An online early warning method and system for lithium-ion battery diving point

Abstract

The present invention belongs to the technical field related to battery storage and discloses an online early warning method and system for lithium-ion battery diving points. The method comprises: combining capacity loss caused by SEI film growth and capacity loss caused by lithium plating to construct a hybrid model reflecting the temporal trend of battery capacity loss; obtaining total discharge capacity loss data from early battery cycles to fit the hybrid model and determine the parameters of the hybrid model; using the output of the hybrid model as the theoretical capacity loss caused by the solid electrolyte interface film, sliding a time window, and obtaining the slope of the linear correlation equation between the theoretical capacity loss and the actual discharge capacity loss within each time window; calculating the first-order autocorrelation coefficient of the slope, and issuing a diving warning when the first-order autocorrelation coefficient reaches a minimum value. The present invention can achieve online early warning of lithium-ion battery diving points through a simple method and has strong physical interpretability.

Description

Technical Field

[0001] The present invention belongs to the technical field related to battery storage, and more specifically, relates to an online early warning method and system for a lithium-ion battery diving point. Background Art

[0002] Lithium-ion batteries are dynamic, time-varying electrochemical systems. As active materials are consumed through cycles, their capacity degrades over time. When battery capacity declines to a certain point, it experiences a capacity drop, a sudden, accelerated decline in capacity, reaching the end of its lifespan over a relatively short period of time. This sudden capacity drop can disrupt battery usage plans and even cause large equipment to cease operation, resulting in unforeseen losses. Therefore, online early warning of lithium battery capacity drop points is essential.

[0003] Existing research on diving point warnings primarily uses a data-driven approach, using collected monitoring data, such as current and voltage, to train neural networks to predict diving points. However, this approach requires a large amount of historical data to train the neural network, and the neural network needs to be retrained for different battery types, making the process complex and difficult to interpret. Summary of the Invention

[0004] In response to the above-mentioned defects or improvement needs of the prior art, the present invention provides an online early warning method and system for the diving point of a lithium-ion battery. The purpose is to provide an online early warning method with physical explainability, which can realize online early warning of the battery diving point through a simple method.

[0005] To achieve the above objectives, according to one aspect of the present invention, there is provided an online early warning method for a lithium-ion battery water drop point, comprising:

[0006] Step S1: combining the capacity loss caused by solid electrolyte interface film growth and the capacity loss caused by lithium plating to construct a hybrid model that reflects the trend of battery capacity loss over time;

[0007] Step S2: obtaining the total discharge capacity loss of the battery in the early cycle as fitting data, wherein the early cycle is a period when lithium plating does not occur, and fitting the hybrid model using the fitting data to determine the parameters of the hybrid model;

[0008] Step S3: Using the output of the fitted hybrid model as the theoretical capacity loss caused by the solid electrolyte interface membrane, sliding the time window, obtaining the theoretical capacity loss and actual discharge capacity loss of different charge and discharge cycles in each time window, and calculating the slope of the linear correlation equation between the theoretical capacity loss and the actual discharge capacity loss to obtain a time series of the slope;

[0009] Step S4: Calculate the first-order autocorrelation coefficient of the slope. When the first-order autocorrelation coefficient reaches a minimum value, issue a price drop warning.

[0010] In some embodiments, in step S1, the capacity loss caused by the growth of the solid electrolyte interface film is expressed as:

[0011]

[0012] Where Q sei is the capacity loss caused by the growth of the solid electrolyte interface film, A sei,gr and B sei,gr are the unknown coefficients of the corresponding terms, and t is the battery charge and discharge cycle time.

[0013] In some embodiments, in step S1, the expression of the hybrid model is:

[0014]

[0015] Where t is the battery charge and discharge cycle time, A and B are the unknown parameters to be fitted.

[0016] In some embodiments, in step S2, obtaining the total discharge capacity loss of the battery in the early cycle as fitting data includes obtaining the total discharge capacity loss of the battery in the first 50 charge and discharge cycles as fitting data.

[0017] In some embodiments, in step S3, each of the time windows contains capacity loss data of a length of 100 charge-discharge cycles, and the moving step length of the time window is one charge-discharge cycle.

[0018] In some embodiments, in step S4, the formula for calculating the first-order autocorrelation coefficient is:

[0019]

[0020] Where ρ1 is the first-order autocorrelation coefficient, y t is the slope value of the slope signal at time t, is the mean slope of the slope signal, and T is the total number of observations of the slope signal.

[0021] In some embodiments, the battery is a lithium iron phosphate battery.

[0022] According to another aspect of the present invention, there is also provided an online early warning system for a lithium-ion battery diving point, comprising:

[0023] A hybrid model building unit is used to combine the capacity loss caused by solid electrolyte interface film growth and the capacity loss caused by lithium plating to build a hybrid model that reflects the trend of battery capacity loss over time;

[0024] a fitting unit, configured to obtain a total discharge capacity loss of the battery in an early cycle as fitting data, wherein the early cycle is a period when lithium plating does not occur, and fit the hybrid model using the fitting data to determine parameters of the hybrid model;

[0025] A slope calculation unit is used to use the output of the fitted hybrid model as the theoretical capacity loss caused by the solid electrolyte interface membrane, slide the time window, obtain the theoretical capacity loss and actual discharge capacity loss of different charge and discharge cycles in each time window, and calculate the slope of the linear correlation equation between the theoretical capacity loss and the actual discharge capacity loss to obtain a time series of the slope;

[0026] The correlation calculation unit is used to calculate the first-order autocorrelation coefficient of the slope. When the first-order autocorrelation coefficient reaches a minimum value, a plunge warning is issued.

[0027] According to another aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0028] According to yet another aspect of the present invention, a computer program product is provided, comprising a computer program or instructions, wherein when the computer program or instructions are executed by a processor, the steps of any of the above methods are implemented.

[0029] In general, the above technical solutions conceived by the present invention, compared with the prior art, provide an online early warning method and system for lithium-ion battery diving points, which have the following beneficial effects:

[0030] 1. The present invention constructs a hybrid model of battery capacity loss over time. The battery capacity loss is mainly caused by SEI film growth and lithium plating. Therefore, the hybrid model constructed by taking into account the capacity loss caused by the above two factors has a higher accuracy. The hybrid model is actually a relationship with unknown parameters. Therefore, the parameters of the hybrid model can be determined by obtaining the historical data of the battery for fitting. It is found that the actual capacity loss caused by SEI film growth is approximately linearly correlated with the total actual capacity loss. Using this rule, the present invention uses the early cycle data of lithium batteries to fit the hybrid model to determine the parameters. Since SEI film growth is the main factor causing capacity decay in the early cycles of lithium batteries, the output of the fitted hybrid model can be approximately considered as the theoretical capacity loss caused by SEI film growth. The slope of the linear correlation equation between the theoretical capacity loss output by the hybrid model and the actual capacity loss caused by SEI film growth in different time windows is calculated online, and the first-order autocorrelation coefficient of the slope obtained in adjacent time windows is calculated. Before lithium plating occurs, the theoretical capacity loss is linearly correlated with the capacity loss caused by actual SEI film growth. The first-order autocorrelation coefficient fluctuates or remains essentially unchanged within a certain interval. After lithium plating occurs, the correlation is reduced, thereby causing the first-order autocorrelation coefficient to decrease. Lithium plating is the main cause of capacity diving. Soon after lithium plating occurs, the battery capacity will experience a diving phenomenon. Therefore, the present invention uses the first-order autocorrelation coefficient reaching an extreme value as an early warning sign, and can achieve online early warning of the lithium-ion battery diving point by a simple method. Moreover, since the method is based on rigorous physical analysis, it also has strong physical interpretability.

[0031] 2. In a preferred embodiment, an empirical formula for capacity loss caused by SEI film growth is provided. This formula can more accurately reflect the capacity loss caused by SEI film growth, thereby improving the accuracy of the hybrid model.

[0032] 3. In a preferred embodiment, a specific expression of a hybrid model is provided, which can well fit the total capacity discharge data of the battery in the data set with high accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a flowchart of the steps of an online early warning method for a lithium-ion battery diving point in one embodiment of the present invention;

[0034] Figure 2 is the R of the fitting curve for fitting data of different data sets using a mixed model in one embodiment of the present invention. 2 Score diagram;

[0035] Figure 3 This is a diagram showing the effect of online warning of the diving point of different battery samples in one embodiment of the present invention. DETAILED DESCRIPTION

[0036] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0037] Example 1

[0038] like Figure 1 The figure shows a flowchart of the steps of the online early warning method for the diving point of a lithium-ion battery in one embodiment of the present invention, which mainly includes steps S1 to S4. The core steps are introduced in detail below.

[0039] Step S1: Combining the capacity loss caused by solid electrolyte interface film growth and the capacity loss caused by lithium plating, a hybrid model is constructed to reflect the time-varying trend of battery capacity loss.

[0040] After use, lithium-ion batteries will experience battery capacity loss, which is mainly caused by two factors: capacity loss caused by the growth of the solid electrolyte interface film (Solid Electrolyte Interphase, referred to as SEI) and capacity loss caused by lithium plating. Therefore, the present invention constructs a hybrid model that reflects the trend of battery capacity loss over time by comprehensively analyzing the capacity loss caused by the SEI film and the capacity loss caused by lithium plating. It should be noted that the hybrid model constructed at this time is related to an equation with battery capacity loss as the dependent variable and time as the independent variable, and the equation parameters are unknown.

[0041] In one embodiment, the lithium-ion battery may specifically be a lithium iron phosphate battery.

[0042] In one embodiment, considering that the capacity loss caused by the SEI film includes two parts: the capacity loss caused by the initial SEI film formation and subsequent growth, and the capacity loss caused by the SEI growth at the cracks, the empirical formulas for the capacity loss of the two parts can be constructed as follows:

[0043]

[0044] Where Q sei,gr is the capacity loss caused by the initial SEI film formation and subsequent growth, Q sei,re is the capacity loss caused by the growth of SEI at the cracks, t is the cycle time, i sei,re is the current density of SEI growth at the crack, A sei,gr and B sei,gr are the unknown coefficients of the corresponding terms.

[0045] By superimposing the above two losses, we can get the capacity loss caused by the SEI film:

[0046]

[0047] Where Q sei The capacity loss is caused by the SEI film.

[0048] In one embodiment, based on the pseudo two-dimensional model and the capacity loss model caused by SEI film growth, a capacity loss model caused by lithium plating is obtained, and the calculation formula is as follows:

[0049]

[0050] Among them, Q lp is the capacity loss caused by lithium plating, i lp is the current density corresponding to lithium plating, B lp is the unknown coefficient of the corresponding term.

[0051] Combining the capacity loss models caused by SEI film and lithium plating, a hybrid model that reflects the trend of battery capacity loss over time is obtained:

[0052]

[0053] Where Q tot is the total capacity loss of the lithium battery, A tot and B tot are the unknown coefficients of the corresponding terms. The above formula can be simplified into the following form:

[0054]

[0055] Where A and B are the unknown parameters to be fitted.

[0056] Specifically, the fitting ability of the hybrid model can be verified, that is, the hybrid model can be verified to fit the total capacity discharge data of the battery in the data set used. Specifically, R 2 The score is used as an evaluation indicator, such as Figure 2 The figure shows the R of the fitting curve of fitting different data sets using a mixed model in one embodiment. 2 Score diagram, the average fitting accuracy of the hybrid model is greater than 98%, which can well fit the total battery capacity discharge data in the dataset, indicating that the hybrid model is relatively accurate.

[0057] Furthermore, analysis revealed that the resulting hybrid model has an approximately linear correlation with the actual capacity loss caused by SEI film growth. The analysis process is as follows:

[0058]

[0059] It can be seen that the output of the hybrid model has an approximately linear correlation with the actual capacity loss.

[0060] Step S2: Obtain the total discharge capacity loss of the battery in the early cycle as fitting data, where the early cycle is a period when lithium plating does not occur. Use the fitting data to fit the hybrid model to determine the parameters of the hybrid model.

[0061] Since SEI film growth is the main factor causing capacity decay in the early cycles of lithium batteries, the present invention only uses the total discharge capacity loss of the battery in the early cycles as fitting data. The parameters A and B of the hybrid model are determined by fitting. The resulting fitting model can be approximately considered to be the theoretical capacity loss caused by SEI film growth. Before lithium plating occurs, the linear correlation between the theoretical capacity loss and the actual capacity loss caused by SEI film growth is highly correlated. When lithium plating occurs, the correlation between the theoretical capacity loss and the actual capacity loss caused by SEI film growth decreases. Therefore, the degree of lithium plating can be identified by judging the correlation of the linear correlation, thereby providing an online warning.

[0062] In a specific embodiment, the total discharge capacity data of the battery in the first 50 cycles may be fitted to obtain the parameters of the hybrid model.

[0063] Step S3: Using the output of the hybrid model as the theoretical capacity loss caused by the solid electrolyte interface membrane, sliding the time window, obtaining the theoretical capacity loss and actual discharge capacity loss of different charge and discharge cycles in each time window, and calculating the slope of the linear correlation equation between the theoretical capacity loss and the actual discharge capacity loss to obtain a time series of the slope.

[0064] In the present invention, a sliding window interception method is used to fix the length of the theoretical capacity loss vector and the actual discharge capacity vector. For example, the window length is 100 charge and discharge cycles, and the window moving step is one charge and discharge cycle, that is, each time a charge and discharge cycle is executed, the time window moves forward one step, and the theoretical capacity loss and the actual discharge capacity loss of different charge and discharge cycles in the current time window are obtained. The theoretical capacity loss can substitute the time corresponding to the charge and discharge cycle into the hybrid model to obtain the theoretical capacity loss of the current charge and discharge cycle. The actual discharge capacity loss can be directly obtained through online acquisition. When the data of at least two charge and discharge cycles (Q sei t1 ,Q tot t1 )、(Q sei t2 ,Q tot t2), where superscripts t1 and t2 are the times corresponding to the two charge and discharge cycles respectively, and the theoretical capacity loss Q is set. tot and Q sei Satisfy the linear correlation equation and calculate the slope k of the linear correlation equation of the current time window t2 , and so on, the slope of each window can be calculated to form a slope signal.

[0065] Step S4: Calculate the first-order autocorrelation coefficient of the slope. When the first-order autocorrelation coefficient reaches a minimum value, issue a price drop warning.

[0066] The first-order autocorrelation coefficient of the slope is calculated. The first-order autocorrelation coefficient reflects the correlation between the slopes before and after. Before lithium plating occurs, the theoretical capacity loss is linearly correlated with the capacity loss caused by actual SEI film growth. The first-order autocorrelation coefficient fluctuates within a certain range or remains essentially unchanged. When lithium plating occurs, the correlation decreases, causing the first-order autocorrelation coefficient to drop. Lithium plating is the main cause of capacity drop. Shortly after lithium plating occurs, the battery capacity will drop. Therefore, in the present invention, the extreme value of the first-order autocorrelation coefficient is used as an early warning sign, which can achieve online early warning of battery capacity drop and the solution is simple.

[0067] Among them, the existing formula can be used to calculate the first-order autocorrelation coefficient (Lag1 autocorrelation coefficient). The following introduces a specific feasible calculation formula:

[0068]

[0069] Where ρ1 is the first-order autocorrelation coefficient, y t is the slope value of the slope signal at time t, is the mean slope of the slope signal, and T is the total number of observations of the slope signal.

[0070] In order to verify the warning effect, 77 LFP / GrA123APR18650M1A batteries with a nominal capacity of 1.1Ah were used as samples. 77 different multi-stage discharge protocols were considered, but the same fast charging protocol was adopted. The online warning method disclosed in the present invention was used to perform early warning of the diving point. Figure 3 The figure shows the effect of online early warning of the diving point of different battery samples in one embodiment. It can be seen from the figure that the early warning method disclosed by the present invention can provide an early warning about 150 cycles before the battery actually dives. This shows that the online early warning of the diving point of the lithium battery of the present invention has a high early warning success rate and the early warning deviation is also within a reasonable range.

[0071] In summary, the online early warning method for the lithium-ion battery diving point mentioned in the present invention can provide online early warning of the occurrence of the lithium battery diving point, and the early warning signal has the characteristics of being able to be obtained in real time and having strong physical interpretability, filling the gap in online early warning of the lithium battery diving point. It can be widely used in battery health management systems to ensure the safe and reliable operation of lithium batteries.

[0072] Example 2

[0073] The present invention also relates to an online early warning system for a lithium-ion battery diving point, the system comprising:

[0074] A hybrid model building unit is used to combine the capacity loss caused by solid electrolyte interface film growth and the capacity loss caused by lithium plating to build a hybrid model that reflects the trend of battery capacity loss over time;

[0075] A fitting unit is used to obtain the total discharge capacity loss of the battery in the early cycle as fitting data, where the early cycle is a period when lithium plating does not occur, and to fit the hybrid model using the fitting data to determine the parameters of the hybrid model;

[0076] A slope calculation unit is used to use the output of the fitted hybrid model as the theoretical capacity loss caused by the solid electrolyte interface membrane, slide the time window, obtain the theoretical capacity loss and actual discharge capacity loss of different charge and discharge cycles in each time window, and calculate the slope of the linear correlation equation between the theoretical capacity loss and the actual discharge capacity loss to obtain a time series of the slope;

[0077] The correlation calculation unit is used to calculate the first-order autocorrelation coefficient of the slope. When the first-order autocorrelation coefficient reaches a minimum value, a plunge warning is issued.

[0078] In specific implementation, different functional units in the early warning system can implement the steps of the above method. For specific details, please refer to the above introduction and will not be repeated here.

[0079] Example 3

[0080] The present invention also relates to a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above method when the computer program is executed by a processor.

[0081] Specifically, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0082] Example 4

[0083] An embodiment of the present invention provides a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps of the method of the above embodiment of the present invention.

[0084] The technical features of the above embodiments can be combined in any manner. To simplify the description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification. It should be noted that the phrases "in one embodiment", "for example", "and another example", etc. of the present invention are intended to illustrate the present invention and are not intended to limit the present invention.

[0085] The above embodiments merely illustrate several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, and all such variations and improvements fall within the scope of protection of the present invention.

Claims

1. An online early warning method for lithium-ion battery diving point, characterized in that: include: Step S1: combining the capacity loss caused by solid electrolyte interface film growth and the capacity loss caused by lithium plating to construct a hybrid model that reflects the trend of battery capacity loss over time; Step S2: obtaining the total discharge capacity loss of the battery in the early cycle as fitting data, wherein the early cycle is a period when lithium plating does not occur, and fitting the hybrid model using the fitting data to determine the parameters of the hybrid model; Step S3: Using the output of the fitted hybrid model as the theoretical capacity loss caused by the solid electrolyte interface membrane, sliding the time window, obtaining the theoretical capacity loss and actual discharge capacity loss of different charge and discharge cycles in each time window, and calculating the slope of the linear correlation equation between the theoretical capacity loss and the actual discharge capacity loss to obtain a time series of the slope; Step S4: Calculate the first-order autocorrelation coefficient of the slope. When the first-order autocorrelation coefficient reaches a minimum value, issue a price drop warning.

2. The online early warning method for lithium-ion battery diving point according to claim 1, characterized in that: In step S1, the capacity loss caused by the growth of the solid electrolyte interface film is expressed as: Where, is the capacity loss caused by the growth of the solid electrolyte interface film. and are the unknown coefficients of the corresponding terms, t It is the battery charge and discharge cycle time.

3. The online early warning method for lithium-ion battery diving point according to claim 1, characterized in that: In step S1, the expression of the hybrid model is: Where, is the total capacity loss of the lithium battery, t is the battery charge and discharge cycle time, A 、 B are the unknown parameters to be fitted.

4. The online early warning method for lithium-ion battery diving point according to claim 1, characterized in that: In step S2, the obtaining of the total discharge capacity loss of the battery in the early cycle as fitting data includes obtaining the total discharge capacity loss of the battery in the first 50 charge and discharge cycles as fitting data.

5. The online early warning method for lithium-ion battery diving point according to claim 1, characterized in that: In step S3, each of the time windows contains capacity loss data of a length of 100 charge-discharge cycles, and the moving step length of the time window is one charge-discharge cycle.

6. The online early warning method for lithium-ion battery diving point according to claim 1, characterized in that: In step S4, the formula for calculating the first-order autocorrelation coefficient is: Where, is the first-order autocorrelation coefficient, is the slope signal in time t The slope value of is the slope mean of the slope signal, T is the total number of observations of the slope signal.

7. The online early warning method for lithium-ion battery diving point according to claim 1, characterized in that: The battery is a lithium iron phosphate battery.

8. An online early warning system for lithium-ion battery diving points, characterized in that: include: A hybrid model building unit is used to combine the capacity loss caused by solid electrolyte interface film growth and the capacity loss caused by lithium plating to build a hybrid model that reflects the trend of battery capacity loss over time; a fitting unit, configured to obtain a total discharge capacity loss of the battery in an early cycle as fitting data, wherein the early cycle is a period when lithium plating does not occur, and fit the hybrid model using the fitting data to determine parameters of the hybrid model; A slope calculation unit is used to use the output of the fitted hybrid model as the theoretical capacity loss caused by the solid electrolyte interface membrane, slide the time window, obtain the theoretical capacity loss and actual discharge capacity loss of different charge and discharge cycles in each time window, and calculate the slope of the linear correlation equation between the theoretical capacity loss and the actual discharge capacity loss to obtain a time series of the slope; The correlation calculation unit is used to calculate the first-order autocorrelation coefficient of the slope. When the first-order autocorrelation coefficient reaches a minimum value, a plunge warning is issued.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images