Lithium ion battery health state estimation method and system based on mechanical information and medium

Abstract

The invention provides a lithium ion battery health state estimation method and system based on mechanical information and a medium, and the method comprises the steps: carrying out the full-life test of a battery cell of a lithium ion battery based on experimental parameters, and obtaining the experimental data of the battery cell; preprocessing the battery cell experiment data, constructing an expansion curve based on the optimized experiment data, extracting battery cell expansion characteristics, and analyzing battery cell response information; constructing a lithium ion battery health state prediction model, and analyzing lithium ion battery capacity attenuation state information according to the battery health state prediction model; analyzing battery health state information based on the lithium ion battery capacity attenuation state information, and analyzing battery performance according to the battery health state information to obtain performance evaluation information; the battery cell response information is calculated by analyzing the battery cell expansion curve, the lithium ion battery capacity attenuation state is analyzed according to the constructed lithium ion battery health state prediction model, the battery health state change is accurately analyzed, the battery performance is accurately analyzed, and the analysis precision is improved.

Description

Technical Field

[0001] The present application relates to the technical field of battery health assessment, and in particular to a method, system, and medium for estimating the health status of a lithium-ion battery based on mechanical information. Background Art

[0002] Existing lithium-ion battery state-of-health (SOH) estimation methods are primarily based on electrical signal features (such as voltage, current, and temperature), and are categorized into feature-driven and end-to-end learning approaches. Feature-driven approaches use commonly used features such as incremental capacity (ICV), differential temperature voltammetry (DTV), and the time constant of the constant voltage (CV) charging phase. While these features exhibit a certain degree of relevance, they are sensitive to operating condition variations in practical applications, resulting in poor generalization across different operating conditions. Furthermore, ICV and DTV features require numerical differentiation of the signal, resulting in poor noise immunity and difficulty in stable application in dynamic scenarios such as electric vehicles. Mechanical information holds particular promise for battery state-of-health (SOH) estimation, as it can directly reflect internal battery reaction processes, such as solid electrolyte interface (SEI) film growth and lithium metal precipitation (Li precipitation). Therefore, this application proposes a SOH estimation method based on expansion information extraction under a constant pressure environment. Its stability and accuracy at different temperatures and rates are systematically verified, addressing the issues of traditional electrical features being susceptible to operating condition perturbations and exhibiting poor generalization, while fully leveraging the advantage of mechanical information in intuitively reflecting internal aging reactions.

[0003] Existing SOH estimation methods based on electrical features have the following disadvantages: (1) The features are sensitive to charging current fluctuations and temperature changes, and have poor generalization capabilities; (2) Some features (such as ICV and DTV) require signal differentiation and have poor noise resistance; (3) Some studies have introduced mechanical features, but have not systematically verified their effectiveness and applicability. Summary of the Invention

[0004] The purpose of the embodiments of the present application is to provide a method, system and medium for estimating the health status of a lithium-ion battery based on mechanical information, by analyzing the battery cell expansion curve to calculate the battery cell response information, and analyzing the lithium-ion battery capacity attenuation state based on the constructed lithium-ion battery health status prediction model, thereby accurately analyzing the changes in the battery health status, and then accurately analyzing the battery performance, and improving the analysis accuracy.

[0005] The present application also provides a method for estimating the health status of a lithium-ion battery based on mechanical information, including:

[0006] Set experimental parameters, perform full life tests on lithium-ion battery cells based on the experimental parameters, and obtain cell experimental data;

[0007] Preprocess the cell experimental data to obtain optimized experimental data;

[0008] Construct an expansion curve based on optimized experimental data, extract the cell expansion characteristics based on the expansion curve, and analyze the cell response information based on the expansion characteristics;

[0009] Build a lithium-ion battery health status prediction model based on big data, and analyze the lithium-ion battery capacity attenuation status information according to the battery health status prediction model;

[0010] The battery health status information is analyzed based on the capacity attenuation status information of the lithium-ion battery, and the battery performance is analyzed according to the battery health status information to obtain performance evaluation information.

[0011] Optionally, in the method for estimating the health status of a lithium-ion battery based on mechanical information described in an embodiment of the present application, setting experimental parameters, and performing a full-life test on the battery cells of the lithium-ion battery based on the experimental parameters to obtain battery cell experimental data specifically includes:

[0012] Setting experimental parameters, including environmental parameters, mechanical parameters, and electrical performance test parameters; the environmental parameters include ambient temperature, ambient humidity, and light intensity; the mechanical parameters include pressure information, and the electrical performance test parameters include voltage, current, charge rate, and discharge rate;

[0013] Set different values ​​of environmental parameters, mechanical parameters and electrical performance test parameters to create multiple different operating conditions;

[0014] The battery cells are tested for their entire life cycle based on different operating conditions to obtain the battery cell experimental data.

[0015] Optionally, in the method for estimating the health status of a lithium-ion battery based on mechanical information described in an embodiment of the present application, preprocessing the cell experimental data to obtain optimized experimental data specifically includes:

[0016] Obtain battery cell experimental data, clean the battery cell experimental data, and obtain experimental data in a standardized format;

[0017] Extract the data features of the experimental data after format standardization, and filter out missing values ​​and outliers of the data features;

[0018] Eliminate the data features corresponding to outliers;

[0019] Analyze the previous moment value and the next moment value based on the linear interpolation method, and average the previous moment value and the next moment value to obtain the mean value;

[0020] Compare the mean with the missing value to obtain the missing difference, fill the missing value based on the missing difference, and obtain the optimized experimental data.

[0021] Optionally, in the method for estimating the health status of a lithium-ion battery based on mechanical information described in an embodiment of the present application, constructing an expansion curve based on optimized experimental data, extracting battery cell expansion characteristics based on the expansion curve, and analyzing battery cell response information based on the expansion characteristics specifically include:

[0022] Set the number of charge and discharge cycles for the battery cell, perform experiments on the battery cell, and obtain the battery cell displacement information;

[0023] Obtain the initial thickness of the battery cell, and divide the expansion displacement information by the initial thickness of the battery cell to obtain the expansion amount;

[0024] The expansion amount is normalized, and an expansion curve is drawn with the number of charge and discharge cycles as the horizontal axis and the normalized expansion amount as the vertical axis;

[0025] Extract the cell expansion characteristics based on the expansion curve, and analyze the relationship between the cell expansion rate during the charging stage, the cell expansion rate during the discharging stage, the cell collision volume during the full cycle, and the cell capacity based on the cell expansion characteristics;

[0026] The cell response information is obtained by the relationship between the cell expansion rate during the cell charging stage, the cell expansion rate during the cell discharging stage, the cell collision volume during the full cycle stage and the cell capacity.

[0027] Optionally, in the method for estimating the health status of a lithium-ion battery based on mechanical information described in an embodiment of the present application, a health status prediction model for a lithium-ion battery is constructed based on big data, and the capacity decay state information of the lithium-ion battery is analyzed according to the health status prediction model of the battery, specifically including:

[0028] Obtain historical lithium-ion battery health status prediction data based on big data and establish a data set;

[0029] Divide the data set into training and test sets;

[0030] Select an initial model framework, iteratively train the initial model framework based on the training set, and obtain training results;

[0031] Analyze whether the training results converge;

[0032] If convergence occurs, an intermediate model is obtained. The intermediate model is tested based on the test set, and the model hyperparameters are dynamically adjusted based on the test results to obtain a health status prediction model for the ion battery and predict the capacity attenuation status information of the ion battery.

[0033] If it does not converge, adjust the data in the training set or the number of iterations.

[0034] Optionally, in the method for estimating the health status of a lithium-ion battery based on mechanical information described in an embodiment of the present application, analyzing the battery health status information based on the capacity decay state information of the lithium-ion battery, and analyzing the battery performance based on the battery health status information to obtain performance evaluation information specifically includes:

[0035] Set the analysis time window and divide it into several sub-windows in equal proportions;

[0036] The starting time of each sub-window is used as the acquisition time to obtain the acquisition time nodes corresponding to several sub-windows;

[0037] Collect the lithium-ion battery capacity decay status information corresponding to different time nodes, compare the lithium-ion battery capacity decay status information of adjacent time nodes, and obtain the battery capacity decay rate of adjacent sub-windows;

[0038] Analyze battery attenuation rate fluctuation information based on the battery capacity attenuation rate of adjacent sub-windows;

[0039] The battery health status information is analyzed based on the battery attenuation rate fluctuation information, the battery performance within the time window is analyzed based on the battery health status information, and the battery performance is evaluated based on the evaluation indicators to obtain performance evaluation information. The evaluation indicators are set according to the battery model and battery factory data.

[0040] In a second aspect, an embodiment of the present application provides a system for estimating the health status of a lithium-ion battery using mechanical information. The system includes: a memory and a processor. The memory includes a program for estimating the health status of a lithium-ion battery using mechanical information. When the program is executed by the processor, the following steps are implemented:

[0041] Set experimental parameters, perform full life tests on lithium-ion battery cells based on the experimental parameters, and obtain cell experimental data;

[0042] Preprocess the cell experimental data to obtain optimized experimental data;

[0043] Construct an expansion curve based on optimized experimental data, extract the cell expansion characteristics based on the expansion curve, and analyze the cell response information based on the expansion characteristics;

[0044] Build a lithium-ion battery health status prediction model based on big data, and analyze the lithium-ion battery capacity attenuation status information according to the battery health status prediction model;

[0045] The battery health status information is analyzed based on the capacity attenuation status information of the lithium-ion battery, and the battery performance is analyzed according to the battery health status information to obtain performance evaluation information.

[0046] Optionally, in the system for estimating the health status of a lithium-ion battery based on mechanical information described in an embodiment of the present application, setting experimental parameters, performing a full-life test on the cells of the lithium-ion battery based on the experimental parameters, and obtaining cell experimental data specifically include:

[0047] Setting experimental parameters, including environmental parameters, mechanical parameters, and electrical performance test parameters; the environmental parameters include ambient temperature, ambient humidity, and light intensity; the mechanical parameters include pressure information, and the electrical performance test parameters include voltage, current, charge rate, and discharge rate;

[0048] Set different values ​​of environmental parameters, mechanical parameters and electrical performance test parameters to create multiple different operating conditions;

[0049] The battery cells are tested for their entire life cycle based on different operating conditions to obtain the battery cell experimental data.

[0050] Optionally, in the lithium-ion battery health status estimation system based on mechanical information according to an embodiment of the present application, preprocessing the cell experimental data to obtain optimized experimental data specifically includes:

[0051] Obtain battery cell experimental data, clean the battery cell experimental data, and obtain experimental data in a standardized format;

[0052] Extract the data features of the experimental data after format standardization, and filter out missing values ​​and outliers of the data features;

[0053] Eliminate the data features corresponding to outliers;

[0054] Analyze the previous moment value and the next moment value based on the linear interpolation method, and average the previous moment value and the next moment value to obtain the mean value;

[0055] Compare the mean with the missing value to obtain the missing difference, fill the missing value based on the missing difference, and obtain the optimized experimental data.

[0056] In a third aspect, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium includes a program for a method for estimating the health status of a lithium-ion battery based on mechanical information. When the program for estimating the health status of a lithium-ion battery based on mechanical information is executed by a processor, the steps of the method for estimating the health status of a lithium-ion battery based on mechanical information as described in any one of the above items are implemented.

[0057] As can be seen from the above, the embodiments of the present application provide a method, system and medium for estimating the health status of a lithium-ion battery based on mechanical information. By setting experimental parameters, the battery cells of the lithium-ion battery are subjected to full-life testing based on the experimental parameters to obtain battery cell experimental data; the battery cell experimental data are preprocessed to obtain optimized experimental data; an expansion curve is constructed based on the optimized experimental data, the battery cell expansion characteristics are extracted based on the expansion curve, and the battery cell response information is analyzed based on the expansion characteristics; a lithium-ion battery health status prediction model is constructed based on big data, and the lithium-ion battery capacity decay state information is analyzed according to the battery health status prediction model; the battery health status information is analyzed based on the lithium-ion battery capacity decay state information, and the battery performance is analyzed according to the battery health status information to obtain performance evaluation information; the battery cell response information is calculated by analyzing the battery cell expansion curve, and the lithium-ion battery capacity decay state is analyzed according to the constructed lithium-ion battery health status prediction model, so as to accurately analyze the changes in the battery health status, and then accurately analyze the battery performance, thereby improving the analysis accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without creative work.

[0059] Figure 1 A flowchart of a method for estimating the health status of a lithium-ion battery using mechanical information provided in an embodiment of the present application;

[0060] Figure 2 A schematic diagram of the cell capacity-cycle charge-discharge evolution of the lithium-ion battery health status estimation method based on mechanical information provided in an embodiment of the present application;

[0061] Figure 3 A comparison chart of the expansion characteristics of the lithium-ion battery health status estimation method based on mechanical information provided in an embodiment of the present application and the traditional characteristics SOH prediction. DETAILED DESCRIPTION

[0062] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work fall within the scope of protection of the present application.

[0063] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and should not be understood as indicating or implying relative importance.

[0064] Please refer to Figure 1-Figure 3 As shown, the method for estimating the health status of a lithium-ion battery using mechanical information is used in a terminal device, and the method for estimating the health status of a lithium-ion battery using mechanical information includes the following steps:

[0065] S101, setting experimental parameters, performing a full life test on the lithium-ion battery cell based on the experimental parameters, and obtaining battery cell experimental data;

[0066] S102, preprocessing the battery cell experimental data to obtain optimized experimental data;

[0067] S103, constructing an expansion curve based on the optimized experimental data, extracting battery cell expansion characteristics based on the expansion curve, and analyzing battery cell response information based on the expansion characteristics;

[0068] S104, constructing a lithium-ion battery health status prediction model based on big data, and analyzing lithium-ion battery capacity attenuation status information according to the battery health status prediction model;

[0069] S105 , analyzing battery health status information based on the lithium-ion battery capacity decay status information, and analyzing battery performance according to the battery health status information to obtain performance evaluation information.

[0070] It should be noted that this dataset uses a publicly available soft-pack battery experimental dataset. This dataset tests 12 soft-pack batteries throughout their lifecycle at a constant pressure (34.5kPa). This dataset covers charging processes at different temperatures (-5°C, 25°C, and 45°C) and at different charging rates (C / 5-2C). The dataset records the expansion curves, current, voltage, and temperature changes of the batteries throughout their entire lifecycle.

[0071] Feature construction: Based on the expansion curve, a variety of statistical features and peak-valley features are extracted to fully reflect the aging response behavior of the battery cell, including:

[0072] Minimum expansion value (min): reflects the expansion degree of irreversible reaction inside the battery cell (such as SEI film, lithium deposition);

[0073] Expansion difference (delta): reflects the expansion amplitude of the reversible intercalation process;

[0074] Mean, standard deviation (SD), root mean square (RMS), and skew: describe the overall level of volatility and symmetry;

[0075] Tail: The residual expansion at the end of the cycle.

[0076] ICV, CV and DTV also extract key features, among which incremental capacity curves (ICV): incremental capacity curve;

[0077] constant voltage (CV): constant voltage curve;

[0078] differential thermal voltammetry (DTV): differential thermal voltammetry curve.

[0079] Modeling: The LightGBM regression model is used to build the SOH prediction model, integrating multiple feature inputs for training.

[0080] Performance Evaluation: MAPE (Mean Absolute Percentage Error) and RMSE (Root Mean Square Error) were used to analyze the model's generalization capabilities under different operating conditions (temperature and current rate). The results showed that the expansion feature outperformed traditional ICV, CV, and DTV features in both relevance and prediction accuracy.

[0081] The expansion response is used to sensitively reflect the microstructural changes produced by the internal reaction of lithium ions, thereby improving the response ability to SOH.

[0082] According to an embodiment of the present invention, experimental parameters are set, and a full life test is performed on a lithium-ion battery cell based on the experimental parameters to obtain cell experimental data, specifically including:

[0083] Set the experimental parameters, including environmental parameters, mechanical parameters, and electrical performance test parameters; environmental parameters include ambient temperature, ambient humidity, and light intensity; mechanical parameters include pressure information, and electrical performance test parameters include voltage, current, charge rate, and discharge rate;

[0084] Set different values ​​of environmental parameters, mechanical parameters and electrical performance test parameters to create multiple different operating conditions;

[0085] The battery cells are tested for their entire life cycle based on different operating conditions to obtain the battery cell experimental data.

[0086] It should be noted that by setting different experimental parameters and establishing different operating conditions, the battery cells can be tested for their entire life under different operating conditions, thereby improving the test accuracy and ensuring the integrity of the battery cell experimental data.

[0087] According to an embodiment of the present invention, the cell experimental data is preprocessed to obtain optimized experimental data, specifically including:

[0088] Obtain battery cell experimental data, clean the battery cell experimental data, and obtain experimental data in a standardized format;

[0089] Extract the data features of the experimental data after format standardization, and filter out missing values ​​and outliers of the data features;

[0090] Eliminate the data features corresponding to outliers;

[0091] Analyze the previous moment value and the next moment value based on the linear interpolation method, and average the previous moment value and the next moment value to obtain the mean value;

[0092] Compare the mean with the missing value to obtain the missing difference, fill the missing value based on the missing difference, and obtain the optimized experimental data.

[0093] It should be noted that cell test data is often affected by noise from test equipment, fluctuations in operating conditions, and sensor errors, which directly affect the accuracy of SOH estimation. Preprocessing improves the data signal-to-noise ratio by removing outliers, correcting biases, and standardizing features, preventing "dirty data" from leading to model misjudgments. Data formats collected under different operating conditions and equipment are unified to enhance comparability. By cleaning cell test data, filling in missing values, and removing outliers, the accuracy of cell test data reflection is improved, thereby enhancing the accuracy of cell health assessments.

[0094] According to an embodiment of the present invention, constructing an expansion curve based on optimized experimental data, extracting battery cell expansion characteristics based on the expansion curve, and analyzing battery cell response information based on the expansion characteristics specifically include:

[0095] Set the number of charge and discharge cycles for the battery cell, perform experiments on the battery cell, and obtain the battery cell displacement information;

[0096] Obtain the initial thickness of the battery cell, and divide the expansion displacement information by the initial thickness of the battery cell to obtain the expansion amount;

[0097] The expansion amount is normalized, and an expansion curve is drawn with the number of charge and discharge cycles as the horizontal axis and the normalized expansion amount as the vertical axis;

[0098] Extract the cell expansion characteristics based on the expansion curve, and analyze the relationship between the cell expansion rate during the charging stage, the cell expansion rate during the discharging stage, the cell collision volume during the full cycle, and the cell capacity based on the cell expansion characteristics;

[0099] The cell response information is obtained by the relationship between the cell expansion rate during the cell charging stage, the cell expansion rate during the cell discharging stage, the cell collision volume during the full cycle stage and the cell capacity.

[0100] It should be noted that by cyclically charging and discharging the battery cells, the expansion of the battery cells during the cyclic charging and discharging process is analyzed to analyze the expansion response of the battery cells, providing an effective basis for subsequent analysis of the battery health status based on the battery cell expansion.

[0101] Furthermore, mechanical characteristics (such as expansion rate δ h ) is linearly related to SOH, and the formula is as follows:

[0102] SOH=k·δ h +c+∈;

[0103] k and c represent regression coefficients (fitted by the least squares method); ∈ represents the error term (caused by interference such as temperature and magnification).

[0104] The irreversible expansion during battery cycling is an important mechanical characteristic, often described by the thickness expansion rate:

[0105]

[0106] δ h represents the thickness expansion rate after the nth cycle; h n represents the battery thickness after the nth cycle; h0 represents the initial thickness.

[0107] Furthermore, the expansion curve describes the following parameters:

[0108] Instantaneous expansion rate (under a certain state): used to describe the expansion degree of the battery at a specific state of charge (SOC) or number of cycles:

[0109]

[0110] L(t) represents the thickness (or volume) at a certain moment / state; L0 represents the initial thickness (or volume, usually the state when SOC=0); ΔL represents the thickness (or volume) change.

[0111] Irreversible expansion rate (cycle accumulation): describes the net expansion that cannot be recovered during the cycle process and is a key indicator for evaluating battery aging:

[0112]

[0113] L n (SOC=0) represents the thickness at the end of the nth cycle (discharge to SOC=0). This value increases monotonically with the number of cycles and is negatively correlated with SOH (the higher the expansion rate, the lower the SOH).

[0114] Reversible expansion amplitude (single charge and discharge): describes the reversible volume fluctuation during the charge and discharge process, reflecting the reversible lithium insertion ability of the active material. It usually decreases after aging. The specific formula is as follows:

[0115] Reversible expansion amplitude = L max -L min ;

[0116] L max Indicates the maximum thickness when single charging to SOC = 100%;

[0117] L min Indicates the minimum thickness when a single discharge is performed to SOC=0.

[0118] According to an embodiment of the present invention, a lithium-ion battery health status prediction model is constructed based on big data, and the lithium-ion battery capacity decay status information is analyzed according to the battery health status prediction model, specifically including:

[0119] Obtain historical lithium-ion battery health status prediction data based on big data and establish a data set;

[0120] Divide the data set into training and test sets;

[0121] Select an initial model framework, iteratively train the initial model framework based on the training set, and obtain training results;

[0122] Analyze whether the training results converge;

[0123] If convergence occurs, an intermediate model is obtained. The intermediate model is tested based on the test set, and the model hyperparameters are dynamically adjusted based on the test results to obtain a health status prediction model for the ion battery and predict the capacity attenuation status information of the ion battery.

[0124] If it does not converge, adjust the data in the training set or the number of iterations.

[0125] It should be noted that the model is iteratively trained by obtaining historical lithium-ion battery health status prediction data through big data, so that the model has learning ability and improves the accuracy of model output.

[0126] Specifically, SOH usually takes capacity decay as the core indicator and is defined as the ratio of current capacity to initial capacity. The formula is as follows:

[0127]

[0128] Q current Indicates the rated capacity of the battery under the current cycle (unit: Ah); Q initial Indicates the initial rated capacity of the battery when it leaves the factory.

[0129] According to an embodiment of the present invention, the battery health status information is analyzed based on the lithium-ion battery capacity decay status information, and the battery performance is analyzed according to the battery health status information to obtain performance evaluation information, specifically including:

[0130] Set the analysis time window and divide it into several sub-windows in equal proportions;

[0131] The starting time of each sub-window is used as the acquisition time to obtain the acquisition time nodes corresponding to several sub-windows;

[0132] Collect the lithium-ion battery capacity decay status information corresponding to different time nodes, compare the lithium-ion battery capacity decay status information of adjacent time nodes, and obtain the battery capacity decay rate of adjacent sub-windows;

[0133] Analyze battery attenuation rate fluctuation information based on the battery capacity attenuation rate of adjacent sub-windows;

[0134] The battery health status information is analyzed based on the battery attenuation rate fluctuation information, the battery performance within the time window is analyzed based on the battery health status information, and the battery performance is evaluated based on the evaluation indicators to obtain performance evaluation information. The evaluation indicators are set according to the battery model and battery factory data.

[0135] It should be noted that by dividing the time window into equal intervals, analyzing the battery capacity decay status information at different acquisition time nodes within the time window, and analyzing the battery decay fluctuation information, the battery health status can be accurately evaluated.

[0136] Specifically, the battery attenuation rate is a key indicator to measure the rate at which the capacity of a lithium-ion battery decreases with usage or number of cycles. Its calculation formula needs to combine the time dimension of attenuation (such as number of cycles, usage time) and the capacity change.

[0137] The battery attenuation rate includes: the capacity attenuation rate based on the number of cycles, which is used to describe the average capacity attenuation ratio per cycle. The formula is:

[0138]

[0139] Q initial Indicates the initial rated capacity of the battery (unit: Ah or mAh);

[0140] Q n Indicates the actual capacity after the nth cycle;

[0141] n represents the number of cycles (such as 100 times, 500 times).

[0142] For example, if the initial capacity of a battery is 2000mAh and the capacity drops to 1600mAh after 500 cycles, the attenuation rate is calculated as follows:

[0143]

[0144] That is, the capacity decays by an average of 0.04% per cycle.

[0145] Capacity decay rate based on usage time: For batteries that are stationary or stored for a long time, the decay rate needs to be calculated based on time (such as years or months). The formula is as follows:

[0146]

[0147] Q t It represents the actual capacity after t years of use;

[0148] t represents the usage time (unit: year, month, etc., which must be consistent with the numerator unit).

[0149] For example, if the initial capacity of a battery is 2000mAh and the capacity is 1880mAh after 2 years of storage, the attenuation rate is calculated as follows:

[0150]

[0151] The cumulative decay rate of the battery from the initial state to the nth cycle is:

[0152]

[0153] For example, if the capacity is 80% of the initial value after 1000 cycles, the cumulative attenuation rate is 1-80%=20%.

[0154] The relationship between attenuation rate and SOH:

[0155] The battery state of health (SOH) is directly related to the decay rate, as shown in the following formula:

[0156] SOH=(1-cumulative decay rate)×100%, that is, when the cumulative decay rate is 20%, SOH=80%.

[0157] In summary, the attenuation nonlinear characteristics: In practice, the battery attenuation rate is not constant (slow in the early stage, stable in the middle stage, and accelerated in the later stage). The above formula is the average attenuation rate, which needs to be analyzed in combination with the specific cycle stage (for example, the attenuation rate in the first 100 cycles may be lower than 0.02% / cycle, and may rise to more than 0.1% / cycle in the later stage).

[0158] Environmental impact: High temperature (such as above 45°C) will accelerate attenuation. The test conditions (such as temperature and charge and discharge rate) must be specified during calculations, and correction factors (such as temperature correction factor α(T)) should be introduced if necessary.

[0159] Capacity test standard: Battery capacity must be measured according to the standard process (such as 0.5C charging to full charge and 0.2C discharging to the cut-off voltage) to ensure data consistency.

[0160] In a second aspect, an embodiment of the present application provides a system for estimating the health status of a lithium-ion battery using mechanical information. The system includes: a memory and a processor. The memory includes a program for estimating the health status of a lithium-ion battery using mechanical information. When the program is executed by the processor, the following steps are implemented:

[0161] Set experimental parameters, perform full life tests on lithium-ion battery cells based on the experimental parameters, and obtain cell experimental data;

[0162] Preprocess the cell experimental data to obtain optimized experimental data;

[0163] Construct an expansion curve based on optimized experimental data, extract the cell expansion characteristics based on the expansion curve, and analyze the cell response information based on the expansion characteristics;

[0164] Build a lithium-ion battery health status prediction model based on big data, and analyze the lithium-ion battery capacity attenuation status information according to the battery health status prediction model;

[0165] The battery health status information is analyzed based on the capacity attenuation status information of the lithium-ion battery, and the battery performance is analyzed according to the battery health status information to obtain performance evaluation information.

[0166] According to an embodiment of the present invention, experimental parameters are set, and a full life test is performed on a lithium-ion battery cell based on the experimental parameters to obtain cell experimental data, specifically including:

[0167] Set the experimental parameters, including environmental parameters, mechanical parameters, and electrical performance test parameters; environmental parameters include ambient temperature, ambient humidity, and light intensity; mechanical parameters include pressure information, and electrical performance test parameters include voltage, current, charge rate, and discharge rate;

[0168] Set different values ​​of environmental parameters, mechanical parameters and electrical performance test parameters to create multiple different operating conditions;

[0169] The battery cells are tested for their entire life cycle based on different operating conditions to obtain the battery cell experimental data.

[0170] According to an embodiment of the present invention, the cell experimental data is preprocessed to obtain optimized experimental data, specifically including:

[0171] Obtain battery cell experimental data, clean the battery cell experimental data, and obtain experimental data in a standardized format;

[0172] Extract the data features of the experimental data after format standardization, and filter out missing values ​​and outliers of the data features;

[0173] Eliminate the data features corresponding to outliers;

[0174] Analyze the previous moment value and the next moment value based on the linear interpolation method, and average the previous moment value and the next moment value to obtain the mean value;

[0175] Compare the mean with the missing value to obtain the missing difference, fill the missing value based on the missing difference, and obtain the optimized experimental data.

[0176] It should be noted that cell test data is often affected by noise from test equipment, fluctuations in operating conditions, and sensor errors, which directly affect the accuracy of SOH estimation. Preprocessing improves the data signal-to-noise ratio by removing outliers, correcting biases, and standardizing features, preventing "dirty data" from leading to model misjudgments. Data formats collected under different operating conditions and equipment are unified to enhance comparability. By cleaning cell test data, filling in missing values, and removing outliers, the accuracy of cell test data reflection is improved, thereby enhancing the accuracy of cell health assessments.

[0177] According to an embodiment of the present invention, constructing an expansion curve based on optimized experimental data, extracting battery cell expansion characteristics based on the expansion curve, and analyzing battery cell response information based on the expansion characteristics specifically include:

[0178] Set the number of charge and discharge cycles for the battery cell, perform experiments on the battery cell, and obtain the battery cell displacement information;

[0179] Obtain the initial thickness of the battery cell, and divide the expansion displacement information by the initial thickness of the battery cell to obtain the expansion amount;

[0180] The expansion amount is normalized, and an expansion curve is drawn with the number of charge and discharge cycles as the horizontal axis and the normalized expansion amount as the vertical axis;

[0181] Extract the cell expansion characteristics based on the expansion curve, and analyze the relationship between the cell expansion rate during the charging stage, the cell expansion rate during the discharging stage, the cell collision volume during the full cycle, and the cell capacity based on the cell expansion characteristics;

[0182] The cell response information is obtained by the relationship between the cell expansion rate during the cell charging stage, the cell expansion rate during the cell discharging stage, the cell collision volume during the full cycle stage and the cell capacity.

[0183] It should be noted that by cyclically charging and discharging the battery cells, the expansion of the battery cells during the cyclic charging and discharging process is analyzed to analyze the expansion response of the battery cells, providing an effective basis for subsequent analysis of the battery health status based on the battery cell expansion.

[0184] According to an embodiment of the present invention, a lithium-ion battery health status prediction model is constructed based on big data, and the lithium-ion battery capacity decay status information is analyzed according to the battery health status prediction model, specifically including:

[0185] Obtain historical lithium-ion battery health status prediction data based on big data and establish a data set;

[0186] Divide the data set into training and test sets;

[0187] Select an initial model framework, iteratively train the initial model framework based on the training set, and obtain training results;

[0188] Analyze whether the training results converge;

[0189] If convergence occurs, an intermediate model is obtained. The intermediate model is tested based on the test set, and the model hyperparameters are dynamically adjusted based on the test results to obtain a health status prediction model for the ion battery and predict the capacity attenuation status information of the ion battery.

[0190] If it does not converge, adjust the data in the training set or the number of iterations.

[0191] It should be noted that the model is iteratively trained by obtaining historical lithium-ion battery health status prediction data through big data, so that the model has learning ability and improves the accuracy of model output.

[0192] According to an embodiment of the present invention, the battery health status information is analyzed based on the lithium-ion battery capacity decay status information, and the battery performance is analyzed according to the battery health status information to obtain performance evaluation information, specifically including:

[0193] Set the analysis time window and divide it into several sub-windows in equal proportions;

[0194] The starting time of each sub-window is used as the acquisition time to obtain the acquisition time nodes corresponding to several sub-windows;

[0195] Collect the lithium-ion battery capacity decay status information corresponding to different time nodes, compare the lithium-ion battery capacity decay status information of adjacent time nodes, and obtain the battery capacity decay rate of adjacent sub-windows;

[0196] Analyze battery attenuation rate fluctuation information based on the battery capacity attenuation rate of adjacent sub-windows;

[0197] The battery health status information is analyzed based on the battery attenuation rate fluctuation information, the battery performance within the time window is analyzed based on the battery health status information, and the battery performance is evaluated based on the evaluation indicators to obtain performance evaluation information. The evaluation indicators are set according to the battery model and battery factory data.

[0198] It should be noted that by dividing the time window into equal intervals, analyzing the battery capacity decay status information at different acquisition time nodes within the time window, and analyzing the battery decay fluctuation information, the battery health status can be accurately evaluated.

[0199] A third aspect of the present invention provides a computer-readable storage medium, which includes a program for a method for estimating the health status of a lithium-ion battery based on mechanical information. When the program for estimating the health status of a lithium-ion battery based on mechanical information is executed by a processor, the steps of the method for estimating the health status of a lithium-ion battery based on mechanical information as described above are implemented.

[0200] The present invention discloses a method, system and medium for estimating the health status of a lithium-ion battery based on mechanical information. The method comprises the following steps: setting experimental parameters, performing a full-life test on a lithium-ion battery cell based on the experimental parameters, and obtaining cell experimental data; preprocessing the cell experimental data, and obtaining optimized experimental data; constructing an expansion curve based on the optimized experimental data, extracting cell expansion characteristics based on the expansion curve, and analyzing cell response information based on the expansion characteristics; constructing a lithium-ion battery health status prediction model based on big data, and analyzing lithium-ion battery capacity decay state information according to the battery health status prediction model; analyzing battery health status information based on the lithium-ion battery capacity decay state information, and analyzing battery performance according to the battery health status information to obtain performance evaluation information; calculating cell response information by analyzing the cell expansion curve, and analyzing the lithium-ion battery capacity decay state according to the constructed lithium-ion battery health status prediction model, thereby accurately analyzing changes in battery health status, and further accurately analyzing battery performance, thereby improving analysis accuracy.

[0201] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of units is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0202] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0203] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0204] Those skilled in the art will appreciate that all or part of the steps of the above-mentioned method embodiments may be implemented by hardware associated with program instructions, and the aforementioned program may be stored in a readable storage medium. When the program is executed, the program executes the steps of the above-mentioned method embodiments. The aforementioned storage medium includes various media that can store program codes, such as mobile storage devices, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0205] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention, or the part that contributes to the existing technology, can be embodied in the form of a software product. The software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program code, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

Claims

1. A method for estimating the health status of a lithium-ion battery based on mechanical information, characterized in that: include: Set experimental parameters, perform full life tests on lithium-ion battery cells based on the experimental parameters, and obtain cell experimental data; Preprocess the cell experimental data to obtain optimized experimental data; Construct an expansion curve based on optimized experimental data, extract the cell expansion characteristics based on the expansion curve, and analyze the cell response information based on the expansion characteristics; Build a lithium-ion battery health status prediction model based on big data, and analyze the lithium-ion battery capacity attenuation status information according to the battery health status prediction model; The battery health status information is analyzed based on the capacity attenuation status information of the lithium-ion battery, and the battery performance is analyzed according to the battery health status information to obtain performance evaluation information.

2. The method for estimating the health status of a lithium-ion battery based on mechanical information according to claim 1, characterized in that: Set experimental parameters, perform full life tests on lithium-ion battery cells based on the experimental parameters, and obtain cell experimental data, including: Setting experimental parameters, including environmental parameters, mechanical parameters, and electrical performance test parameters; the environmental parameters include ambient temperature, ambient humidity, and light intensity; the mechanical parameters include pressure information, and the electrical performance test parameters include voltage, current, charge rate, and discharge rate; Set different values ​​of environmental parameters, mechanical parameters and electrical performance test parameters to create multiple different operating conditions; The battery cells are tested for their entire life cycle based on different operating conditions to obtain the battery cell experimental data.

3. The method for estimating the health status of a lithium-ion battery based on mechanical information according to claim 2, characterized in that: Preprocess the cell experimental data to obtain optimized experimental data, including: Obtain battery cell experimental data, clean the battery cell experimental data, and obtain experimental data in a standardized format; Extract the data features of the experimental data after format standardization, and filter out missing values ​​and outliers of the data features; Eliminate the data features corresponding to outliers; Analyze the previous moment value and the next moment value based on the linear interpolation method, and average the previous moment value and the next moment value to obtain the mean value; Compare the mean with the missing value to obtain the missing difference, fill the missing value based on the missing difference, and obtain the optimized experimental data.

4. The method for estimating the health status of a lithium-ion battery based on mechanical information according to claim 3, characterized in that: Construct an expansion curve based on optimized experimental data, extract the cell expansion characteristics based on the expansion curve, and analyze the cell response information based on the expansion characteristics, specifically including: Set the number of charge and discharge cycles for the battery cell, perform experiments on the battery cell, and obtain the battery cell displacement information; Obtain the initial thickness of the battery cell, and divide the expansion displacement information by the initial thickness of the battery cell to obtain the expansion amount; The expansion amount is normalized, and an expansion curve is drawn with the number of charge and discharge cycles as the horizontal axis and the normalized expansion amount as the vertical axis; Extract the cell expansion characteristics based on the expansion curve, and analyze the relationship between the cell expansion rate during the charging stage, the cell expansion rate during the discharging stage, the cell collision volume during the full cycle, and the cell capacity based on the cell expansion characteristics; The cell response information is obtained by the relationship between the cell expansion rate during the cell charging stage, the cell expansion rate during the cell discharging stage, the cell collision volume during the full cycle stage and the cell capacity.

5. The method for estimating the health status of a lithium-ion battery based on mechanical information according to claim 4, characterized in that: Build a lithium-ion battery health status prediction model based on big data, and analyze the lithium-ion battery capacity attenuation status information according to the battery health status prediction model, including: Obtain historical lithium-ion battery health status prediction data based on big data and establish a data set; Divide the data set into training and test sets; Select an initial model framework, iteratively train the initial model framework based on the training set, and obtain training results; Analyze whether the training results converge; If convergence occurs, an intermediate model is obtained. The intermediate model is tested based on the test set, and the model hyperparameters are dynamically adjusted based on the test results to obtain a health status prediction model for the ion battery and predict the capacity attenuation status information of the ion battery. If it does not converge, adjust the data in the training set or the number of iterations.

6. The method for estimating the health status of a lithium-ion battery based on mechanical information according to claim 5, characterized in that: The battery health status information is analyzed based on the lithium-ion battery capacity attenuation status information, and the battery performance is analyzed based on the battery health status information to obtain performance evaluation information, including: Set the analysis time window and divide it into several sub-windows in equal proportions; The starting time of each sub-window is used as the acquisition time to obtain the acquisition time nodes corresponding to several sub-windows; Collect the lithium-ion battery capacity decay status information corresponding to different time nodes, compare the lithium-ion battery capacity decay status information of adjacent time nodes, and obtain the battery capacity decay rate of adjacent sub-windows; Analyze battery attenuation rate fluctuation information based on the battery capacity attenuation rate of adjacent sub-windows; The battery health status information is analyzed based on the battery attenuation rate fluctuation information, the battery performance within the time window is analyzed based on the battery health status information, and the battery performance is evaluated based on the evaluation indicators to obtain performance evaluation information. The evaluation indicators are set according to the battery model and battery factory data.

7. A lithium-ion battery health status estimation system based on mechanical information, characterized in that: The system includes: a memory and a processor, wherein the memory includes a program of a method for estimating the health state of a lithium-ion battery using mechanical information, and when the program of the method for estimating the health state of a lithium-ion battery using mechanical information is executed by the processor, the following steps are implemented: Set experimental parameters, perform full life tests on lithium-ion battery cells based on the experimental parameters, and obtain cell experimental data; Preprocess the cell experimental data to obtain optimized experimental data; Construct an expansion curve based on optimized experimental data, extract the cell expansion characteristics based on the expansion curve, and analyze the cell response information based on the expansion characteristics; Build a lithium-ion battery health status prediction model based on big data, and analyze the lithium-ion battery capacity attenuation status information according to the battery health status prediction model; The battery health status information is analyzed based on the capacity attenuation status information of the lithium-ion battery, and the battery performance is analyzed according to the battery health status information to obtain performance evaluation information.

8. The lithium-ion battery health status estimation system based on mechanical information according to claim 7, characterized in that: Set experimental parameters, perform full life tests on lithium-ion battery cells based on the experimental parameters, and obtain cell experimental data, including: Setting experimental parameters, including environmental parameters, mechanical parameters, and electrical performance test parameters; the environmental parameters include ambient temperature, ambient humidity, and light intensity; the mechanical parameters include pressure information, and the electrical performance test parameters include voltage, current, charge rate, and discharge rate; Set different values ​​of environmental parameters, mechanical parameters and electrical performance test parameters to create multiple different operating conditions; The battery cells are tested for their entire life cycle based on different operating conditions to obtain the battery cell experimental data.

9. The lithium-ion battery health status estimation system based on mechanical information according to claim 8, characterized in that: Preprocess the cell experimental data to obtain optimized experimental data, including: Obtain battery cell experimental data, clean the battery cell experimental data, and obtain experimental data in a standardized format; Extract the data features of the experimental data after format standardization, and filter out missing values ​​and outliers of the data features; Eliminate the data features corresponding to outliers; Analyze the previous moment value and the next moment value based on the linear interpolation method, and average the previous moment value and the next moment value to obtain the mean value; Compare the mean with the missing value to obtain the missing difference, fill the missing value based on the missing difference, and obtain the optimized experimental data.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a program for a method for estimating the health status of a lithium-ion battery using mechanical information. When the program is executed by a processor, the steps of the method for estimating the health status of a lithium-ion battery using mechanical information are implemented as described in any one of claims 1 to 6.

Images