity loss caused by crack extension changes with the number of cycles:

[0069]

[0070] Among them, n SEI is the number of lithium atoms lost per SEI molecule to form, ρSEI is the density of the formed SEI film, M SEI is the molecular weight of the compounds that make up the SEI.

[0071] The lithium loss caused by SEI formation on the newly formed surface due to crack propagation in the subsequent cycle is:

[0072]

[0073] Where F is the Faraday constant, l cr is the crack length, ρ cr is the number of cracks per unit area of ​​the particle, a 0 is the depth of the initial crack, m is the fracture mechanics factor of the carbon anode, b is the stress constant 1.12, k is the Paris formula constant, σ θ.max is the maximum diffusion stress at the crack, n SEI is the number of lithium atoms lost per SEI molecule to form, ρ SEI is the density of the formed SEI film, M SEI is the molecular weight of the compounds that make up the SEI, is the thickness of the SEI film formed on the crack surface.

[0074] The existing SEI will slowly grow thicker and cause capacity loss. The life test and model assume that the thickness of the SEI film is related to the number of cycles and is positively correlated with the square root of the number of cycles. Therefore, if the accurate measurement formula is not obtained, it is assumed to be L SEI =K th N 1 / 2 , where K th is a temperature-dependent constant. It is assumed that the SEI on the crack surface exposed after each discharge cycle quickly grows to the final thickness of the cycle. During the first charge and discharge cycle, it is assumed that all surfaces are covered with a thickness of L. SEI 0 The SEI film grows thicker from the second cycle. The initial exposure area of ​​the spherical particles after the first cycle is the sum of the spherical area and the existing crack area.

[0075] For commercial batteries, the radius of electrode particles is about 5 microns. Generally only a few nanometers, so but The area is A initial =4πR 2 (1+2ρ cr a 0 l cr), that is, the lithium loss caused by the continuous growth of SEI thickness formed in the first cycle is

[0076]

[0077] Among them, dL SEI is the thickness of the SEI film, K th is the thickness growth constant of the SEI film, and N is the number of battery cycles.

[0078] The surface area of ​​the electrode of a lithium-ion battery increases after each charge and discharge cycle. Once a new area is exposed to the electrolyte, a new SEI will form in this area during the subsequent lithiation process. The SEI formed in this crack area will increase in thickness in subsequent cycles. For the Nth cycle, the crack area opened in the first cycle will experience N-1 cycles, so the area will have a thickness growth associated with the N-1 cycle. Similarly, in cycle N, the crack area opened in the second cycle will experience N-2 cycles, so the thickness growth is related to the N-2 cycles. For N cycles, the crack accumulation experiment needs to be calculated from the second cycle. For N cycles, the capacity loss caused by the thickening of the cracked SEI film can be expressed as i is a positive integer; and the crack accumulation experiment needs to be calculated from the second cycle, then the lithium loss caused by the continuous growth of the thickness of the new SEI formed on the surface due to crack extension is:

[0079]

[0080] Where F is the Faraday constant, ρ cr is the number of cracks per unit area of ​​the particle, m is the fracture mechanics factor of the carbon anode, b is the stress constant 1.12, k is the Paris formula constant, σ θ.max is the maximum diffusion stress at the crack, n SEI is the number of lithium atoms lost per SEI molecule to form, ρ SEI is the density of the formed SEI film, M SEI is the molecular weight of the compounds that make up the SEI, is the thickness of the SEI film formed on the crack surface, K th is the thickness growth constant of the SEI film.

[0081] The total capacity loss is obtained by integrating the lithium loss caused by the formation of SEI on the newly formed surface due to crack expansion in the above subsequent cycles, the lithium loss caused by the continuous growth of the SEI thickness formed in the first cycle, and the lithium loss caused by the continuous growth of the new SEI thickness formed on the surface due to crack expansion:

[0082]

[0083] The fracture mechanics factor m is between 2 and 5, and the capacity of the second cycle of the battery is recorded as Q 0 , which is different from the capacity Q before the formation of SEI film 0 initial The relationship is in The cell efficiency of the first cycle when SEI film is formed.

[0084] For lithium batteries, the capacity of the negative electrode material is generally over-capacitated, so the capacity of the entire battery is controlled by the positive electrode material. Among them, q r =negative electrode capacity / positive electrode capacity, generally 1.1.

[0085] Will Defined as A, Defined as B. Defined as C, the capacity loss formula is simplified to:

[0086]

[0087] Integrating the above formula gives the capacity Q',

[0088]

[0089] Thus, the relationship function between capacity Q' and cycle number N is obtained.

[0090] It is particularly important to note that if the SEI reaches its final thickness immediately after each discharge cycle and does not grow thicker, then

[0091] Furthermore, crack propagation and SEI film growth and thickening are processes that are closely related to the temperature of the battery, so the constant k in the Paris formula and the constant K in the SEI film growth formula are th The relationship with temperature follows the Arrhenius formula:

[0092]

[0093]

[0094] The k in the model 0 and k 0 th As the fitting parameter, the activation energy of crack growth E a1 and the activation energy E of SEI film growth a2 It can be found by searching relevant literature data that the higher the temperature, the thinner the SEI film is. By substituting m, ρ cr , lcr and a 0 The corresponding k 0 and k th .

Claims

1. A method for predicting the service life of a lithium-ion battery based on an electrochemical model, characterized in that: The following steps are involved: Step 1: Place several new lithium-ion batteries in a constant temperature and humidity chamber with an air-conditioning system and leave them for 20 to 60 minutes; Step 2, charging each lithium-ion battery in step 1 at a constant current of 0.5C to 2C until the voltage reaches 3.65V, and then charging at a constant voltage until the cut-off current of each lithium-ion battery drops to 0.05C, and then stopping, and leaving for 20 minutes, and starting the air conditioning system to cool the battery during the charging process, so that the battery temperature is controlled within the range of 20 to 30°C; Step 3, discharge each lithium-ion battery in step 2 to 2.4V at 0.5C, start the air conditioning device to cool the battery during the discharge process, so that the battery temperature is controlled within the range of 20-30°C, and let it stand for 20-60 minutes after the discharge is completed; Step 4, numbering several new lithium-ion batteries in the order of 1, 2, 3..., n, where n is equal to the total number of lithium-ion batteries, and performing a cycle test on all lithium-ion batteries according to the method of steps 1 to 3 above for 50 to 300 cycles, wherein the number of cycle tests of all lithium-ion batteries is not exactly the same, and recording the voltage, current and capacity data of each lithium-ion battery in each cycle according to the numbering, and then disassembling the lithium-ion batteries that have completed the cycle test in a nitrogen environment to obtain the electrode plates and clean them for standby use; Step 5, the composition and thickness of the solid electrolyte interface on the negative electrode sheet of the lithium-ion battery with different test cycle numbers, as well as the length and depth of the surface cracks of the test negative electrode material, and then the lithium-ion battery usage capacity is predicted through electrochemical models and mathematical algorithms.

2. A method for predicting the service life of a lithium-ion battery based on an electrochemical model as claimed in claim 1, characterized in that: The lithium-ion battery includes one or more of a lithium iron ion battery, a ternary lithium-ion battery, a lithium manganate battery and a lithium titanate battery.

3. A method for predicting the service life of a lithium-ion battery based on an electrochemical model as claimed in claim 1, characterized in that: There are at least 6 lithium-ion batteries, and all the lithium-ion batteries are divided into at least 6 groups according to their numbers. In step 4, all the lithium-ion batteries are cycled for 50 to 300 times according to the methods of steps 1 to 3 above. Specifically, the lithium-ion batteries in the 6 groups are tested for 50, 100, 150, 200, 250, and 300 cycles, respectively.

4. The method for predicting the service life of a lithium-ion battery based on an electrochemical model as claimed in claim 1, characterized in that: The disassembled electrode plates in step 4 are soaked in unused electrolyte for 1 to 2 hours, then soaked in alcohol for 30 minutes, and repeated 2 to 3 times.

5. The method for predicting the service life of a lithium-ion battery based on an electrochemical model as claimed in claim 1, characterized in that: In step 5, the electrochemical model couples the chemical degradation of the interface between the electrolyte and the negative electrode material with the mechanical degradation of the negative electrode material.

6. The method for predicting the service life of a lithium-ion battery based on an electrochemical model as claimed in claim 1, characterized in that: In step 5, the electrochemical model divides the active lithium loss of the lithium-ion battery into lithium loss during SEI formation in the first cycle, lithium loss caused by SEI formation on the newly formed surface due to crack extension in subsequent cycles, lithium loss caused by continued growth in the thickness of the SEI formed in the first cycle, and lithium loss caused by continued growth in the thickness of the new SEI formed on the surface due to crack extension.

7. A method for predicting the service life of a lithium-ion battery based on an electrochemical model as claimed in claim 6, characterized in that: Solid electrolyte interface A on the negative electrode initial =4πR 2 +8πR 2 l cr ρ cr a0=4πR 2 (1+2l cr ·ρ cr a0), lithium loss caused by the continuous growth of SEI thickness formed in the first cycle Where F is the Faraday constant, ρ SEI is the density of the formed SEI film, M SEI is the molecular weight of the compounds that make up the SEI; is the thickness of the SEI film formed on the surface of the electrode plate, Among them, Q0 initial is the active lithium capacity before the formation of SEI film, The battery efficiency of forming SEI film in the first cycle; Right now Among them, l cr is the crack length, ρ cr is the number of cracks per unit area of ​​the particle, a0 is the depth of the initial crack, n SEI is the number of lithium atoms lost per SEI molecule formed.

8. A method for predicting the service life of a lithium-ion battery based on an electrochemical model as claimed in claim 6 or 7, characterized in that: The lithium loss caused by SEI formation on the newly formed surface during the subsequent cycles Where dQ is the differential of the battery capacity, dN is the differential of the number of cycles, and dA is cr is the differential of the negative electrode area, F is the Faraday constant, ρ SEI is the density of the formed SEI film, M SEI is the molecular weight of the compounds that make up the SEI; is the thickness of the SEI film formed on the surface of the electrode, n SEI is the number of lithium atoms lost per SEI molecule formed.

9. The method for predicting the service life of a lithium-ion battery based on an electrochemical model as claimed in claim 6, characterized in that: During the first charge-discharge cycle, assume that all surfaces are covered with a layer with a thickness of L. SEI 0 The SEI film grows thicker from the second cycle. The initial exposure area of ​​the spherical particles after the first cycle is the sum of the spherical area and the existing crack area. For commercial batteries, the radius of the electrode particles is 5 microns, and Generally only a few nanometers, so but The area is A inital =4πR 2 (1+2ρ cr a0l cr ), that is, the lithium loss caused by the continuous growth of SEI thickness formed in the first cycle is: Among them, dL SEI is the thickness of the SEI film, K th is the thickness growth constant of the SEI film, and N is the number of battery cycles.

10. The method for predicting the service life of a lithium-ion battery based on an electrochemical model according to claim 7, characterized in that: After N cycles, the capacity loss caused by the thickening of the SEI film due to cracks can be expressed as i is a positive integer; and the crack accumulation experiment needs to be calculated from the second cycle, then the lithium loss caused by the continuous growth of the thickness of the new SEI formed on the surface due to crack extension is: Where F is the Faraday constant, ρ cr is the number of cracks per unit area of ​​the particle, m is the fracture mechanics factor of the carbon anode, b is the stress constant 1.12, k is the Paris formula constant, σ θ.max is the maximum diffusion stress at the crack, n SEI is the number of lithium atoms lost per SEI molecule to form, ρ SEI is the density of the formed SEI film, M SEI is the molecular weight of the compounds that make up the SEI, is the thickness of the SEI film formed on the crack surface, K th is the thickness growth constant of the SEI film.

Images