Method and module for determining lithium-plating mass of battery, and model training method and module
By extracting the electrochemical curve features during the battery charging and discharging process and utilizing machine learning models, the sensitivity and accuracy issues of battery lithium plating detection have been resolved, enabling non-destructive and accurate detection of battery lithium plating quality and supporting efficient monitoring of battery safety.
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- UNIV OF SCI & TECH OF CHINA
- Filing Date
- 2025-03-03
- Publication Date
- 2026-07-02
Smart Images
Figure CN2025080214_02072026_PF_FP_ABST
Abstract
Description
Methods and modules for determining lithium plating quality in batteries, as well as model training methods and modules. Technical Field
[0001] This disclosure relates to the field of power battery safety testing technology, and more specifically, to a method and module for determining the quality of lithium plating in a non-destructive battery, as well as a model training method and module. Background Technology
[0002] With the rapid development of new energy vehicles and distributed energy storage systems, lithium-ion batteries have gradually become a crucial energy device. However, battery safety has always been one of the factors restricting their widespread application. Therefore, the safety hazards caused by lithium plating inside batteries have gradually attracted widespread attention.
[0003] In realizing the concept disclosed herein, the inventors discovered that the related technologies for detecting lithium plating state have technical problems such as low sensitivity and low accuracy. Summary of the Invention
[0004] In view of this, this disclosure provides a method and module for determining the quality of lithium plating in batteries, as well as a model training method and module.
[0005] One aspect of this disclosure provides a method for determining the lithium plating quality of a battery, comprising: acquiring an electrochemical curve of a target battery during a charge / discharge process; extracting target electrochemical features from the electrochemical curve, wherein the target electrochemical features characterize electrochemical features related to lithium plating quality, and the target electrochemical features include time difference data and integral data corresponding to a predetermined current range in the electrochemical curve; inputting the target electrochemical features into a trained target lithium plating quality determination model to obtain the lithium plating quality of the target battery, wherein the lithium plating quality is the mass of metallic lithium deposited on the graphite surface when the electrochemical reduction rate and diffusion transport rate of lithium ions on the graphite surface do not match during the charge / discharge process of the target battery.
[0006] According to embodiments of this disclosure, the electrochemical curve includes a constant voltage current-time curve and a constant current voltage-time curve, and the predetermined current range includes a first current and a second current. Extracting target electrochemical features from the electrochemical curve includes: extracting a first time point corresponding to the first current and a second time point corresponding to the second current from the constant voltage current-time curve based on the first current and the second current; calculating the difference between the first time point and the second time point to obtain time difference data corresponding to the predetermined current range; and calculating integral data corresponding to the predetermined current range based on the first time point and the second time point.
[0007] Another aspect of this disclosure provides a model training method, including:
[0008] A sample database is acquired, comprising a sample electrochemical feature set and a sample lithium plating state set for multiple sample batteries corresponding to different predetermined aging conditions. The sample electrochemical feature set includes multiple sample electrochemical features, and the sample lithium plating state set includes multiple sample lithium plating state information. Using the multiple sample lithium plating state information and multiple sample electrochemical features, an initial model is trained to obtain the correlation information between each sample electrochemical feature and the multiple lithium plating state information, and an intermediate model. Based on the correlation information between each sample electrochemical feature and the multiple lithium plating state information, a target electrochemical feature is determined from the sample electrochemical feature set. Based on the target electrochemical feature, the intermediate model is trained to obtain a target lithium plating quality determination model, wherein the target lithium plating quality determination model is the aforementioned trained target lithium plating quality determination model.
[0009] According to embodiments of this disclosure, obtaining a sample database includes: performing cyclic charge / discharge treatment on multiple sample batteries at a first predetermined rate to obtain electrochemical curves of multiple sample batteries; extracting multiple sample electrochemical features from the electrochemical curves of multiple sample batteries to obtain a sample electrochemical feature set; analyzing multiple sample batteries to obtain a sample lithium plating state set; and obtaining a sample database based on the sample electrochemical feature set and the sample lithium plating state set.
[0010] According to embodiments of this disclosure, multiple sample batteries are analyzed to obtain a sample lithium plating state set, including: observing the negative electrode sheets of multiple sample batteries using scanning electron microscopy and nuclear magnetic resonance analysis to obtain lithium plating structure information; using titration gas chromatography to perform quantitative analysis on the negative electrode sheets of multiple sample batteries to obtain lithium plating weight information; and verifying the lithium plating weight information based on the lithium plating structure information to obtain a sample lithium plating state set.
[0011] According to embodiments of this disclosure, the initial battery includes a first initial battery, a second initial battery, a third initial battery, and a fourth initial battery. Multiple sample batteries are obtained through the following operations: maintaining the initial state of the first initial battery to obtain a first sample battery; performing cyclic overcharge / discharge simulation on the second initial battery at a second predetermined rate to obtain a second sample battery; performing cyclic charge / discharge simulation below the operating temperature on the third initial battery at a second predetermined rate to obtain a third sample battery; and performing cyclic overcharge / discharge simulation below the operating temperature on the fourth initial battery at a second predetermined rate to obtain a fourth sample battery.
[0012] According to embodiments of this disclosure, multiple sample electrochemical features include: voltage-related features of the constant current charging process, current-related features of the constant voltage charging process, diffusion resistance coefficient of the intermittent current interruption process, DC internal resistance, and charging cutoff voltage. The current-related features of the constant voltage charging process include time difference data and integral data corresponding to a predetermined current range. Based on the correlation information between each sample electrochemical feature and multiple lithium plating state information, a target electrochemical feature is determined from the sample electrochemical feature set. This includes: sorting the correlation information between each sample electrochemical feature and multiple lithium plating state information to obtain a sorting result; and based on the sorting result, selecting sample electrochemical features whose correlation information meets a predetermined threshold from the sample electrochemical feature set and determining them as target electrochemical features.
[0013] According to embodiments of this disclosure, an initial model is trained using multiple sample lithium plating state information and multiple sample electrochemical features to obtain correlation information between each sample electrochemical feature and multiple sample lithium plating state information and an intermediate model. This includes: splitting multiple sample lithium plating state information and multiple sample electrochemical features into a validation sample set and a training sample set according to a predetermined ratio; performing initial training on the initial model using the training sample set; and optimizing the parameters of the initially trained model using the validation sample set until the parameters meet predetermined conditions to obtain the intermediate model and correlation information.
[0014] Another aspect of this disclosure provides an apparatus for determining the quality of lithium plating in a battery, comprising:
[0015] The first acquisition module is used to acquire the electrochemical curve of the target battery during the charging / discharging process;
[0016] An extraction module is used to extract target electrochemical features from an electrochemical curve. The target electrochemical features characterize electrochemical features related to lithium plating quality, and the target electrochemical features include time difference data and integral data corresponding to a predetermined current range in the electrochemical curve.
[0017] The first module is used to input the target electrochemical characteristics into the trained target lithium plating quality determination model to obtain the lithium plating quality of the target battery. The lithium plating quality is the mass of metallic lithium deposited on the graphite surface when the electrochemical reduction rate and diffusion transport rate of lithium ions on the graphite surface do not match during the charging / discharging process of the target battery.
[0018] Another aspect of this disclosure provides a model training apparatus, comprising:
[0019] The second acquisition module is used to acquire a sample database, wherein the sample database includes a sample electrochemical feature set and a sample lithium plating state set of multiple sample batteries corresponding to different predetermined aging conditions. The sample electrochemical feature set includes multiple sample electrochemical features, and the sample lithium plating state set includes multiple sample lithium plating state information.
[0020] The training module is used to train the initial model using lithium plating state information and electrochemical features of multiple samples, and to obtain the correlation information between the electrochemical features of each sample and multiple lithium plating state information and the intermediate model.
[0021] The determination module is used to determine the target electrochemical feature from the sample electrochemical feature set based on the correlation information between the electrochemical features of each sample and multiple lithium plating state information.
[0022] The second module is used to train the intermediate model based on the target electrochemical characteristics to obtain the target lithium plating quality determination model, wherein the target lithium plating quality determination model is the target lithium plating quality determination model trained above.
[0023] According to embodiments of this disclosure, by extracting target electrochemical features from the electrochemical curves of the target battery during the charge / discharge process, and inputting the target electrochemical features into a trained target lithium plating quality determination model, the mass of metallic lithium deposited on the graphite surface of the target battery can be obtained. Since the lithium plating mass can be output simply by inputting the target electrochemical features into the model, it is not limited by the battery operating conditions and can obtain the specific lithium plating mass without damaging the battery. This solves the problem that related technologies can only identify a large amount of lithium plating after the middle stage, thereby improving the sensitivity and accuracy of detecting the lithium plating mass of the target battery. As a result, the obtained lithium plating mass can be used to make more accurate and efficient predictions or monitoring of the safety of the target battery. Attached Figure Description
[0024] Figure 1 schematically illustrates a flowchart of a method for determining the lithium plating quality of a battery according to an embodiment of the present disclosure.
[0025] Figure 2 schematically illustrates a flowchart of a model training method according to an embodiment of the present disclosure.
[0026] Figure 3 schematically illustrates a scanning electron microscope (SEM) view of the negative electrode of a sample battery in different lithium plating states according to an embodiment of the present disclosure.
[0027] Figure 4 schematically illustrates a schematic diagram of electrochemical curves according to an embodiment of the present disclosure.
[0028] Figure 5 schematically illustrates the diffusion resistance coefficients of sample cells in different lithium plating states according to embodiments of the present disclosure.
[0029] Figure 6 schematically illustrates the lithium plating quality of sample batteries under different lithium plating states before and after cycle testing according to embodiments of the present disclosure.
[0030] Figure 7 schematically illustrates sample batteries with different lithium plating states according to embodiments of the present disclosure. 7 Li spectrum.
[0031] Figure 8 schematically illustrates the hydrogen content of a sample battery according to an embodiment of the present disclosure.
[0032] Figure 9 schematically illustrates the experimental results of titration gas chromatography and balance weighing according to embodiments of the present disclosure.
[0033] Figure 10 schematically illustrates a diagram of the sorting results of associated information according to an embodiment of the present disclosure.
[0034] Figure 11 schematically illustrates a block diagram of an apparatus for determining battery lithium plating quality according to an embodiment of the present disclosure.
[0035] Figure 12 schematically illustrates a block diagram of an apparatus for determining battery lithium plating quality according to an embodiment of the present disclosure. Detailed Implementation
[0036] The embodiments of the present disclosure will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the disclosure. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the present disclosure for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.
[0037] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.
[0038] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.
[0039] When using expressions such as "at least one of A, B, and C", they should generally be interpreted in accordance with the meaning that is commonly understood by a person skilled in the art (e.g., "a system having at least one of A, B, and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B, and C, etc.).
[0040] In the embodiments disclosed herein, the collection, updating, analysis, processing, use, transmission, provision, disclosure, and storage of data (e.g., including but not limited to user personal information) comply with relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. In particular, necessary measures have been taken to prevent unauthorized access to user personal information data and to safeguard user personal information security and network security.
[0041] The lithium plating reaction in lithium-ion batteries differs from the conventional intercalation process. Even a small amount of lithium plating can cause capacity loss and increased internal resistance. As the lithium branches grow further, they may puncture the separator, leading to short circuits, fires, or even explosions. These safety incidents seriously affect the reliability of lithium batteries in practical applications and user safety.
[0042] To ensure the safety of lithium-ion batteries, extensive research has been conducted on battery safety monitoring and early warning technologies, primarily focusing on non-invasive in-situ physical characterization of lithium and safety characterization based on external electrical signals. In-situ physical characterization can be used to identify metallic lithium and the processes and mechanisms of lithium plating reactions in complex battery components, mainly utilizing techniques such as synchrotron X-ray diffraction, surface-enhanced Raman spectroscopy, electron paramagnetic resonance, and nuclear magnetic resonance. However, the instruments and equipment used for these in-situ physical characterizations are complex, potentially causing the battery structure to deviate from the actual battery structure used in practical applications. Furthermore, these characterization methods rely on costly experimental equipment and are difficult to implement online monitoring, making it challenging to meet the safety testing requirements of power batteries under complex operating conditions.
[0043] In addition, characterization based on external electrical signals, such as ohmic drop analysis, differential voltage analysis (DVA), relaxation voltage analysis (VRP), and incremental capacity analysis (ICA), are simpler, less expensive, and more portable than the aforementioned in-situ physical characterization, making them more suitable for industrial applications. However, they still have certain limitations. For example, they lack the ability to diagnose early lithium plating, only detecting mid-to-late-stage lithium plating, still relying on external voltage signals, and are difficult to effectively detect the lithium plating state inside the battery. Secondly, the repeatability of data testing is poor, and the accuracy is low. Test results are greatly affected by battery operating conditions, battery structure, and battery system type. At the same time, it is difficult to quantify lithium plating based on electrochemical characterization, and the detection sensitivity is low.
[0044] In view of this, embodiments of the present disclosure provide a method for determining the quality of lithium plating in a battery, including:
[0045] The electrochemical curves of the target battery during the charge / discharge process are obtained; target electrochemical features are extracted from the electrochemical curves, wherein the target electrochemical features characterize the electrochemical features related to the lithium plating quality, and the target electrochemical features include time difference data and integral data corresponding to the predetermined current range in the electrochemical curve; the target electrochemical features are input into the trained target lithium plating quality determination model to obtain the lithium plating quality of the target battery, wherein the lithium plating quality is the mass of metallic lithium deposited on the graphite surface when the electrochemical reduction rate and diffusion transport rate of lithium ions on the graphite surface do not match during the charge / discharge process of the target battery.
[0046] Figure 1 schematically illustrates a flowchart of a method for determining the lithium plating quality of a battery according to an embodiment of the present disclosure.
[0047] As shown in Figure 1, the method includes operations S110 to S130.
[0048] Using S110, electrochemical curves of the target battery during the charge / discharge process are obtained.
[0049] In operation S120, the target electrochemical characteristics are extracted from the electrochemical curve.
[0050] In operation S130, the target electrochemical characteristics are input into the trained target lithium plating quality determination model to obtain the lithium plating quality of the target battery.
[0051] According to embodiments of this disclosure, the electrochemical curves of the target battery during the charging / discharging process are obtained in real time by corresponding testing equipment, and may include constant voltage current-time curves and constant current voltage-time curves.
[0052] According to embodiments of this disclosure, the target electrochemical feature characterizes electrochemical features related to lithium plating quality, and the target electrochemical feature includes time difference data and integral data corresponding to a predetermined current range in the electrochemical curve, for example, the predetermined current range is 2.5A to 3.5A.
[0053] According to embodiments of this disclosure, the lithium plating reaction of the target battery differs from the conventional intercalation process. When the electrochemical reduction rate and diffusion transport rate of lithium ions on the graphite surface of the target battery do not match, lithium ions will be deposited on the graphite surface in the form of metallic lithium, and the lithium plating mass is the mass of metallic lithium produced by the lithium plating reaction.
[0054] According to embodiments of this disclosure, the target lithium plating quality determination model can be trained using a machine learning method, wherein the independent variables in the training process are various electrochemical features, including the target electrochemical features, and the dependent variable is the lithium plating quality.
[0055] According to embodiments of this disclosure, machine learning methods include, but are not limited to, supervised learning algorithms, unsupervised learning algorithms, and deep learning algorithms. Examples include: Support Vector Machines, deep neural networks, Adaptive Boosting, decision trees, and other models.
[0056] According to embodiments of this disclosure, for lithium-ion batteries, when the lithium plating quality increases, there is a risk that the generated lithium dendrites will puncture the battery separator, leading to a short circuit and causing a more serious thermal runaway problem. Simultaneously, the threshold temperature for thermal runaway will also decrease significantly. Therefore, the safety of the target battery can be assessed by determining the lithium plating quality output by the model based on the target lithium plating quality.
[0057] According to embodiments of this disclosure, by extracting target electrochemical features from the electrochemical curves of the target battery during the charge / discharge process, and inputting the target electrochemical features into a trained target lithium plating quality determination model, the mass of metallic lithium deposited on the graphite surface of the target battery can be obtained. Since the lithium plating mass can be output simply by inputting the target electrochemical features into the model, it is not limited by the battery operating conditions and can obtain the specific lithium plating mass without damaging the battery. This solves the problem that related technologies can only identify a large amount of lithium plating after the middle stage, thereby improving the sensitivity and accuracy of detecting the lithium plating mass of the target battery. As a result, the obtained lithium plating mass can be used to make more accurate and efficient predictions or monitoring of the safety of the target battery.
[0058] According to embodiments of this disclosure, the electrochemical curve includes a constant voltage current-time curve and a constant current voltage-time curve, and the predetermined current range includes a first current and a second current. Extracting target electrochemical features from the electrochemical curve includes: extracting a first time point corresponding to the first current and a second time point corresponding to the second current from the constant voltage current-time curve based on the first current and the second current; calculating the difference between the first time point and the second time point to obtain time difference data corresponding to the predetermined current range; and calculating integral data corresponding to the predetermined current range based on the first time point and the second time point.
[0059] For example, if the first current in the predetermined current range is 2.5A and the second current is 3.5A, according to the constant voltage current-time curve, the current is 2.5A at 3000s, that is, the first moment is 3000s, and the current is 3.5A at 3500s, that is, the first moment is 3500s. The difference between the first moment and the second moment is 500s, which is the time difference data. The curve within this range is integrated to obtain the integral data corresponding to the predetermined current range.
[0060] According to embodiments of this disclosure, taking the random forest model as an example, statistical methods can be used to convert the target electrochemical features into numerical vectors, such as calculating the maximum value, minimum value, mean, standard deviation, skewness, etc., or dimensionality reduction techniques (such as PCA) can be used to extract the main features.
[0061] Figure 2 schematically illustrates a flowchart of a model training method according to an embodiment of the present disclosure.
[0062] As shown in Figure 2, the method includes operations S210 to S240.
[0063] In operation S210, the sample database is obtained.
[0064] In operation S220, the initial model is trained using lithium plating state information and electrochemical features of multiple samples to obtain the correlation information between the electrochemical features of each sample and the lithium plating state information, as well as the intermediate model.
[0065] In operation S230, the target electrochemical feature is determined from the sample electrochemical feature set based on the correlation information between the electrochemical features of each sample and multiple lithium plating state information.
[0066] By operating the S240, the intermediate model is trained based on the target electrochemical characteristics to obtain a model for determining the target lithium plating quality.
[0067] According to embodiments of this disclosure, the sample database includes a sample electrochemical feature set and a sample lithium plating state set of multiple sample batteries corresponding to different predetermined aging conditions. The sample electrochemical feature set includes multiple sample electrochemical features, and the sample lithium plating state set includes multiple sample lithium plating state information.
[0068] According to embodiments of this disclosure, the sample battery may include batteries with different lithium plating states, such as: no lithium plating state, small amount of lithium plating state, medium lithium plating state, and severe lithium plating state.
[0069] To better illustrate the structure of the negative electrode of sample batteries in different lithium plating states, a comparison can be made according to Figure 3 below.
[0070] Figure 3 schematically illustrates a scanning electron microscope (SEM) view of the negative electrode of a sample battery in different lithium plating states according to an embodiment of the present disclosure.
[0071] As shown in Figure 3, the sample battery was disassembled, and the graphite negative electrode was removed. The structure of the negative electrode was characterized using a scanning electron microscope (SEM). The structural characteristics of the deposited lithium metal, including the morphology and distribution of the lithium plating, were observed using SEM. A represents a state with no lithium plating, B represents a state with a small amount of lithium plating, C represents a state with moderate lithium plating, and D represents a state with severe lithium plating. It is evident that the quality and structure of lithium plating differ among negative electrode sheets in different lithium plating states.
[0072] According to embodiments of this disclosure, the correlation information between the electrochemical characteristics of each sample and multiple lithium plating state information characterizes the importance of the sample electrochemical characteristics to the lithium plating state information. Different sample electrochemical characteristics have different predictive abilities for the lithium plating quality output by the target lithium plating quality determination model.
[0073] According to embodiments of this disclosure, the obtained initial sample electrochemical features can be preprocessed to remove missing and outlier values, thus obtaining the sample electrochemical features. Meanwhile, the scale differences between different sample electrochemical features are significant. To avoid the model being affected by these scale differences, multiple sample electrochemical features can be standardized, for example, by using zero-mean, unit variance, or normalization, such as scaling all features to the [0,1] interval, to ensure more stable and efficient model training.
[0074] According to embodiments of this disclosure, one-hot encoding or embedding should also be used to convert sample electrochemical features into numerical features so that machine learning algorithms can process them.
[0075] According to embodiments of this disclosure, the lithium plating state information of a sample may include lithium plating quality information and lithium plating structure information. The target electrochemical feature is the electrochemical feature that has the greatest influence on the lithium plating state information among all sample electrochemical features.
[0076] According to embodiments of this disclosure, the hyperparameters of intermediate models can be optimized through network search or random search, and the model can be verified and evaluated using the K-fold cross-validation method to avoid overfitting or underfitting.
[0077] According to embodiments of this disclosure, an initial model is trained using lithium plating state information and electrochemical features of multiple samples in a sample database. This yields the correlation information between the electrochemical features of each sample and the multiple lithium plating state information, as well as an intermediate model. The target electrochemical feature is then determined, and the intermediate model is trained based on the target electrochemical feature to obtain a target lithium plating quality determination model. This establishes the correlation between electrochemical features and lithium plating quality, allowing the lithium plating quality to be output simply by inputting the target electrochemical feature into the model. This method is not limited by battery operating conditions and does not require damage to the battery to obtain the specific lithium plating quality. It solves the problem that related technologies can only identify a large amount of lithium plating after the mid-term stage, thereby improving the sensitivity and accuracy of detecting the lithium plating quality of the target battery. Consequently, the obtained lithium plating quality can be used to more accurately and efficiently predict or monitor the safety of the target battery.
[0078] According to embodiments of this disclosure, obtaining a sample database includes: performing cyclic charge / discharge treatment on multiple sample batteries at a first predetermined rate to obtain electrochemical curves of multiple sample batteries; extracting multiple sample electrochemical features from the electrochemical curves of multiple sample batteries to obtain a sample electrochemical feature set; analyzing multiple sample batteries to obtain a sample lithium plating state set; and obtaining a sample database based on the sample electrochemical feature set and the sample lithium plating state set.
[0079] According to embodiments of this disclosure, multiple sample electrochemical characteristics include: voltage-related characteristics of the constant current charging process, current-related characteristics of the constant voltage charging process, diffusion resistance coefficient of the intermittent current interruption process, DC internal resistance, and charging cutoff voltage. The current-related characteristics of the constant voltage charging process include time difference data and integral data corresponding to a predetermined current range.
[0080] For example, the electrochemical characteristics of the constant current range include different voltage ranges (V i To V i+1 The charge (ΔQ) and the slope of the current curve (dV / dt(t)) i )), the integral area of the current curve (S) i and average voltage V a The electrochemical characteristics of the constant voltage range include different current ranges (I... i to I i+1 The time difference (Δt) and the slope of the voltage curve (dI / dt(t)) i )), voltage curve integral area (S) v and average current I a The diffusion resistance coefficient during the intermittent current interruption process is the diffusion resistance coefficient (k) obtained by intermittent current interruption method (ICI), the DC internal resistance (R) obtained by DC internal resistance test at near full charge (DCIR), and the charging cut-off voltage is the charging cut-off voltage (EOCV) after relaxation.
[0081] For example, the constant current charging rate can be 0.7C. When the battery is charged to 4.2V, an intermittent current interruption test is performed. This involves applying a charging current of 500mA to the sample battery for 5 minutes, followed by a 10-second relaxation period, and repeating this cycle until the voltage reaches its upper limit. The constant current charging cutoff voltage is 4.45V, followed by constant voltage charging until the current decreases to 100mA. A DC internal resistance test is then performed at full charge by continuously applying a current of 5A for 1 second, with a relaxation time of 5 minutes.
[0082] Figure 4 schematically illustrates a schematic diagram of electrochemical curves according to an embodiment of the present disclosure.
[0083] Figure 4 shows the electrochemical curves for the constant current charging stage, intermittent current interruption stage, constant voltage charging stage, and DC internal resistance testing stage, with the horizontal axis representing time and the vertical axis representing current or voltage. I0 a =S v / Δt v V a =S a / Δt i .
[0084] Figure 5 schematically illustrates the diffusion resistance coefficients of sample cells in different lithium plating states according to embodiments of the present disclosure.
[0085] As shown in Figure 5, the horizontal axis represents different lithium plating states, and the vertical axis represents the diffusion resistance coefficient.
[0086] According to embodiments of this disclosure, the sample electrochemical characteristics may further include the energy density of the battery obtained from the integral or gradient of the voltage-capacity curve, or the degradation rate and decay capacity extracted based on battery cycle charging data.
[0087] According to embodiments of this disclosure, the initial battery includes a first initial battery, a second initial battery, a third initial battery, and a fourth initial battery. Multiple sample batteries are obtained through the following operations: maintaining the initial state of the first initial battery to obtain a first sample battery; performing overcharge / discharge cycle simulation on the second initial battery at a second predetermined rate to obtain a second sample battery; performing charge / discharge cycle simulation below the operating temperature on the third initial battery at a second predetermined rate to obtain a third sample battery; and performing overcharge / discharge cycle simulation below the operating temperature on the fourth initial battery at a second predetermined rate to obtain a fourth sample battery.
[0088] According to embodiments of this disclosure, the initial battery can be multiple commercial power batteries with a capacity of 5Ah. The positive electrode material of the battery is lithium cobalt oxide and the negative electrode material is graphite. The initial battery includes multiple different lithium plating states.
[0089] According to embodiments of this disclosure, the first sample battery does not need to undergo aging simulation and can remain in its initial factory state.
[0090] According to embodiments of this disclosure, the overcharge voltage of the second sample battery can be 4.6V, 4.65V, 4.7V, etc.
[0091] According to embodiments of this disclosure, the predetermined rate can be 0.7C charging and 0.2C discharging. Accordingly, the electrochemical characteristics of the sample can be extracted from the electrochemical curve by performing an intermittent current interruption test near full charge during the constant current charging phase. First, the battery is charged at 500mA for 5 minutes, followed by a 10s relaxation period, and this cycle is repeated until the voltage reaches its upper limit. The voltage response to time is recorded to extract the lithium-ion diffusion resistance coefficient. After reaching 100% state of charge (SOC), a 5A current is applied for 1 second, and the DC internal resistance of the battery is measured. The battery is then left to stand for 5 minutes, and the open-circuit voltage is recorded.
[0092] According to embodiments of this disclosure, the number of charge / discharge cycles for each sample battery may include 0 cycles, 5 cycles, 10 cycles, and 30 cycles. Cyclic testing can be performed on batteries with different lithium plating states to verify the impact of charge / discharge cycles on the quality of lithium plating in the battery.
[0093] Figure 6 schematically illustrates the lithium plating quality of sample batteries under different lithium plating states before and after cycle testing according to embodiments of the present disclosure.
[0094] As shown in Figure 6, the horizontal axis represents the lithium plating state, and the vertical axis represents the lithium plating mass. The comparison shows that the lithium plating mass of the sample batteries in different lithium plating states remained essentially unchanged after cycle testing, indicating that cycle charging / discharging has a relatively small impact on the lithium plating mass.
[0095] According to embodiments of this disclosure, simulations of initial batteries in different lithium plating states under predetermined aging conditions are performed at a predetermined rate to obtain multiple sample batteries. By comparing the changes in lithium plating quality of the multiple sample batteries, the influence of predetermined aging conditions, i.e., battery health, on the lithium plating state can be obtained.
[0096] According to embodiments of this disclosure, multiple sample batteries are analyzed to obtain a sample lithium plating state set, including: observing the negative electrode sheets of multiple sample batteries using scanning electron microscopy and nuclear magnetic resonance analysis to obtain lithium plating structure information; using titration gas chromatography to perform quantitative analysis on the negative electrode sheets of multiple sample batteries to obtain lithium plating weight information; and verifying the lithium plating weight information based on the lithium plating structure information to obtain a sample lithium plating state set.
[0097] According to embodiments of this disclosure, the microstructure of the negative electrode of the sample battery can be observed first using a scanning electron microscope (SEM). Electrode sheets of 0.5 cm × 0.5 cm are randomly cut from electrodes in different lithium plating states to preliminarily analyze the growth of lithium plating on the negative electrode. The change in the proportion of lithium metal in lithium-containing species can be determined by solid-state nuclear magnetic resonance (NMR).
[0098] Figure 7 schematically illustrates sample batteries with different lithium plating states according to embodiments of the present disclosure. 7Li spectrum.
[0099] As shown in Figure 7, 3 mg to 4 mg of powder was scraped from the surface of the negative electrode sheet in different lithium plating states and its properties were tested. 7 Lithium spectra can be used to preliminarily determine the proportion of metallic lithium in lithium-containing species based on different chemical shifts, thus identifying the lithium deposition state. 7 The Li spectrum shows peaks corresponding to metallic lithium, as well as other peaks corresponding to lithium graphite precursors and lithium salts. The comparison shows that the peak values of the sample battery corresponding to the severe lithium plating state (D) are the highest.
[0100] According to embodiments of this disclosure, to further quantitatively analyze the "dead lithium" deposited in the sample battery, the sample battery can be disassembled in a glove box. The negative electrode and separator are cut into strips and divided into 10 equal portions, each placed in a 1L headspace vial, and sealed with a rubber stopper and aluminum ring to maintain a stable pressure of approximately 1.2 atm inside the glove box. 10 mL of water is added to each headspace vial using a syringe, and the mixture is thoroughly shaken to allow it to fully react with the electrode and separator. Then, 2 mL of gas is extracted using a sampling needle and injected into the inlet of a gas chromatograph to detect the hydrogen content.
[0101] According to embodiments of this disclosure, titration gas chromatography (TGC) is used to quantitatively analyze negative electrode samples of each lithium plating state to obtain the specific content range of lithium metal deposited under different lithium plating states, and multiple samples are measured for each lithium plating state to verify consistency.
[0102] Figure 8 schematically illustrates the hydrogen content of a sample battery according to an embodiment of the present disclosure.
[0103] As shown in Figure 8, the horizontal axis represents the lithium plating mass, and the vertical axis represents the detected hydrogen area. It can be seen that the higher the lithium plating mass, the larger the hydrogen area.
[0104] To verify the rationality of the TGC method in the quantitative analysis of lithium plating, the results of weighing the lithium plating mass using a balance can be compared with the quantitative results of the TGC method.
[0105] Figure 9 schematically illustrates the experimental results of titration gas chromatography and balance weighing according to embodiments of the present disclosure.
[0106] As shown in Figure 9, the horizontal axis represents the sample number of the sample battery, and the vertical axis represents the mass of lithium metal deposited. A comparison shows that the mass of lithium metal obtained by titration gas chromatography is approximately equal to the mass of lithium metal weighed by balance. Therefore, the TGC method has a certain degree of rationality in the quantitative analysis of lithium deposition.
[0107] According to embodiments of this disclosure, lithium plating structure information is obtained through nuclear magnetic resonance analysis, and lithium plating quality information is obtained through titration gas chromatography. Qualitative and quantitative analyses of sample batteries are achieved respectively, resulting in a more comprehensive sample database to ensure the accuracy of model training.
[0108] According to embodiments of this disclosure, a target electrochemical feature is determined from a set of sample electrochemical features based on the correlation information between each sample electrochemical feature and multiple lithium plating state information. This includes: sorting the correlation information between each sample electrochemical feature and multiple lithium plating state information to obtain a sorting result; and, based on the sorting result, selecting sample electrochemical features from the set of sample electrochemical features whose correlation information meets a predetermined threshold and determining them as target electrochemical features.
[0109] According to embodiments of this disclosure, the sample electrochemical characteristics may include 43 sample electrochemical characteristics as shown in Table 1 below.
[0110] Table 1
[0111] Figure 10 schematically illustrates a diagram of the sorting results of associated information according to an embodiment of the present disclosure.
[0112] As shown in Figure 10, among the ranking results of the correlation information of features 1 to 43, features 40 and 39 have the largest correlation information, so they can be used as target electrochemical features.
[0113] According to embodiments of this disclosure, by statistically analyzing and ranking the correlation information of different sample electrochemical characteristics, the target electrochemical characteristics can be determined. This allows for obtaining the sample electrochemical characteristics with the highest correlation to lithium plating quality. By utilizing the trained target lithium plating quality determination model and target electrochemical characteristics, the lithium plating quality of the target battery can be output, thereby assessing the safety of the target battery.
[0114] According to embodiments of this disclosure, an initial model is trained using multiple sample lithium plating state information and multiple sample electrochemical features to obtain correlation information between each sample electrochemical feature and multiple sample lithium plating state information and an intermediate model. This includes: splitting multiple sample lithium plating state information and multiple sample electrochemical features into a validation sample set and a training sample set according to a predetermined ratio; performing initial training on the initial model using the training sample set; and optimizing the parameters of the initially trained model using the validation sample set until the parameters meet predetermined conditions to obtain an intermediate model and correlation information.
[0115] According to embodiments of this disclosure, the predetermined ratio can be 2:8, that is, for 100% of the sample lithium plating state information and 100% of the sample electrochemical characteristics, it can be divided into 20% as the verification sample set and 80% as the training sample set.
[0116] According to embodiments of this disclosure, the performance of intermediate models can be evaluated through cross-validation or by using independent validation sample sets. Common evaluation metrics include mean squared error (MSE), coefficient of determination (R2), etc., until the evaluation metrics can reach a high and stable level, i.e., satisfy the predetermined conditions.
[0117] Figure 11 schematically illustrates a block diagram of an apparatus for determining battery lithium plating quality according to an embodiment of the present disclosure.
[0118] As shown in Figure 11, the device 1100 includes a first acquisition module 1110, an extraction module 1120, and a first obtaining module 1130.
[0119] The first acquisition module 1110 is used to acquire the electrochemical curve of the target battery during the charging / discharging process.
[0120] Extraction module 1120 is used to extract target electrochemical features from electrochemical curves, wherein the target electrochemical features characterize electrochemical features related to lithium plating quality, and the target electrochemical features include time difference data and integral data corresponding to a predetermined current range in the electrochemical curves.
[0121] The first module 1130 is used to input the target electrochemical characteristics into the trained target lithium plating quality determination model to obtain the lithium plating quality of the target battery. The lithium plating quality is the mass of metallic lithium deposited on the graphite surface when the electrochemical reduction rate and diffusion transport rate of lithium ions on the graphite surface do not match during the charging / discharging process of the target battery.
[0122] According to embodiments of this disclosure, the extraction module 1120 includes an extraction submodule, an acquisition submodule, and a calculation submodule.
[0123] The extraction submodule is used to extract the first moment corresponding to the first current and the second moment corresponding to the second current from the constant voltage current-time curve based on the first current and the second current.
[0124] A submodule is obtained to calculate the difference between the first and second time points, and to obtain time difference data corresponding to the predetermined current range.
[0125] The calculation submodule is used to calculate the integral data corresponding to the predetermined current range based on the first time point and the second time point.
[0126] Figure 12 schematically illustrates a block diagram of an apparatus for determining battery lithium plating quality according to an embodiment of the present disclosure.
[0127] As shown in Figure 12, the device 1200 includes a second acquisition module 1210, a training module 1220, a determination module 1230, and a second obtaining module 1240.
[0128] The second acquisition module 1210 is used to acquire a sample database, wherein the sample database includes a sample electrochemical feature set and a sample lithium plating state set of multiple sample batteries corresponding to different predetermined aging conditions. The sample electrochemical feature set includes multiple sample electrochemical features, and the sample lithium plating state set includes multiple sample lithium plating state information.
[0129] Training module 1220 is used to train the initial model using lithium plating state information and electrochemical features of multiple samples, and to obtain the correlation information between the electrochemical features of each sample and the lithium plating state information, as well as the intermediate model.
[0130] The determination module 1230 is used to determine the target electrochemical feature from the sample electrochemical feature set based on the correlation information between the electrochemical features of each sample and multiple lithium plating state information.
[0131] The second module 1240 is used to train the intermediate model based on the target electrochemical characteristics to obtain the target lithium plating quality determination model, wherein the target lithium plating quality determination model is the trained target lithium plating quality determination model of claim 1 or 2.
[0132] According to embodiments of this disclosure, the second acquisition module 1210 includes an acquisition submodule, an extraction submodule, a first obtaining submodule, and a second obtaining submodule.
[0133] The acquisition submodule is used to acquire electrochemical curves of multiple sample batteries.
[0134] The extraction submodule is used to extract multiple sample electrochemical features from the electrochemical curves of multiple sample batteries to obtain a sample electrochemical feature set.
[0135] The first submodule is used to analyze multiple sample batteries and obtain a set of lithium plating states for the samples.
[0136] The second submodule is used to obtain a sample database based on the sample electrochemical feature set and the sample lithium plating state set.
[0137] According to embodiments of this disclosure, the first obtaining submodule includes an analysis unit, a quantitative analysis unit, and a verification unit.
[0138] The analysis unit is used to perform scanning electron microscopy and nuclear magnetic resonance analysis on the negative electrode sheets of multiple sample batteries to obtain lithium plating structure information.
[0139] The quantitative analysis unit is used to perform quantitative analysis on the negative electrode of multiple sample batteries using titration gas chromatography technology to obtain lithium deposition weight information.
[0140] The verification unit is used to verify the lithium deposition weight information based on the lithium deposition structure information to obtain the sample lithium deposition state set.
[0141] According to embodiments of this disclosure, the determining module 1230 includes a sorting submodule and a determining submodule.
[0142] The sorting submodule is used to sort the correlation information between the electrochemical characteristics of each sample and multiple lithium plating states to obtain the sorting results.
[0143] The determination submodule is used to filter sample electrochemical features whose correlation information meets a predetermined threshold from the sample electrochemical feature set according to the sorting results, and determine them as target electrochemical features.
[0144] According to embodiments of this disclosure, training module 1220 includes a partitioning submodule, a training submodule, and an optimization submodule.
[0145] The sub-module is used to split the lithium plating state information and electrochemical characteristics of multiple samples into a validation sample set and a training sample set according to a predetermined ratio.
[0146] The training submodule is used to perform initial training on the initial model using the training sample set.
[0147] The optimization submodule is used to optimize the parameters of the initially trained model using the validation sample set until the parameters meet the predetermined conditions, thereby obtaining the intermediate model and related information.
[0148] Any one or more of the modules, submodules, units, and subunits according to embodiments of the present disclosure, or at least part of the functions of any one or more of them, can be implemented in one module. Any one or more of the modules, submodules, units, and subunits according to embodiments of the present disclosure can be implemented by dividing them into multiple modules. Any one or more of the modules, submodules, units, and subunits according to embodiments of the present disclosure can be at least partially implemented as hardware circuitry, such as Field Programmable Gate Arrays (FPGAs), Programmable Logic Arrays (PLAs), Systems-on-Chip, Systems-on-Substrate, Systems-on-Package, Application-Specific Integrated Circuits (ASICs), or implemented in hardware or firmware by any other reasonable means of integrating or packaging circuitry, or implemented in software, hardware, or firmware, or in any suitable combination of any of these three implementation methods. Alternatively, one or more of the modules, submodules, units, and subunits according to embodiments of the present disclosure can be at least partially implemented as computer program modules, which, when run, can perform corresponding functions.
[0149] For example, any plurality of the first acquisition module 1110, extraction module 1120, and first obtaining module 1130 can be combined into one module / unit / subunit, or any one of these modules / units / subunits can be split into multiple modules / units / subunits. Alternatively, at least part of the functionality of one or more of these modules / units / subunits can be combined with at least part of the functionality of other modules / units / subunits and implemented in one module / unit / subunit. According to embodiments of this disclosure, at least one of the first acquisition module 1110, extraction module 1120, and first obtaining module 1130 can be at least partially implemented as hardware circuitry, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or implemented in hardware or firmware by any other reasonable means of integrating or packaging the circuitry, or implemented in software, hardware, or firmware, or in any suitable combination of any of these three implementation methods. Alternatively, at least one of the first acquisition module 1110, extraction module 1120, and first obtaining module 1130 may be implemented at least partially as a computer program module, which can perform corresponding functions when the computer program module is run.
[0150] For example, any plurality of the second acquisition module 1210, training module 1220, determination module 1230, and second obtaining module 1240 can be combined into one module / unit / subunit, or any one of these modules / units / subunits can be split into multiple modules / units / subunits. Alternatively, at least part of the functionality of one or more of these modules / units / subunits can be combined with at least part of the functionality of other modules / units / subunits and implemented in one module / unit / subunit. According to embodiments of this disclosure, at least one of the second acquisition module 1210, training module 1220, determination module 1230, and second obtaining module 1240 can be at least partially implemented as hardware circuitry, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or any other reasonable means of integrating or packaging circuitry, or implemented in software, hardware, or firmware, or in any suitable combination of any of these three implementation methods. Alternatively, at least one of the second acquisition module 1210, training module 1220, determination module 1230, and second obtaining module 1240 may be implemented at least partially as a computer program module that can perform corresponding functions when the computer program module is run.
[0151] It should be noted that the device for determining the lithium plating quality in the embodiments of this disclosure corresponds to the method for determining the lithium plating quality in the embodiments of this disclosure. For a detailed description of the device for determining the lithium plating quality, please refer to the method for determining the lithium plating quality, which will not be repeated here. Similarly, the model training device in the embodiments of this disclosure corresponds to the model training method in the embodiments of this disclosure. For a detailed description of the model training device, please refer to the model training method, which will not be repeated here.
[0152] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions. Those skilled in the art will understand that the features described in the various embodiments of the present disclosure can be combined and / or combined in various ways, even if such combinations are not explicitly described in the present disclosure. In particular, the features described in the various embodiments of this disclosure may be combined and / or combined in various ways without departing from the spirit and teachings of this disclosure. All such combinations and / or combinations fall within the scope of this disclosure.
[0153] The embodiments of this disclosure have been described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of this disclosure. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of this disclosure, and all such substitutions and modifications should fall within the scope of this disclosure.
Claims
1. A method for determining the quality of lithium plating in a battery, characterized in that, include: Obtain the electrochemical curves of the target battery during the charge / discharge process; Target electrochemical features are extracted from the electrochemical curve, wherein the target electrochemical features characterize electrochemical features related to lithium plating quality, and the target electrochemical features include time difference data and integral data corresponding to a predetermined current range in the electrochemical curve; The target electrochemical characteristics are input into a trained target lithium plating quality determination model to obtain the lithium plating quality of the target battery. The lithium plating quality is the mass of metallic lithium deposited on the graphite surface when the electrochemical reduction rate and diffusion transport rate of lithium ions on the graphite surface do not match during the charging / discharging process of the target battery.
2. The method according to claim 1, characterized in that, The electrochemical curves include constant-voltage current-time curves and constant-current voltage-time curves, the predetermined current range includes a first current and a second current, and the extraction of target electrochemical features from the electrochemical curves includes: Based on the first current and the second current, extract the first moment corresponding to the first current and the second moment corresponding to the second current from the constant voltage current-time curve; Calculate the difference between the first time point and the second time point to obtain the time difference data corresponding to the predetermined current range; Based on the first time point and the second time point, calculate the integral data corresponding to the predetermined current range.
3. A model training method, characterized in that, include: Obtain a sample database, wherein the sample database includes a sample electrochemical feature set and a sample lithium plating state set of multiple sample batteries corresponding to different predetermined aging conditions, the sample electrochemical feature set includes multiple sample electrochemical features, and the sample lithium plating state set includes multiple sample lithium plating state information; Using the lithium plating state information and electrochemical features of the multiple samples, the initial model is trained to obtain the correlation information between the electrochemical features of each sample and the multiple lithium plating state information, and an intermediate model. Based on the correlation information between the electrochemical characteristics of each sample and the multiple lithium plating state information, the target electrochemical characteristics are determined from the set of sample electrochemical characteristics. Based on the target electrochemical characteristics, the intermediate model is trained to obtain a target lithium plating quality determination model, wherein the target lithium plating quality determination model is the trained target lithium plating quality determination model as described in claim 1 or 2.
4. The method according to claim 3, characterized in that, The acquisition of the sample database includes: The plurality of sample batteries were subjected to cyclic charge / discharge treatment at a first predetermined rate to obtain the electrochemical curves of the plurality of sample batteries. The electrochemical features of the multiple sample batteries are extracted from their electrochemical curves to obtain the sample electrochemical feature set. The lithium plating state set of the samples was obtained by analyzing the multiple sample batteries; The sample database is obtained based on the sample electrochemical feature set and the sample lithium plating state set.
5. The method according to claim 4, characterized in that, The analysis of the multiple sample batteries to obtain the lithium plating state set of the samples includes: The negative electrode sheets of the multiple sample batteries were observed by scanning electron microscopy and nuclear magnetic resonance analysis to obtain the lithium plating structure information; The negative electrode of the multiple sample batteries was quantitatively analyzed using titration gas chromatography to obtain the lithium deposition weight information. Based on the lithium plating structure information, the lithium plating weight information is verified to obtain the sample lithium plating state set.
6. The method according to claim 4, characterized in that, The initial cells include a first initial cell, a second initial cell, a third initial cell, and a fourth initial cell, and the plurality of sample cells are obtained through the following operations: Maintaining the initial state of the first initial battery, a first sample battery is obtained; According to the second predetermined rate, the second initial battery is subjected to overcharge / discharge cycle simulation to obtain the second sample battery; According to the second predetermined rate, the third initial battery is subjected to charge / discharge cycle simulation below the operating temperature to obtain the third sample battery; According to the second predetermined rate, the fourth initial battery is subjected to overcharge / discharge cycle simulation below the operating temperature to obtain the fourth sample battery.
7. The method according to claim 3, characterized in that, The multiple sample electrochemical features include: voltage-related features of the constant current charging process, current-related features of the constant voltage charging process, diffusion resistance coefficient, DC internal resistance, and charging cutoff voltage of the intermittent current interruption process. The current-related features of the constant voltage charging process include time difference data and integral data corresponding to a predetermined current range. The step of determining the target electrochemical feature from the sample electrochemical feature set based on the correlation information between each of the sample electrochemical features and the multiple lithium plating state information includes: The correlation information between the electrochemical characteristics of each sample and the multiple lithium plating state information is sorted to obtain the sorting result; Based on the sorting results, sample electrochemical features whose correlation information meets a predetermined threshold are selected from the sample electrochemical feature set and determined as the target electrochemical features.
8. The method according to claim 3, characterized in that, The process of training an initial model using the lithium plating state information and electrochemical features of the multiple samples to obtain the correlation information between each sample's electrochemical feature and the multiple lithium plating state information, and an intermediate model, includes: According to a predetermined ratio, the lithium plating state information and electrochemical characteristics of the multiple samples are split into a verification sample set and a training sample set; The initial model is initially trained using the training sample set; Using the validation sample set, the parameters of the initially trained model are optimized until the parameters meet predetermined conditions to obtain the intermediate model and the associated information.
9. A device for determining the quality of lithium plating in batteries, characterized in that, include: The first acquisition module is used to acquire the electrochemical curve of the target battery during the charging / discharging process; An extraction module is used to extract target electrochemical features from the electrochemical curve, wherein the target electrochemical features characterize electrochemical features related to lithium plating quality, and the target electrochemical features include time difference data and integral data corresponding to a predetermined current range in the electrochemical curve; The first obtaining module is used to input the target electrochemical characteristics into a trained target lithium plating quality determination model to obtain the lithium plating quality of the target battery. The lithium plating quality is the mass of metallic lithium deposited on the graphite surface when the electrochemical reduction rate and diffusion transport rate of lithium ions on the graphite surface do not match during the charging / discharging process of the target battery.
10. A model training device, characterized in that, include: The second acquisition module is used to acquire a sample database, wherein the sample database includes a sample electrochemical feature set and a sample lithium plating state set of multiple sample batteries corresponding to different predetermined aging conditions. The sample electrochemical feature set includes multiple sample electrochemical features, and the sample lithium plating state set includes multiple sample lithium plating state information. The training module is used to train the initial model using the lithium plating state information and electrochemical features of the multiple samples, so as to obtain the correlation information between the electrochemical features of each sample and the multiple lithium plating state information and the intermediate model. The determination module is used to determine the target electrochemical feature from the sample electrochemical feature set based on the correlation information between each sample electrochemical feature and the multiple lithium plating state information; The second obtaining module is used to train the intermediate model based on the target electrochemical characteristics to obtain a target lithium plating quality determination model, wherein the target lithium plating quality determination model is the trained target lithium plating quality determination model as described in claim 1 or 2.