Lithium battery state evaluation method based on alternating current impedance and online operation data
By combining AC impedance and online operation data, the lithium battery status evaluation model is established using the SVM algorithm, which solves the problems of low evaluation accuracy and poor real-time performance in the existing technology, and achieves a more accurate and reliable lithium battery status evaluation to ensure the safety of the battery. and efficient operation.
Patent Information
- Application Number
- CN202411159267.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-22
- Publication Date
- 2024-11-29
AI Technical Summary
The existing lithium battery status evaluation method cannot fully reflect the changes in the internal state of the battery, resulting in low evaluation accuracy and poor real-time performance, and poses safety hazards.
A lithium battery state evaluation method based on AC impedance and online operating data is adopted. A state evaluation model is established through a support vector machine (SVM) algorithm, and preprocessing and model training is carried out in combination with AC impedance characteristics and actual operating data to achieve real-time state evaluation.
It improves the accuracy and reliability of lithium battery status evaluation, and can continuously monitor and evaluate the status during battery operation, ensuring the safe and efficient operation of the battery, and reducing maintenance costs.
Smart Images

Figure CN119044788A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of battery status evaluation, and in particular to a lithium battery status evaluation method based on AC impedance and online operation data. Background Art
[0002] With the rapid development of electric vehicles and renewable energy, lithium batteries are widely used as efficient, lightweight, and high-energy-density battery devices. However, during the use of lithium batteries, the performance gradually decays with the increase of usage time and charge and discharge cycles. Therefore, the evaluation of the status of lithium batteries is crucial to ensure their safe and reliable operation.
[0003] At present, the status assessment methods of lithium batteries are mainly based on analysis of parameters such as voltage, current, and temperature, but these methods often cannot fully reflect the changes in the internal status of the battery. There are still problems such as low assessment accuracy and poor real-time performance. They cannot accurately feedback the status of lithium batteries, and there are certain safety hazards in battery use projects. Summary of the invention
[0004] The present invention aims to provide a lithium battery status assessment method based on AC impedance and online operation data to solve the problems raised in the above background technology.
[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a lithium battery status evaluation method based on AC impedance and online operation data, specifically comprising:
[0006] S1, perform AC impedance test on the lithium battery to obtain the AC impedance characteristics of the lithium battery;
[0007] S2, analyzing the AC impedance characteristics of the lithium battery obtained in step S1 to extract relevant characteristic parameters;
[0008] S3, collecting and analyzing the actual operation data of the lithium battery to establish a lithium battery working state model;
[0009] S4, based on the lithium battery AC impedance characteristics obtained in step S1 and the actual operation data of the lithium battery obtained in step S3, a lithium battery status assessment model is established, and the specific steps are as follows:
[0010] S41, preprocessing the lithium battery AC impedance characteristic obtained in step S1 and the actual operation data of the lithium battery obtained in step S3 to obtain preprocessed data;
[0011] S42, dividing the preprocessed data in step S41 into a training set and a test set, and performing data division to ensure the objectivity of model evaluation;
[0012] S43, select N sample data related to the AC impedance characteristics of lithium batteries from the data, each sample data contains d features, represented by x i =(x i1 , x i2 , …, x id ), the state label corresponding to the sample data is y i , where i = 1, 2, ..., N;
[0013] S44. According to the sample data x i With state label y i , the optimal hyperplane is found through the SVM algorithm, and the hyperplane is expressed as:
[0014] w t *x+b=0:
[0015] Among them, w is the normal vector of the hyperplane, and b is the bias term;
[0016] The decision function of SVM is expressed as:
[0017] f(x)=sign(w t *x+b);
[0018] Among them, the sign() function is the sign function. When w t * When x+b>0, f(x) is positive, when w t *When x+b<0, f(x) is negative;
[0019] The SVM model is optimized through convex optimization to maximize the distance from the support vector to the hyperplane. The specific formula is:
[0020]
[0021] Among them, ||w|| is the L2 norm of w, C is the regularization parameter, and the trade-off between margin and misclassification is controlled by model optimization;
[0022] S5, training and optimizing the lithium battery status assessment model established in S4;
[0023] S6, deploys the trained model to the online evaluation system to perform real-time evaluation of the lithium battery status.
[0024] Furthermore, step S1 is further:
[0025] S11. Ensure that the lithium battery is at a specific temperature and charge and discharge state, and apply voltage and current signals of different frequencies to both ends of the lithium battery;
[0026] S12, record the voltage and current response at each frequency;
[0027] S13, perform impedance calculation to obtain the AC impedance characteristics of the lithium battery, and the calculation formula is:
[0028]
[0029] Where V(f) and I(f) are the voltage and current at frequency f, respectively.
[0030] Furthermore, the characteristic parameters in step S2 include charge transfer resistance, double layer capacitance, cut-off frequency, electrolyte resistance, diffusion time constant, and polarization impedance.
[0031] Furthermore, step S3 is further:
[0032] S31. Collect data of the lithium battery in actual operation, including voltage data, current data, temperature data, capacity data, and cycle data;
[0033] S32, performing data cleaning, normalization, and smoothing preprocessing on the data to obtain preprocessed data to improve data quality;
[0034] S33, extracting key features of statistical features, time domain features and frequency domain features from the acquired pre-processed data for model construction;
[0035] Furthermore, the preprocessing in step S41 specifically includes:
[0036] S411, processing missing values by deleting missing value samples and filling them with interpolation methods, processing outliers by statistical methods, processing duplicate values by direct deletion, and performing denoising and filtering on the data;
[0037] S412, merging data from multiple data sources by connection for data integration to ensure data field alignment and data consistency;
[0038] S413, extracting features from the integrated data to obtain feature values;
[0039] S414: normalize the extracted feature values to eliminate dimensional differences between different features.
[0040] Furthermore, step S5 is further as follows:
[0041] S51. Hyperparameter tuning of SVW model is performed through cross validation.
[0042] S52. Use the indicator evaluation model to evaluate the SVW model.
[0043] Furthermore, the lithium battery status assessment includes health status assessment, remaining service life assessment, performance stability assessment, safety performance assessment, and environmental adaptability assessment.
[0044] Beneficial effects of the present invention: AC impedance reflects the electrochemical characteristics inside the battery, while online operation data reflects the dynamic changes of the battery during actual working. This scheme utilizes a method combining lithium battery AC impedance data with online operation data, which can make full use of different types of information, more accurately evaluate the status of the lithium battery, and improve the accuracy and reliability of the evaluation; through real-time collection and processing of online operation data, the SVW evaluation model is used to perform real-time detection and evaluation of the lithium battery operation status, which can continuously monitor and evaluate the battery status during battery operation, and can fully understand the status of the lithium battery, provide strong data support for battery maintenance and management, ensure the safe and efficient operation of the battery, and promptly discover problems with the lithium battery and take corresponding measures to reduce maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] Figure 1 It is a schematic diagram of the overall process of the present invention.
[0046] Figure 2 It is a schematic flow chart of step S3 of the present invention.
[0047] Figure 3 It is a schematic flow chart of step S4 of the present invention. DETAILED DESCRIPTION
[0048] In the following description, a large number of specific details are provided to provide a more thorough understanding of the present invention. However, it is apparent to those skilled in the art that the present invention can be implemented without one or more of these details. In other examples, in order to avoid confusion with the present invention, some technical features well known in the art are not described.
[0049] The present invention provides a lithium battery status evaluation method based on AC impedance and online operation data, which establishes a model for the AC impedance characteristics of the lithium battery and the online operation data, evaluates the lithium battery operation status, and Figure 1 , the specific steps are:
[0050] S1, perform AC impedance test on the lithium battery to obtain the AC impedance characteristics of the lithium battery;
[0051] S2, analyzing the AC impedance characteristics of the lithium battery obtained in step S1 to extract relevant characteristic parameters;
[0052] S3, collecting and analyzing the actual operation data of the lithium battery to establish a lithium battery working state model;
[0053] S4, establishing a lithium battery status assessment model based on the lithium battery AC impedance characteristics obtained in step S1 and the actual operation data of the lithium battery obtained in step S3;
[0054] S5, training and optimizing the lithium battery status assessment model established in S4;
[0055] S6, deploys the trained model to the online evaluation system to perform real-time evaluation of the lithium battery status.
[0056] This embodiment uses a binary classification model of a support vector machine to construct an algorithm formula for a lithium battery status evaluation model to evaluate the lithium battery status. Through the optimal hyperplane parameters w and b, the lithium battery status characteristics are reflected in real time, the lithium battery status is evaluated more accurately, and the accuracy and reliability of the evaluation are improved. The evaluation method is simple to operate and easy to implement and apply in actual production.
[0057] In a further embodiment, the AC impedance test is to analyze the electrochemical process inside the battery by applying a small amplitude AC voltage or current signal at both ends of the battery and measuring the corresponding response. Performing AC impedance testing on lithium batteries to obtain the AC impedance characteristics of lithium batteries is a process involving electrochemical analysis and data processing. This embodiment uses an AC impedance spectrometer to apply an AC signal and measure the response, covering a frequency range from mHz to kHz, covering multiple frequency points, and using a small amplitude to avoid a significant impact on the battery state. The specific test steps are:
[0058] S11. Ensure that the lithium battery is at a specific temperature and charge and discharge state, and apply voltage and current signals of different frequencies to both ends of the lithium battery;
[0059] S12, record the voltage and current response at each frequency;
[0060] S13, perform impedance calculation to obtain the AC impedance characteristics of the lithium battery, and the calculation formula is:
[0061]
[0062] Where V(f) and I(T) are the voltage and current at frequency f, respectively.
[0063] The real part Z′ and imaginary part Z″ of the impedance are plotted on the complex plane to form an impedance spectrum. The Nyquist diagram of the impedance spectrum is visually analyzed to identify the impedance characteristics at different frequencies, mainly including:
[0064] High frequency region: Ohmic impedance (R s ), which appears as a vertical line segment;
[0065] Mid-frequency region: Charge transfer impedance (R ct ), which appears as a semicircle;
[0066] Low frequency region: Warburg impedance (Zw ), which appears as a slash;
[0067] Extract the main parameters from the impedance spectrum, including:
[0068] The ohmic impedance R is obtained directly from the intercept of the vertical line segment in the high frequency region s , R s It is related to the material and structure of the battery and reflects the pure resistance inside the battery. The charge transfer impedance R is obtained from the diameter of the semicircle in the medium frequency region. ct , R ct It is related to the active materials and interface state of the battery, and reflects the impedance of the charge transfer process at the interface between the electrode and the electrolyte. The double-layer capacitance C is calculated by fitting the semicircle in the intermediate frequency region and combining the charge transfer impedance. dl . C dl It is related to the properties of the battery's electrode materials and electrolyte, and reflects the charge storage capacity of the electrode surface; the Warburg impedance Z is obtained from the slope of the oblique line in the low-frequency region. w , Z w It is related to the properties of the battery's electrode materials and electrolyte, and reflects the impedance of the ion diffusion process in the electrode material.
[0069] In this embodiment, the impedance spectrum is visually analyzed to identify the impedance characteristics at different frequencies and extract characteristic parameters. The health status of the battery is evaluated by comparing the characteristic parameters of different batteries or under different conditions, and the trend of characteristic parameters over time is analyzed to predict the remaining service life of the battery. The fault or unevenness inside the battery is indicated based on the abnormal characteristic parameters. The state evaluation and performance analysis of the lithium battery are performed by analyzing the AC impedance characteristics of the lithium battery.
[0070] In a further embodiment, actual operation data of the lithium battery is collected and analyzed, including voltage data, current data, temperature data, capacity data, and charge and discharge cycle times; and the collected data is preprocessed to improve data quality. The specific steps are as follows:
[0071] Use statistical methods to identify and remove outliers and perform data cleaning;
[0072] Use minimum-maximum normalization to normalize the data and scale the data to the same range. The calculation formula is:
[0073]
[0074] Use moving average to smooth data and reduce data fluctuations. The formula is:
[0075]
[0076] Extract key features from the preprocessed data, including:
[0077] Statistical feature extraction, the mean calculation formula is:
[0078]
[0079] The variance calculation formula is:
[0080]
[0081] The calculation formula of the time domain characteristic peak is:
[0082] P = max(x);
[0083] The calculation formula of frequency domain spectrum characteristics after Fourier transform is:
[0084]
[0085] In this embodiment, based on the voltage data, current data, temperature data, capacity data, and number of charge and discharge cycles during the actual operation of the lithium battery, the performance change trend of battery capacity attenuation, internal resistance increase, etc. can be more accurately predicted, which is helpful to adjust the battery management strategy in time and extend the battery life; analysis of actual operation data can help identify potential problems in material selection and process flow in battery design and manufacturing, which helps to improve battery design and improve product quality and performance.
[0086] In a further embodiment, a lithium battery status assessment model is established based on the acquired lithium battery AC impedance characteristics and the actual operation data of the lithium battery, and the specific steps are as follows:
[0087] S41, preprocessing the lithium battery AC impedance characteristics obtained in step S1 and the actual operation data of the lithium battery obtained in step S3 to obtain preprocessed data. The specific preprocessing process includes:
[0088] S411, fill missing values, delete outliers and perform interpolation operations. These steps are aimed at repairing or removing possible defects in the data obtained by the drone to ensure the integrity and quality of the data; the filling of missing values can be done by using the mean, median or interpolation methods based on other relevant features to minimize the impact on the distribution characteristics of the data; the detection and deletion of outliers are performed through statistical analysis or methods based on domain knowledge to avoid the adverse effects of these extreme values on the model training results. By filling missing values, deleting outliers and interpolating methods, the missing values, outliers and error values in the data packet are processed to ensure the integrity and accuracy of the data; a Gaussian filter is selected to remove high-frequency noise, retain a smooth surface, and perform denoising to remove noise in the data;
[0089] S412. Combine records from multiple data sets by using a common key, ensuring that the primary key or specified association key is retained during the merge process to ensure data field alignment; merge data from different data sources based on index or column values; ensure that field names, data types, and formats are consistent after the merge by aligning data fields; ensure that the meaning, format, and quality of the data in the integrated data set are consistent through steps such as data cleaning, standardization, and validation;
[0090] S413. Convert time domain signals into frequency domain signals through Fourier transform and other methods, analyze their frequency response curves, and help understand the energy distribution and response of signals at different frequencies; harmonic distortion analysis is used to identify and eliminate undesirable frequency components in the system, thereby improving the efficiency and performance of the system. The maximum and minimum values reflect the extreme conditions of the data, while the average value provides the central trend of the data. These characteristics are calculated and used for basic description and preliminary analysis of the data to obtain characteristic values;
[0091] S414, normalizing the extracted feature values using a maximum-minimum normalization method to eliminate dimensional differences between different features;
[0092] S42, dividing the preprocessed data in step S41 into a training set and a test set, and performing data division to ensure the objectivity of model evaluation;
[0093] S43, select N sample data related to the AC impedance characteristics of lithium batteries from the data, each sample data contains d features, represented by x i =(x i1 , x i2 , …, x id ), the state label corresponding to the sample data is y i , where i = 1, 2, ..., N;
[0094] S44. According to the sample data x i With state label y i , the optimal hyperplane is found through the SVM algorithm, and the hyperplane is expressed as:
[0095] w t *x+b=0;
[0096] Among them, w is the normal vector of the hyperplane, and b is the bias term;
[0097] The decision function of SVM is expressed as:
[0098] f(x)=sign(w t *x+b);
[0099] Among them, the sign() function is the sign function. When w t* When x+b>0, f(x) is positive, when w t *When x+b<0, f(x) is negative;
[0100] The SVM model is optimized through convex optimization to maximize the distance from the support vector to the hyperplane. The specific formula is:
[0101]
[0102] Among them, ||w|| is the L2 norm of w, and C is the regularization parameter used to control the trade-off between margin and misclassification.
[0103] The established SVW model is trained and optimized as follows:
[0104] K-fold cross validation is used to evaluate the model performance. The specific formula is:
[0105]
[0106] Among them, Performance is the performance evaluation function;
[0107] Select the mean square error indicator evaluation model for model evaluation, and the calculation formula is:
[0108]
[0109] This embodiment can remove or repair outliers, missing values and noise in the data by preprocessing the acquired data, improve the quality and accuracy of the data, and make the preprocessed data easier to analyze and process, speed up the model training, and reduce the overfitting of the model to the noise and irrelevant information in the training data, thereby improving the generalization ability of the model. By using the cross-validation method to tune hyperparameters, the generalization ability and prediction accuracy of the model can be improved.
[0110] In a further embodiment, the lithium battery status is evaluated by analyzing the evaluation results output by the evaluation model, specifically including:
[0111] By evaluating the ratio of the current capacity of the battery to the capacity of a new battery, the capacity retention rate is obtained; by evaluating the change of the internal resistance of the battery relative to the internal resistance of a new battery, the internal resistance change data is obtained; by evaluating the ratio of the number of charge and discharge cycles the battery has gone through to the number of cycles of the designed life, the cycle life is obtained; by evaluating the capacity retention rate, the change of internal resistance and the cycle life, the degree of battery retention relative to a new battery is reflected, and the health status assessment result of the lithium battery is obtained;
[0112] Predict the remaining time of the battery in hours, days or charge and discharge cycles, perform time prediction, predict the trend of battery capacity decay over time by capacity decline rate, and obtain the capacity decay trend prediction result. Through the above prediction, reflect the expected time that the battery can continue to be used in the current state, and obtain the remaining service life prediction result;
[0113] By evaluating the energy conversion efficiency of the battery during the charging and discharging process, the charging and discharging efficiency evaluation results are obtained. By evaluating the performance changes of the battery at different temperatures, the temperature impact evaluation results are obtained. Through the above evaluations, the lithium battery performance stability evaluation results are obtained.
[0114] The safety threshold is used to evaluate the safety of the battery under overcharge or over-discharge conditions, and the evaluation results of lithium battery charge and over-discharge are obtained. The protection mechanism of the battery under short-circuit conditions is evaluated by the response time, and the evaluation results of short-circuit protection are obtained. Through the above evaluations, the safety performance evaluation results of the lithium battery are obtained, which reflect the safety performance of the lithium battery under abnormal conditions;
[0115] The performance change rate is used to evaluate the performance stability of the battery under extreme temperatures to obtain the temperature adaptability evaluation results. The performance change rate is used to evaluate the performance stability of the battery under different humidity conditions to obtain the humidity adaptability evaluation results. Through the above evaluations, the environmental adaptability evaluation results of the lithium battery are obtained.
[0116] Based on the above evaluation results, a reasonable charging and discharging strategy can be formulated for the charging and discharging management of lithium batteries to avoid overcharging and over-discharging, and appropriate temperature control measures can be formulated to keep the battery within the optimal operating temperature range. Through regular lithium battery performance inspections and maintenance, the status of lithium batteries can be fully understood, providing strong data support for battery maintenance and management, ensuring the safety and efficient operation of the battery.
[0117] As described above, although the present invention has been shown and described with reference to specific preferred embodiments, it should not be construed as limiting the present invention itself. Various changes may be made to it in form and detail without departing from the spirit and scope of the present invention as defined in the appended claims.
Claims
1. A lithium battery status assessment method based on AC impedance and online operation data, characterized in that: Specifically include: S1, perform AC impedance test on the lithium battery to obtain the AC impedance characteristics of the lithium battery; S2, analyzing the AC impedance characteristics of the lithium battery obtained in step S1 to extract relevant characteristic parameters; S3, collecting and analyzing the actual operation data of the lithium battery to obtain the actual operation status of the lithium battery; S4, based on the lithium battery AC impedance characteristics obtained in step S1 and the actual operation data of the lithium battery obtained in step S3, a lithium battery status assessment model is established, and the specific steps are as follows: S41, preprocessing the lithium battery AC impedance characteristic obtained in step S1 and the actual operation data of the lithium battery obtained in step S3 to obtain preprocessed data; S42, dividing the preprocessed data in step S41 into a training set and a test set, and performing data division to ensure the objectivity of model evaluation; S43, select N sample data related to the AC impedance characteristics of lithium batteries from the data, each sample data contains d features, represented by x i =(x i1 , x i2 , …, x id ), the state label corresponding to the sample data is y i , where i = 1, 2, ..., N; S44. According to the sample data x i With state label y i , the optimal hyperplane is found through the SVM algorithm, and the hyperplane is expressed as: w t *x+b=0; Among them, w is the normal vector of the hyperplane, and b is the bias term; The decision function of SVM is expressed as: f(x)=sign(w t *x+b); Among them, the sign() function is the sign function. When w t * When x+b>0, f(x) is positive, when w t *When x+b<0, f(x) is negative; The SVM model is optimized through convex optimization to maximize the distance from the support vector to the hyperplane. The specific formula is: Among them, ∥w∥ is the L2 norm of w, C is the regularization parameter, and the trade-off between margin and misclassification is controlled by model optimization; S5, training and optimizing the lithium battery status assessment model established in S4; S6, deploys the trained model to the online evaluation system to perform real-time evaluation of the lithium battery status.
2. The lithium battery status assessment method based on AC impedance and online operation data according to claim 1, characterized in that: Step S1 is further as follows: S11. Ensure that the lithium battery is at a specific temperature and charge and discharge state, and apply voltage and current signals of different frequencies to both ends of the lithium battery; S12, record the voltage and current response at each frequency; S13, perform impedance calculation to obtain the AC impedance characteristics of the lithium battery, and the calculation formula is: Where V(f) and I(f) are the voltage and current at frequency f, respectively.
3. The lithium battery status assessment method based on AC impedance and online operation data according to claim 1, characterized in that: The characteristic parameters in step S2 include charge transfer resistance, double layer capacitance, cut-off frequency, electrolyte resistance, diffusion time constant, and polarization impedance.
4. The lithium battery status assessment method based on AC impedance and online operation data according to claim 1, characterized in that: Step S3 is further as follows: S31. Collect data of the lithium battery in actual operation, including voltage data, current data, temperature data, capacity data, and cycle data; S32, performing data cleaning, normalization, and smoothing preprocessing on the data to obtain preprocessed data to improve data quality; S33, extracting key features of statistical features, time domain features and frequency domain features from the acquired pre-processed data for model construction.
5. The lithium battery status assessment method based on AC impedance and online operation data according to claim 1, characterized in that: The preprocessing in step S41 specifically includes: S411, processing missing values by deleting missing value samples and filling them with interpolation methods, processing outliers by statistical methods, processing duplicate values by direct deletion, and performing denoising and filtering on the data; S412, merging data from multiple data sources by connection for data integration to ensure data field alignment and data consistency; S413, extracting features from the integrated data to obtain feature values; S414: normalize the extracted feature values to eliminate dimensional differences between different features.
6. The lithium battery status assessment method based on AC impedance and online operation data according to claim 1, characterized in that: Step S5 is further as follows: S51. Hyperparameter tuning of SVW model is performed through cross validation. S52. Use the indicator evaluation model to evaluate the SVW model.
7. The lithium battery status assessment method based on AC impedance and online operation data according to claim 1, characterized in that: The lithium battery status assessment includes health status assessment, remaining service life assessment, performance stability assessment, safety performance assessment, and environmental adaptability assessment.
Citation Information
Cited By
Lithium ion battery performance rapid evaluation method based on time domain impedance spectroscopy
CN120385947A
