Lithium ion battery super-linear cycle life prediction method based on rate acceleration

Abstract

The invention discloses a lithium ion battery super-linear cycle life prediction method based on rate acceleration, and the method comprises the following steps: S100, placing lithium batteries of the same system in a constant-temperature incubator with a constant temperature, connecting the lithium batteries to a test cabinet, and carrying out charging and discharging circulation at a set rate gradient to obtain data of circulation parameters and a circulation curve; s200, creating a capacity degradation model of the cycle time and the charge-discharge rate stress factor according to the cycle parameters; s300, fitting the data of the cycle parameters, and obtaining a cycle life curve model under a target condition by using the capacity degradation model; and S400, verifying the error between the actual cycle curve under the target condition and the cycle curve obtained by the cycle life curve model. According to the method, the nonlinear cycle life prediction of the lithium battery under a relatively low rate can be realized, the method is simple and easy to operate, the test period is short, the evaluation progress can be accelerated, resources are saved, and the purpose of rapidly predicting the life of the battery cell is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of lithium battery detection, and in particular to a method for predicting superlinear cycle life of lithium ion batteries based on rate acceleration. Background Art

[0002] Lithium-ion batteries are rechargeable energy storage devices with high energy density and based on intercalation structures. They are widely used in small portable electronic devices, electric vehicles, energy storage products, etc. However, as lithium batteries are continuously charged and discharged, the combined effects of various failure mechanisms cause their performance to gradually decline and their capacity to decay, including the formation and growth of solid electrolyte membranes (SEI), the reduction of electrode material porosity, the cracking of particles, the decomposition of electrolytes, etc. At the same time, the impedance of the battery increases and the coulomb efficiency decreases. Therefore, it is very necessary to predict the life of the battery. However, the actual evaluation cycle of lithium batteries is too long, which greatly delays the progress of product development and verification, resulting in waste of resources and increased costs. Therefore, accelerated life experiments are used to speed up the evaluation progress and quickly predict the life of the battery cell.

[0003] Accelerated cycle testing is a method of finding the relationship between battery life and accelerated stress without changing the failure mechanism, and using the life characteristics under high accelerated stress to extrapolate the life characteristics under normal or lower stress. Generally, temperature or rate is used as the accelerated stress. When the battery cycle life curve is linear or sublinear attenuation, the single power function Y=1-K1*X ​​is currently used. b1 To simulate the curve trend, K1 is a function related to temperature or rate. When the life curve of the lithium battery shows a two-phase decay and an obvious inflection point appears in the middle, the model is no longer applicable. Therefore, a super-linear cycle life prediction method for lithium-ion batteries that includes an inflection point is needed. Summary of the invention

[0004] The purpose of the present invention is to provide a super-linear cycle life prediction method for lithium-ion batteries based on rate acceleration, which is simple and easy to operate, has a short test cycle, and can realize battery life prediction under different charge and discharge rates.

[0005] In one aspect of the present invention, the present invention proposes a method for predicting the superlinear cycle life of lithium-ion batteries based on rate acceleration. According to an embodiment of the present invention, the method comprises the following steps:

[0006] S100, placing lithium batteries of the same system in a constant temperature thermostat, connecting the battery to a test cabinet, and performing charge and discharge cycles at a set rate gradient to obtain cycle parameter data and cycle curves;

[0007] S200, creating a capacity degradation model of cycle time and charge / discharge rate stress factor according to the cycle parameters;

[0008] S300, fitting the data of the cycle parameters, and obtaining a cycle life curve model under target conditions using a capacity degradation model;

[0009] S400, verifying the error between the actual cycle curve under target conditions and the cycle curve obtained by the cycle life curve model.

[0010] In addition, the method for predicting the super-linear cycle life of a lithium-ion battery based on rate acceleration according to the above embodiment of the present invention may also have the following additional technical features:

[0011] In some embodiments of the present invention, in step S100, the temperature of the thermostat is 10°C-60°C, and the rate gradient is no greater than 3C.

[0012] In some embodiments of the present invention, in step S100, the cycle parameters include the number of cycles, cycle time and rate, and normalized discharge capacity.

[0013] In some embodiments of the present invention, in step S200, the capacity degradation model created is as follows: Y = 1-Arate1*X b1 -Arate2*X b2 , where Ln(Arate1)=M1*Ln(C rate )+N1,Ln(Arate2)=M2*Ln(C rate )+N2,b2=E*C rate +F; Y is the current normalized discharge capacity of the battery, X is the cycle time, and the model is built using nonlinear regression technology. Arate1, b1, Arate2, and b2 are model coefficients. Arate1 and Arate2 have an inverse power law function relationship with the charge and discharge rate. b2 is a linear function related to the rate. M1, N1, M2, N2, E, and F are model parameters. C rate It is the charge and discharge rate of the battery.

[0014] In some embodiments of the present invention, the first power function Arate1*X b1 The slow decay process of the exponent less than 1 before the inflection point on the simulated charge and discharge cycle curve is attributed to the capacity loss caused by the SEI film; the second power function Arate2*X b2 The rapid decay process after the inflection point on the charge-discharge cycle curve is simulated to simulate irreversible losses, which include lithium deposition or increased internal resistance.

[0015] In some embodiments of the present invention, b1 is independent of stress factors, and b2 is greater than 1. Specifically, in the first power function, the capacity loss caused by the SEI film is generally proportional to the square root of the cycle time, so b1 is independent of stress factors and is generally about 0.5; in the second power function, the capacity loss caused by irreversible losses such as lithium deposition or increase in internal resistance increases rapidly after reaching a certain threshold, and the impact gradually exceeds the SEI film loss, resulting in an obvious inflection point on the cycle curve, so b2 is greater than 1, and presents a linear function relationship with the rate, b2 = E*C rate +F. At the same time, when the acceleration factor is a non-thermal factor (such as current, voltage, power), some life characteristics of the research object and the factor conform to the inverse power law formula, § = Av -c Here, it is assumed that Arate1, Arate2 and the charge and discharge rate show an inverse power law function relationship, that is, Ln(Arate1)=M1*Ln(C rate )+N1,Ln(Arate2)=M2*Ln(C rate )+N2, verify the adaptability of the model.

[0016] In some embodiments of the present invention, in step S300, the fitting method used is global fitting.

[0017] In some embodiments of the present invention, the premise of global fitting is that the attenuation mechanisms of the battery group are consistent.

[0018] Compared with the prior art, the present invention has the following beneficial effects:

[0019] 1) The prediction method of the present invention is applicable to lithium-ion batteries that can produce two-phase attenuation behavior, wherein the positive electrode of the battery is LFP, LiCoO2, Li(NiCoAl)O2(NCA) or Li(NiMnCo)O2(NCM), and the negative electrode is graphite or Si / C material, wherein the lithium-ion battery includes all lithium-ion full battery systems, specifically including square lithium batteries, soft-pack lithium batteries, and cylindrical lithium batteries.

[0020] 2) The present invention is of great significance for the life prediction and status evaluation of lithium batteries. At present, there are certain limitations in the empirical model of lithium batteries with diving cycle curves containing inflection points, and most accelerated experiments use temperature as the accelerating stress. The present invention uses the charge and discharge rate as the accelerating stress, creates a double power function model based on the inverse power law relationship, and can achieve the life prediction of super-linear cycles of lithium batteries at relatively low rates (such as when Crate≤1). This method is simple and easy to operate, with a short test cycle, and can achieve battery life prediction at different charge and discharge rates.

[0021] 3) The present invention proposes a double power function model based on rate acceleration, which simulates the two-phase attenuation cycle curve, that is, the diving curve with obvious inflection point with high accuracy; and when a relatively high rate is used as the acceleration condition, the parameters at a lower rate can be extrapolated from the fitting parameters at a higher rate to create a lower rate cycle curve. It has been verified that the error with the actual cycle curve is small, which greatly saves the lithium battery cycle evaluation time, thereby accelerating the product development and verification progress and saving costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a flow chart of a method for predicting super-linear cycle life of a lithium-ion battery based on accelerated charge and discharge rate in an embodiment of the present invention;

[0023] Figure 2 This is a graph of cycle fitting results of the first group of iron-lithium batteries at different charge and discharge rates in Example 1 of the present invention;

[0024] Figure 3 This is a graph of cycle fitting results of the second group of iron-lithium batteries in Example 2 of the present invention at different charge and discharge rates. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] Example 1

[0027] like Figure 1 As shown, a method for predicting the ultra-linear cycle life of a lithium-ion battery based on accelerated charge and discharge rate comprises the following steps:

[0028] S100, select at least 6 52Ah lithium iron phosphate batteries, divide them into three groups, each group has two parallel samples, all are placed in a constant temperature box at 25℃, and cycle at 1C / 1C, 1.5C / 1.5C, and 2C / 2C charge and discharge rates, respectively. The capacity is fixed every 200 cycles, and the relationship between normalized discharge capacity and cycle time is calculated, and the cycle curve is drawn, such as Figure 2 shown.

[0029] S200, based on the cycle data, create a double power function capacity degradation model Y=1-Arate1*X of cycle time and charge / discharge rate stress factor b1 -Arate2*X b2 , where Ln(Arate1)=M1*Ln(C rate)+N1,Ln(Arate2)=M2*Ln(C rate )+N2,b2=E*C rate +F. Y is the current normalized discharge capacity of the battery, X is the cycle time, and the model is built using nonlinear regression technology. Arate1, b1, Arate2, and b2 are model coefficients. Arate1 and Arate2 have an inverse power law function relationship with the charge and discharge rate. b2 is a linear function related to the rate. M1, N1, M2, N2, E, and F are model parameters. C rate The first power function Arate1*X is the rate at which the battery is charged and discharged. b1 It mainly simulates the slow decay process of the index less than 1 before the inflection point on the cycle curve, which is attributed to the capacity loss caused by the SEI film; the second power function Arate2*X b2 Simulate the rapid decay process after the inflection point on the cycle curve, and simulate irreversible losses such as lithium deposition or increased internal resistance.

[0030] S300, normalized cycle capacity retention data of 1.5C / 1.5C and 2C / 2C after global fitting.

[0031] Global fitting process: 1) In the Origin software, the x-axis is the cycle time and the y-axis is the normalized discharge capacity. The three groups of cycle curves at different rates are uniformly plotted; 2) In the Analysis-Fitting-Nonlinear Curve Fit interface, a new function formula is created: Y = 1-K1X b1 -K2X b2 ; and given appropriate initial values ​​of K1 and K2, after iterative operation, the double power function is successfully created; 3) For the two sets of cycle curves under higher rates, determine the fitting range through the "Slection on all plots" button on the left, call out the fitting formula through "Crtl+Y", select the double power function just created, and in the "Data selection" interface, click the "Multi-Data Fit Mode" button to select "Globle Fit" global fitting; 4) In the "Parameters" interface, use the "Fit until converged" button to preliminarily fit the two sets of curves until convergence; 5) Select the "Share" button to share the b1 value, fix the previous power value of the two sets of data, and click the "Fit" button to obtain the fitting results when global fitting (fixed b1), that is, K1, b1 and K2, b2 under two sets of different cycle rates (1.5C / 1.5C and 2C / 2C), that is, Arate1, b1, Arate2, b2 in the formula, such as Figure 2 6) According to Arate1 and Arate2 and C rateThe inverse power law relationship between b2 and C rate The linear function relationship between them is used to obtain M1, N1, M2, N2, E, and F, as shown in Table 1:

[0032] Table 1 Fitting results of 1.5C / 1.5C and 2C / 2C cycle data of 52Ah iron-lithium battery in Example 1

[0033]

[0034] When C rate =1, substitute M1, N1, M2, N2, E, F into the formula Ln(Arate1)=M1*Ln(C rate )+N1,Ln(Arate2)=M2*Ln(C rate )+N2,b2=E*C rate +F,; the capacity degradation trend under 1C / 1C cycle rate can be obtained, and the formula is as follows:

[0035] Y=1-4.32E-4*X 0.6844 -2.97E-16*X 3.8053 .

[0036] S400, such as Figure 2 As shown in the figure, the 1C / 1C prediction curve is more consistent with the 1C / 1C actual cycle curve trajectory, with a small error. As the cycle progresses, the error increases slightly, which may be the error generated during the actual cycle. If this method is used to implement the prediction, according to the 70% capacity cutoff point, the 1C / 1C test of this group of batteries is expected to save about 58% of the test time.

[0037] Example 2

[0038] like Figure 1 As shown, a method for predicting the ultra-linear cycle life of a lithium-ion battery based on accelerated charge and discharge rate comprises the following steps:

[0039] S100. Select 6 iron-lithium 102Ah battery systems and divide them into three groups. Each group has two parallel samples. All of them are placed in a constant temperature box at 45°C and cycled at charge and discharge rate gradients of 0.5C / 0.5C, 1C / 1C, and 1.5C / 1.5C. The capacity is constant every 200 cycles, and the relationship between the capacity retention rate and the number of cycles is calculated.

[0040] S200, create a capacity degradation model of cycle time and charge and discharge rate stress factor, that is, the inverse power law double power function degradation model Y = 1-Arate1*X b1 -Arate2*X b2 , where Ln(Arate1)=M1*Ln(C rate)+N1,Ln(Arate2)=M2*Ln(C rate )+N2,b2=E*C rate +F. Y is the current normalized discharge capacity of the battery, X is the cycle time, and the model is built using nonlinear regression technology. Arate1, b1, Arate2, and b2 are model coefficients. Arate1 and Arate2 have an inverse power law function relationship with the charge and discharge rate. b2 is a linear function related to the rate. M1, N1, M2, N2, E, and F are model parameters. C rate The first power function Arate1*X is the rate at which the battery is charged and discharged. b1 It mainly simulates the slow decay process of the index less than 1 before the inflection point on the cycle curve, which is attributed to the capacity loss caused by the SEI film; the second power function Arate2*X b2 Simulate the rapid decay process after the inflection point on the cycle curve, and simulate irreversible losses such as lithium deposition or increased internal resistance.

[0041] S300, according to the global fitting method in Example 1, after fitting the data of 1C / 1C and 1.5C / 1.5C, as shown in FIG. Figure 3 As shown in Table 2, M1, N1, M2, N2, E, and F are obtained:

[0042] Table 2 Fitting results of 102Ah 1C / 1C, 1.5C / 1.5C cycle data in Example 2

[0043]

[0044] When C rate =0.5, substitute M1, N1, M2, N2, E, F into the formula Ln(Arate1)=M1*Ln(C rate )+N1,Ln(Arate2)=M2*Ln(C rate )+N2,b2=E*C rate +F; the capacity degradation model under 0.5C / 0.5C charge and discharge rate can be obtained, and its formula is:

[0045] Y=1-3.22E-3*X 0.4289 -1.24E-21*X 5.6581 .

[0046] S400, such as Figure 3 As shown, the 0.5C / 0.5C prediction curve can be verified to simulate the actual cycle curve more accurately with a small error.

[0047] In summary, a superlinear cycle life prediction method for lithium-ion batteries based on charge and discharge rate acceleration is created by a double power function model based on the inverse power law, which well simulates the superlinear cycle trajectory lithium battery life curve containing an inflection point. Based on cycle rate acceleration, through the global fitting method in Example 1, the cycle life characteristics at a lower rate cycle can be extrapolated through the cycle data and fitting model at a higher rate, which can effectively predict the approximate position of the inflection point and shorten the test time. And its feasibility has been verified by actual testing of lithium batteries. The method is simple and easy to operate, has a short test cycle, can speed up the evaluation progress, save resources, and achieve the purpose of quickly predicting the life of the battery cell.

[0048] The above contents are merely examples and explanations of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the present invention or exceed the scope defined by the claims, they shall all fall within the protection scope of the present invention.

Claims

1. A method for predicting the superlinear cycle life of lithium-ion batteries based on rate acceleration, characterized in that: The following steps are involved: S100, placing lithium batteries of the same system in a constant temperature thermostat, connecting the battery to a test cabinet, and performing charge and discharge cycles at a set rate gradient to obtain cycle parameter data and cycle curves; S200, creating a capacity degradation model of cycle time and charge / discharge rate stress factor according to the cycle parameters; S300, fitting the data of the cycle parameters, and obtaining a cycle life curve model under target conditions using a capacity degradation model; S400, verifying the error between the actual cycle curve under target conditions and the cycle curve obtained by the cycle life curve model.

2. The method for predicting the super-linear cycle life of a lithium-ion battery based on rate acceleration according to claim 1, characterized in that: In step S100, the temperature of the thermostat is 10°C-60°C, and the rate gradient is not greater than 3°C.

3. The method for predicting the super-linear cycle life of a lithium-ion battery based on rate acceleration according to claim 1, characterized in that: In the step S100, the cycle parameters include the number of cycles, cycle time, rate, and normalized discharge capacity.

4. The method for predicting the super-linear cycle life of a lithium-ion battery based on rate acceleration according to claim 1, characterized in that: In step S200, the capacity degradation model created is as follows: Y = 1-Arate1*X b1 -Arate2*X b2 , where Ln(Arate1)=M1*Ln(C rate )+N1,Ln(Arate2)=M2*Ln(C rate )+N2,b2=E*C rate +F; Y is the current normalized discharge capacity of the battery, X is the cycle time, and the model is built using nonlinear regression technology. Arate1, b1, Arate2, and b2 are model coefficients. Arate1 and Arate2 have an inverse power law function relationship with the charge and discharge rate. b2 is a linear function related to the rate. M1, N1, M2, N2, E, and F are model parameters. C rate It is the charge and discharge rate of the battery.

5. The method for predicting the super-linear cycle life of a lithium-ion battery based on rate acceleration according to claim 4, characterized in that: The first power function Arate1*X b1 The slow decay process of the exponent less than 1 before the inflection point on the simulated charge and discharge cycle curve is attributed to the capacity loss caused by the SEI film; the second power function Arate2*X b2 The rapid decay process after the inflection point on the charge-discharge cycle curve is simulated to simulate irreversible losses, which include lithium deposition or increased internal resistance.

6. The method for predicting the super-linear cycle life of a lithium-ion battery based on rate acceleration according to claim 4, characterized in that: b1 has nothing to do with stress factors, and b2 is greater than 1.

7. The method for predicting super-linear cycle life of lithium-ion batteries based on rate acceleration according to claim 1, characterized in that: In step S300, the fitting method used is global fitting.

8. The method for predicting the super-linear cycle life of a lithium-ion battery based on rate acceleration according to claim 7, characterized in that: The premise of global fitting is that the attenuation mechanism of the battery group is consistent.

Images