Lithium ion single battery monitoring system and method based on weak magnetic detection technology

Abstract

The invention provides a lithium ion single battery monitoring system and method based on a weak magnetic detection technology, and the system comprises a multi-channel magnetic field measurement module which is used for measuring the magnetic signal data of a lithium ion single battery through a distributed fluxgate sensor array; the magnetic signal acquisition module is used for acquiring magnetic signal data generated by the lithium ion single battery and measured by the multi-channel magnetic field measurement module; and the signal processing module is used for converting the preprocessed magnetic signal data into a curve graph and completing classification evaluation of the health state of the lithium ion single battery in combination with a battery health state evaluation model. According to the technical scheme, the accuracy rate of classification and recognition of the lithium ion pilot-scale test or small-scale test batteries in different health states and before and after different working conditions reaches 87.5%, and the method has the application space in the fields of electric automobiles and the like and is of great significance in guaranteeing safe operation of equipment.

Description

Technical Field

[0001] The present invention belongs to the technical field of battery health status monitoring, and in particular relates to a lithium-ion single battery monitoring system and method based on weak magnetic field detection technology. Background Art

[0002] Currently, the main magnetic-related monitoring methods for single-cell lithium-ion batteries include magnetic resonance technology and magnetic field measurement technology. Magnetic resonance technologies include nuclear magnetic resonance (NMR), electron paramagnetic resonance (EPR), and magnetic resonance imaging (MRI). NMR is a spectroscopic analysis method based on the interaction between the magnetic moment of an atomic nucleus and an applied magnetic field. It can quantitatively characterize the changing behavior of chemical substances during the cycling of lithium-ion batteries. Electron paramagnetic resonance, a spectroscopic analysis method based on the interaction between unpaired electrons and an applied magnetic field, can detect free radicals, transition metal ions, and other paramagnetic species in materials. Therefore, it can be applied to the monitoring of lithium-ion battery materials, reaction mechanisms, and failure mechanisms. Magnetic resonance imaging is a non-destructive imaging method based on the principle of nuclear magnetic resonance. It can provide high-resolution spatial information for visual monitoring of the microstructure and dynamic processes within lithium-ion batteries. Magnetic field measurement technology has been a method that has gradually emerged in recent years and has been applied to lithium-ion batteries. Its principle is to use a magnetic sensor to measure the external magnetic field of the battery and analyze the changes in the external magnetic field to determine and evaluate the battery's operating performance.

[0003] The magnetic correlation monitoring technology that has emerged in recent years provides an innovative solution for lithium-ion battery status monitoring. Compared with traditional detection methods, this technology has significant advantages: first, it can achieve multi-dimensional real-time monitoring by using low-cost equipment; second, it can intuitively reflect the current distribution characteristics inside the battery by relying on the visualization of magnetic field distribution; third, the spatial arrangement of the magnetic sensor array can be flexibly configured according to the characteristics of the battery system and the differences in monitoring requirements. However, there are still some problems in current research:

[0004] Existing technologies lack statistically significant reproducible experimental data for dynamic monitoring of battery charge and discharge cycles, and no repeated complete charge and discharge cycle studies have been conducted on the batteries. No in-depth research and analysis has been conducted on single magnetic induction intensity curves. As a direct representation of the electrochemical process, the dynamic evolution characteristics of a single-channel magnetic induction intensity curve are strongly correlated with the multi-physical field coupling within the battery, but the related time-frequency domain characteristics have not been fully analyzed. Current magnetic signal-based state assessments have not yet established a diagnostic model that integrates electromagnetic time-domain features with battery health status. There is a lack of dedicated extraction algorithms for the time-domain features of magnetic induction curves, as well as feature classification models based on machine learning algorithms. Summary of the Invention

[0005] In response to the problems existing in the above-mentioned prior art, the present invention provides a lithium-ion single cell battery monitoring system and method based on weak magnetic detection technology, which performs in-situ non-destructive weak magnetic monitoring of the complete charge and discharge cycle of the single lithium-ion battery, and classifies and evaluates the health status of the lithium-ion single cell according to the magnetic induction data curve.

[0006] The lithium-ion single battery monitoring system based on weak magnetic detection technology includes:

[0007] A multi-channel magnetic field measurement module is used to measure the magnetic signal data of lithium-ion single batteries through a distributed fluxgate sensor array;

[0008] A magnetic signal acquisition module is used to collect magnetic signal data generated by the lithium-ion single battery measured by the multi-channel magnetic field measurement module;

[0009] The signal processing module is used to convert the pre-processed magnetic signal data into a curve graph, and complete the classification evaluation of the health status of the lithium-ion single battery in combination with the battery health status evaluation model.

[0010] Preferably, the detection sensitive direction of the distributed fluxgate sensor array used by the multi-channel magnetic field measurement module is perpendicular to the end face of the cylinder, and the magnetic signal data actually measured is the external induced magnetic field signal at different positions on the surface of the lithium-ion single battery at the same time during the charge and discharge cycle.

[0011] Preferably, the multi-channel magnetic field measurement module includes:

[0012] An influencing factor acquisition unit is used to obtain influencing factors of the external induced magnetic field of the lithium-ion single cell based on Maxwell's equations, magnetic permeability, and the electrochemical reaction principle of the lithium-ion single cell; wherein the influencing factors include the internal current of the lithium-ion single cell and the magnetic susceptibility of the lithium-ion single cell material, and the magnetic susceptibility is an interference signal;

[0013] A magnetic field change acquisition unit, configured to obtain a surface magnetic field change of the lithium-ion single cell based on the internal current of the lithium-ion single cell;

[0014] The monitoring position acquisition unit is used to use a distributed fluxgate sensor array to scan the external induced magnetic field of the lithium-ion single battery during the charge and discharge cycle, and obtain the position where the magnetic field change on the surface of the lithium-ion single battery meets the preset conditions and is far away from the interference signal as the monitoring position.

[0015] Preferably, the signal processing module includes:

[0016] The signal preprocessing unit is used to select 100 magnetic signal data within one second, remove the highest and lowest values, and calculate the average value of the remaining magnetic signal data as the magnetic field monitoring data value of this second;

[0017] a curve generating unit, configured to generate a magnetic field change data curve based on the magnetic field monitoring data values ​​and corresponding time data;

[0018] A feature extraction unit, configured to extract segmented time-domain magnetic feature values ​​of the magnetic field change data curve;

[0019] The model building unit is used to build a battery health status assessment model by using the segmented time-domain magnetic eigenvalues ​​in combination with the k-NN algorithm.

[0020] Preferably, the model building unit includes:

[0021] A data set construction subunit is used to divide the segmented time-domain magnetic eigenvalues ​​into a training data set and a test data set; wherein the training data set is the segmented time-domain magnetic eigenvalues ​​that mark the health status categories of lithium-ion single batteries; and the test data set is the segmented time-domain magnetic eigenvalues ​​to be evaluated;

[0022] The k value determination subunit is used to determine the k value of the k-NN algorithm through a cross-validation method;

[0023] a Euclidean distance calculation subunit, configured to calculate the Euclidean distances of the segmented time-domain magnetic characteristic values ​​in the test data set and the training data set using a k-NN algorithm with a determined k value;

[0024] A neighboring sample acquisition subunit, configured to obtain neighboring sample data points in the training dataset that meet a similarity threshold with the test dataset based on the Euclidean distance;

[0025] The health status evaluation subunit is configured to obtain a health status evaluation result of the lithium-ion single battery of the test data set based on the health status category of the adjacent sample data points.

[0026] Preferably, the improvement of the k-NN algorithm includes introducing an automatic order disruption layer and a data normalization transposition layer, and adding a confusion matrix to evaluate the health status assessment results of lithium-ion single batteries.

[0027] The present invention also provides a lithium-ion single-cell battery monitoring method based on weak magnetic field detection technology, and the system includes:

[0028] Measure the magnetic signal data of lithium-ion single batteries through a distributed fluxgate sensor array;

[0029] Collect magnetic signal data generated by lithium-ion single batteries measured by a multi-channel magnetic field measurement module;

[0030] The pre-processed magnetic signal data is converted into a curve graph, and combined with the battery health status assessment model to complete the classification assessment of the health status of the lithium-ion single battery.

[0031] Preferably, the method for obtaining the monitoring position of the distributed fluxgate sensor array includes:

[0032] Based on Maxwell's equations, magnetic permeability, and the electrochemical reaction principle of lithium-ion single cells, factors affecting the external induced magnetic field of the lithium-ion single cells are obtained; wherein the influencing factors include the internal current of the lithium-ion single cells and the magnetic susceptibility of the lithium-ion single cell material, and the magnetic susceptibility is an interference signal;

[0033] Obtaining a change in the surface magnetic field of the lithium-ion single cell based on the internal current of the lithium-ion single cell;

[0034] A distributed fluxgate sensor array is used to scan the external induced magnetic field of the lithium-ion single battery during the charge and discharge cycle, and the position where the magnetic field change on the surface of the lithium-ion single battery meets the preset conditions and is far away from the interference signal is obtained as the monitoring position.

[0035] Compared with the existing technology, the beneficial effects of the present invention are as follows: by designing a special magnetic detection sensor device package, the present invention can achieve weak magnetic non-destructive monitoring of the charge and discharge cycle of single lithium-ion batteries of different shapes, sizes and requirements. Based on the magnetic field monitoring data, a rough judgment can be made on the health status of the lithium-ion battery; combined with the magnetic eigenvalue extraction method and the k-NN algorithm model, an accurate classification and evaluation of the health status of the lithium-ion battery can be achieved. As a supplementary means to existing monitoring technology, it can achieve non-destructive in-situ monitoring of the complete charge and discharge cycle of a single lithium-ion battery, and classify and evaluate the health status of the battery in combination with the eigenvalue extraction method and the k-NN algorithm model. The classification and identification accuracy rate of lithium-ion batteries in different health states and before and after different working conditions reaches 87.5%. It has room for application in fields such as electric vehicles and is of great significance to ensuring the safe operation of equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0037] Figure 1 This is a schematic structural diagram of a lithium-ion single battery monitoring system based on weak magnetic field detection technology according to an embodiment of the present invention;

[0038] Figure 2 Schematic diagram of components of a lithium-ion single battery monitoring system based on weak magnetic field detection technology according to an embodiment of the present invention;

[0039] Figure 3This is a schematic diagram of a probe according to an embodiment of the present invention;

[0040] Figure 4 This is a schematic diagram of packaging of an embodiment of the present invention;

[0041] Figure 5 This is a typical magnetic anomaly schematic diagram of an embodiment of the present invention;

[0042] Figure 6 This is a schematic diagram of the scanning direction of an embodiment of the present invention;

[0043] Figure 7 This is a monitoring diagram of an embodiment of the present invention;

[0044] Figure 8 This is a schematic diagram of a magnetic monitoring data curve according to an embodiment of the present invention;

[0045] Figure 9 16 battery magnetic monitoring data results of the embodiment of the present invention; wherein, (a) is the pilot weak magnetic monitoring results diagram, (b) is the small test weak magnetic monitoring results diagram;

[0046] Figure 10 Graphs showing the electrochemical monitoring data results for pilot and small-scale test cells No. 16 of Example 16 of the present invention; (a) shows the electrochemical monitoring results during the charging phase, and (b) shows the electrochemical monitoring results during the discharging phase;

[0047] Figure 11 Graphs showing performance evaluation results of a training dataset according to an embodiment of the present invention; (a) is a line graph of prediction results, and (b) is a confusion matrix.

[0048] Figure 12 Graphs showing the performance evaluation results of the test dataset for an embodiment of the present invention; (a) is a line graph of the prediction results, and (b) is a confusion matrix. DETAILED DESCRIPTION

[0049] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0050] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0051] Example 1

[0052] like Figure 1 、 Figure 2As shown, the lithium-ion single battery monitoring system based on weak magnetic field detection technology includes: a multi-channel magnetic field measurement module, a magnetic signal acquisition module and a signal processing module (host computer); the host computer and the magnetic signal acquisition module are connected via Ethernet.

[0053] The multi-channel magnetic field measurement module is used to measure the magnetic signal data of lithium-ion single batteries through a distributed fluxgate sensor array.

[0054] A further embodiment is that the detection sensitive direction of the distributed fluxgate sensor array used in the multi-channel magnetic field measurement module is perpendicular to the end face of the cylinder. The actual measured magnetic signal data is the external induced magnetic field signal at different positions on the surface of the lithium-ion single battery at the same time during the charge and discharge cycle. The magnetic field signals of multiple sensors can be obtained simultaneously. Depending on the size and shape of the battery to be measured and the monitoring requirements, fluxgate sensors with different numbers of channels can be selected. The structure of the multi-channel fluxgate sensor is as follows: Figure 3 shown.

[0055] According to information such as the position structure of the detection object, i.e., the actual weld seam, a certain length of cable is reserved for connecting the sensor array and the magnetic signal acquisition board module.

[0056] The system also includes a probe fixture, which is used to package and fix the sensor. The sensor is packaged in the fixture. The package is made of nylon material with very weak magnetism and conductivity to ensure that it has no effect on the monitored magnetic signal. The structure of the fixture is as follows: Figure 4 Use screws to fix the sensor on the outside of the tooling housing to prevent the probe packaging position or magnetism and conductivity from affecting the monitoring results.

[0057] In a further embodiment, the multi-channel magnetic field measurement module includes:

[0058] The influencing factor acquisition unit is used to obtain the influencing factors of the external induced magnetic field of the lithium-ion single cell based on Maxwell's equations, magnetic permeability, and the electrochemical reaction principle of the lithium-ion single cell. The influencing factors include the internal current of the lithium-ion single cell and the magnetic susceptibility of the lithium-ion single cell material. The magnetic susceptibility is an interference signal. Specifically, the full current law expression in Maxwell's equations is as follows:

[0059]

[0060] In the formula, the left part ∮ l H·dl is the line integral of the magnetic field on the closed path l, which represents the circulation of the magnetic field on the closed path, that is, the induced magnetic field generated by the internal current of the battery in space. Among them, H represents the magnetic field intensity, l represents the closed loop, that is, the current path, and dl represents the displacement element vector; the first term on the right side of the formula ∫∫ sJ·ds represents the conduction current, J represents the current density, s is an arbitrary surface surrounded by l, and ds represents the area element vector; the second term on the right side of the formula represents the time-varying displacement current, It represents the inverse of the electric displacement field D relative to time t, that is, the density of the displacement current, and ds also represents the area element vector.

[0061] Positive electrode electrochemical reaction formula:

[0062]

[0063] Negative electrode electrochemical reaction formula:

[0064]

[0065] Overall reaction formula:

[0066]

[0067] The relationship between the induced magnetic field and current in the circuit: ∮ l H·dl=∫∫ s J·ds

[0068] The relationship between magnetic field change and magnetic permeability change:

[0069] B=μ0(1+χ m )H

[0070] μ0—magnetic permeability of vacuum.

[0071] χ m —Magnetic susceptibility, which indicates the magnetization ability of a substance to an external magnetic field.

[0072] H— represents the magnetic field strength.

[0073] B—represents the magnetic induction intensity.

[0074] A magnetic field change acquisition unit, used to obtain the magnetic field change on the surface of the lithium-ion single cell based on the internal current of the lithium-ion single cell;

[0075] The monitoring position acquisition unit is used to scan the external induced magnetic field of the lithium-ion single battery during the charge and discharge cycle using a distributed fluxgate sensor array, and obtain the location where the magnetic field change on the lithium-ion single battery surface meets the preset conditions and is away from interference signals as the monitoring position. The preset condition is the location where the magnetic field change on the battery surface is the most obvious and stable.

[0076] like Figure 6, and finally the following monitoring points are selected: the monitoring points need to avoid the edge of the battery (the edge of the battery is in contact with the air, and may be more complex in design and physical structure. For example, the edge may contain connection points, contact points, shell packaging, etc. These structures may cause irregular or restricted current flow. Therefore, the current distribution at the edge is often different from that in the center of the battery, resulting in more unstable changes in the edge magnetic field); avoid battery electrodes (changes in the magnetic susceptibility of the electrode material during battery charging and discharging will cause large changes in the battery's induced magnetic field, which will interfere with the battery's induced magnetic field monitoring results); cover the battery as much as possible (to avoid problems with a single probe that will cause the monitoring results to be invalid). The placement method can be designed according to the shape and size of the monitored battery and the monitoring requirements. Taking the embodiment as an example, according to the size and shape of the monitored battery, a three-channel probe is selected. After the probe is inserted into the package and fixed with screws, the package is parallel to the front surface of the battery, tightly fitted in the middle of the battery and fixed. As Figure 7 shown.

[0077] The weak magnetic monitoring technology used in the present invention is a kind of magnetic field signal, namely the physical quantity of magnetic induction intensity (unit, Tesla, T, actual measurement is generally nT, 1nT = 10 -9 T) measurement non-destructive monitoring technology, which does not require the application of an excitation field source when implemented. When the intrinsic magnetic characteristic parameters (such as magnetic susceptibility, coercive force, etc.) of the monitored object change, there are internal defects, and the parameters that cause the magnetic field change (such as voltage, current, temperature, etc.) within the object change, a magnetic anomaly signal will be generated, such as Figure 5As shown. Magnetic anomaly is defined as the curve of the magnetic induction intensity signal measured by the sensor showing an upward or downward bulge. During the electrochemical reaction of lithium-ion batteries, there is a significant current distribution in the positive and negative electrode materials and the current collector. Based on the full current law in Maxwell's equations, the movement of charges inside the battery and the time-varying characteristics of the current will excite an induced magnetic field in the space around it. In view of the low time-varying gradient of the current distribution inside the lithium-ion battery (the migration of lithium ions in the electrolyte depends on the diffusion process, and its diffusion coefficient is very small, which makes it difficult for the lithium ion current distribution to change rapidly during the charging and discharging process, and the current changes slowly over time. The time-varying gradient of the current distribution inside the lithium-ion battery is low. The "lower" here is a relative concept, and there is no absolutely fixed threshold to define it. In specific research and applications, a relatively reasonable range will be determined through experimental measurement and data analysis based on different battery systems, working conditions, and specific performance indicators of concern, so as to judge whether the gradient is at a "low" level. For example For some lithium-ion batteries, when the rate of change of current density per unit time and unit space is less than a certain value (e.g., less than 1 milliampere per second per square centimeter), the time-varying gradient of the current distribution may be considered low, but this value varies depending on the specific battery. Furthermore, the circuit exhibits significant large-scale structural characteristics. The large-scale structural characteristics of the internal circuit of a lithium-ion battery mainly include the following aspects: electrode structure, current collector, internal battery connection structure, and overall battery packaging structure. It is reasonable to infer that the induced magnetic field is mainly derived from the internal current. When a lithium-ion battery is charged and discharged, changes in the internal battery current directly affect changes in the external induced magnetic field. Changes in the internal battery current are the source of changes in the external magnetic field, and changes in the internal battery current represent changes in various battery parameters. Therefore, by placing a high-precision distributed magnetic sensor array on the surface of lithium-ion single cells, the external induced magnetic field of the battery is continuously monitored throughout the battery's complete charge and discharge cycle. By analyzing the changes and patterns in the monitored magnetic field data, the internal changes of the lithium-ion battery during the charge and discharge cycle are obtained, thereby assessing the health status of the lithium-ion battery.

[0078] The magnetic signal acquisition module is used to collect the magnetic signal data generated by the lithium-ion single battery measured by the multi-channel magnetic field measurement module.

[0079] The signal processing module is used to convert the pre-processed magnetic signal data into a curve graph, and complete the classification evaluation of the health status of lithium-ion single batteries in combination with the battery health status evaluation model.

[0080] A further embodiment is that the signal processing module includes:

[0081] The signal preprocessing unit is used to select 100 magnetic signal data within one second, remove the highest and lowest values, and calculate the average value of the remaining magnetic signal data as the magnetic field monitoring data value of this second;

[0082] A curve generating unit, for generating a magnetic field change data curve based on magnetic field monitoring data values ​​and corresponding time data;

[0083] The feature extraction unit is used to extract the segmented time-domain magnetic characteristic values ​​of the magnetic field change data curve; specifically, the monitoring magnetic field data is read, wherein the first column of data is read and converted into the X-axis time value in seconds (s); the 2-N columns of data are read and converted into the Y-axis magnetic field magnitude value in nanotesla (nT), the data is segmented by inputting the X-axis value of the segmentation point, and the segmented time-domain characteristic values ​​of the data are extracted through a pre-written program, including variance, standard deviation, root mean square, peak-to-peak value, and kurtosis value.

[0084] Specifically, the collected raw magnetic field data is first read and preprocessed, which includes merging data worksheets, deleting NaN values ​​in the data of each channel through loop traversal to ensure data continuity, and downsampling the processed data by a factor of 100 to reduce the data volume.

[0085] Then, for each channel's data, the user manually specifies four split points, dividing the data into three intervals, and calculates the time domain statistical characteristics (variance, standard deviation, mean, peak-to-peak value) and waveform characteristics (kurtosis value, root mean square, kurtosis factor). A total of five magnetic field characteristic values ​​are extracted using the following formula to characterize the battery charging and discharging status information:

[0086] F1=Var(x)=E[(x-μ) 2 ]

[0087]

[0088] F5=ptp=max(data(interval))-min(data(interval))

[0089] Here, x represents the sample value, μ represents the sample mean, N represents the total number of samples, data(interval) represents the data set within the specified interval, F1 represents the degree of data variation, F2 represents the average degree of data deviation from the mean, F3 represents the "amplitude" of the data, and is often used to describe the energy or fluctuation amplitude of the signal, and F4 is used to measure the sharpness of the data distribution or the thickness of the tail. F5 is used to represent the total amplitude of data or signal fluctuations and is often used to describe the maximum range of variation of the signal. The above five eigenvalues ​​are constructed into an eigenvector F = (F1, F2, F3, F4, F5), and finally the time domain waveforms of the three channels are plotted to intuitively display the data characteristics.

[0090] The model building unit is used to build a battery health status assessment model by using segmented time-domain magnetic eigenvalues ​​combined with the k-NN algorithm.

[0091] Specifically, the original magnetic field monitoring data stored in the host computer is converted into a curve graph. During the conversion process, the data needs to be preprocessed. 100 magnetic field data points are selected every second. After removing the highest and lowest values, the average value of the remaining magnetic field data is obtained as the magnetic field monitoring data value of this second. The time data of the original magnetic field monitoring data stored in the host computer is the actual time in the form of hours: minutes: seconds. The time data is converted into second data starting from 0 through the code. Secondly, the monitoring magnetic field change data curve is analyzed. The curve can be divided into five stages, such as Figure 8 As shown in the figure: the first stage is a gradual and steady extension of the magnetic field intensity or a slight decrease, corresponding to the constant current charging stage of the battery charging and discharging process; the second stage is a sudden increase in the magnetic field intensity followed by a gradual stabilization, corresponding to the constant voltage charging stage of the battery charging and discharging process; the third stage is a cliff-like rise in the magnetic field intensity, corresponding to the rest stage between charging and discharging in the battery charging and discharging process; the fourth stage is when the magnetic field intensity stops its cliff-like rise and then decreases and then extends horizontally or directly horizontally, corresponding to the constant current discharge stage of the battery charging and discharging process; the fifth stage is when the horizontal trend of the magnetic field intensity ends and then drops abruptly, corresponding to the rest stage after discharge in the battery charging and discharging process. After the five stages are completed, the battery completes a charge and discharge cycle. Based on the above content, the variation pattern of the induced magnetic field intensity monitoring data is summarized; then, feature extraction is performed on the monitored magnetic field change data curve. According to the characteristics and curve analysis of magnetic field monitoring data: magnetic field monitoring data is collected for fixed-frequency monitoring, and the time domain eigenvalues ​​of the data curve are extracted; the battery charge and discharge exist in different stages, and the data curve is segmented and the time domain magnetic eigenvalues ​​are extracted; finally, the extracted segmented time domain magnetic eigenvalues ​​are extracted and input into the constructed data set, and the health status of the monitored lithium-ion batteries is classified and evaluated in combination with the k-NN algorithm model.

[0092] A further embodiment is that the model building unit includes:

[0093] The dataset construction subunit is used to divide the segmented time-domain magnetic eigenvalues ​​into a training dataset and a test dataset; the training dataset is the segmented time-domain magnetic eigenvalues ​​that mark the health status categories of lithium-ion single cells; the test dataset is the segmented time-domain magnetic eigenvalues ​​to be evaluated; specifically, the battery health status categories are divided and the sample order is randomly shuffled to avoid the data order affecting the fairness of the model; then, the training set and test set are extracted separately according to the battery health status category to ensure that the proportion of each category in the training dataset and the test dataset is consistent.

[0094] The k value determination subunit is used to determine the k value of the k-NN algorithm through a cross-validation method;

[0095] a Euclidean distance calculation subunit, configured to calculate the Euclidean distances of the segmented time-domain magnetic characteristic values ​​in the test data set and the training data set using a k-NN algorithm with a determined k value;

[0096] A neighboring sample acquisition subunit is used to obtain neighboring sample data points in the training dataset that meet a similarity threshold with the test dataset based on Euclidean distance;

[0097] The health status assessment subunit is used to obtain the health status assessment result of the lithium-ion single battery of the test data set based on the health status category of the adjacent sample data points.

[0098] Specifically, the dataset is normalized (k-NN is based on distance measurement, and differences in feature scales will lead to distance calculation deviations); then the k-NN model is trained, and the number of neighbors is set to 1, that is, the nearest neighbor is selected for classification. Then, the distance between samples is automatically calculated through the built-in k-NN classifier, and the category is determined by "majority voting" to achieve prediction of the training dataset and the test dataset.

[0099] According to the formula:

[0100] The accuracy of the training and test datasets is calculated and displayed as a prediction comparison chart. A confusion matrix (rows: true class, columns: predicted class, diagonal elements: number of correct classifications, off-diagonal elements: number of misclassifications) is then used to display the misclassification between classes. The evaluation criterion is based on the closest training data point within the selected "k" value range. The type of the training data point is the evaluation result of the test data point, and the evaluation form is a graded category (good / bad).

[0101] A further implementation method is to improve the k-NN algorithm by introducing an automatic order-shuffling layer and a data normalization transposition layer, and adding a confusion matrix to evaluate the health status assessment results of lithium-ion single cells.

[0102] Specifically, the core of the automatic shuffling function is to use the randperm function based on a pseudo-random number generator and the Fisher-Yates shuffling algorithm to generate a random permutation sequence containing all sample indices, and then rearrange the rows of the dataset according to this sequence to eliminate deviations such as category concentration, temporal dependence, or experimental batches that may be caused by the original data order. This process first obtains the total number of samples in the dataset, uses the algorithm to generate random indexes, and maps row by row to randomize the sample order while keeping the correspondence between features and labels unchanged. For category-imbalanced data, stratified sampling can be combined after shuffling to ensure that the proportion of each state category in the training and test datasets is balanced; for battery cycle data with strong temporal sequence, block shuffling can be used to retain local dependencies rather than completely shuffling the sample order; for multi-table related data, the same random index sequence must be used for synchronous shuffling to avoid misalignment of samples and features. This operation has low computational complexity and can ensure reproducible results by setting a random seed, which can effectively improve the fairness and reliability of model training and evaluation.

[0103] Normalization uses the formula: Scaling the data to the range [0,1] ensures that different features have the same scale, thus preventing some features from having too much influence on the model due to their large numerical range.

[0104] The data is transposed because the k-NN classifier requires the input feature matrix to be in sample × feature format, and the normalized data format is in feature × sample format. If the transposition is not performed, the classification logic will be completely wrong. The transposition is to avoid confusion of sample dimensions, so that the normalization function can correctly handle the dimension of each feature and ensure that all sample values ​​of the same feature are uniformly scaled.

[0105] The evaluation criteria is to see which training data point is closest to the test data point within the selected "k" value range. The type of the training data point is the evaluation result of the test data point, and the evaluation form is a grade category (good / bad).

[0106] Example 2

[0107] The present invention also provides a lithium-ion single-cell battery monitoring method based on weak magnetic field detection technology, applying the system of the first embodiment, including:

[0108] Measure the magnetic signal data of lithium-ion single batteries through a distributed fluxgate sensor array;

[0109] Collect magnetic signal data generated by lithium-ion single batteries measured by a multi-channel magnetic field measurement module;

[0110] The preprocessed magnetic signal data is converted into a curve graph, and combined with the battery health status assessment model to complete the classification assessment of the health status of lithium-ion single batteries.

[0111] A further embodiment is that the method for obtaining the monitoring position of the distributed fluxgate sensor array includes:

[0112] Based on Maxwell's equations, magnetic permeability, and the electrochemical reaction principle of lithium-ion cells, the factors affecting the external induced magnetic field of lithium-ion cells are obtained. The influencing factors include the internal current of the lithium-ion cell and the magnetic susceptibility of the lithium-ion cell material. The magnetic susceptibility is an interference signal.

[0113] Based on the internal current of the lithium-ion single cell, the magnetic field change on the surface of the lithium-ion single cell is obtained;

[0114] A distributed fluxgate sensor array is used to scan the external induced magnetic field of the lithium-ion single battery during the charge and discharge cycle, and the position where the magnetic field change on the surface of the lithium-ion single battery meets the preset conditions and is far away from the interference signal is obtained as the monitoring position.

[0115] This embodiment also provides specific operating steps for the system of embodiment 1:

[0116] (1) First, it is necessary to pre-design the battery charge and discharge cycle operation steps and various specific parameters based on the battery parameters.

[0117] (2) Determine the position where the magnetic sensor probe is fixed on the battery based on the shape and size of the battery and monitoring requirements.

[0118] (3) After the monitored battery is connected and fixed to the magnetic sensor, place it at the monitoring position, then open the upper computer software of the weak magnetic monitoring and check the connection status. If there is any abnormality, reconnect the battery to the instrument and check the battery status.

[0119] (4) The fourth step is to add the designed parameters and operation steps to the host computer software and set the data saving format, that is, to complete the setting of the battery charging and discharging process parameters.

[0120] (5) In the fifth step, select the channel for monitoring the battery and set the name of the work step to be executed. After confirmation, enter the experimental information of the monitored battery (including the information of the monitored battery, the monitoring date and conditions); then click the multi-channel magnetic field measurement system software to receive and save. The magnetic measurement system starts to synchronously collect data and stores the data transmission in the host computer.

[0121] (6) After completing the battery charge and discharge cycle monitoring, the magnetic field monitoring data and electrochemical monitoring data stored in the host computer are exported and organized into new data files respectively. The data are then plotted as curves using a self-written drawing program and compared and analyzed.

[0122] Example 3

[0123] This embodiment monitors and verifies five groups of lithium-ion batteries. Since the differences between the magnetic field monitoring data curves presented by each group are extremely subtle, in order to more concisely and effectively display the monitoring results, it is decided to select the magnetic field monitoring data curve of battery No. 16 as the final monitoring verification result graph for presentation (this embodiment uses a 3-channel fluxgate sensor for monitoring based on the size and shape of the tested battery and the monitoring requirements). The magnetic monitoring data result graph of battery No. 16 is shown in the figure below. Figure 9 (a) and Figure 9 As shown in (b), the electrochemical monitoring data results of battery No. 16 are shown in the figure Figure 10 (a) and Figure 10 (b) shown.

[0124] Comparing the electrochemical monitoring data verification results with the results of the monitoring method proposed by the present invention, the monitoring results of the present invention are more intuitive and obvious.

[0125] The magnetic monitoring data curves for the pilot and pilot tests show significant differences within the red circle. The pilot test cell exhibits sharp fluctuations within the red circle, while the pilot cell's curve is smooth and stable. While the electrochemical monitoring data curves exhibit minor differences in data size or duration, the same significant fluctuations as observed in the magnetic monitoring results are not observed.

[0126] Based on the induced magnetic field intensity change data of the lithium-ion battery charge and discharge cycle obtained by monitoring, the established time-domain magnetic signal feature extraction method is used to extract the magnetic signal eigenvalues ​​(in this embodiment, five eigenvalues ​​are extracted for each of the three curve data segments, for a total of fifteen eigenvalues), and an input magnetic signal eigenvalue data set is constructed. Category "1" represents lithium-ion batteries in the pilot state, and category "2" represents lithium-ion batteries in the small-scale test state. Then, the k-NN algorithm is used to perform classification and recognition calculations on the data set, with the k value selected as "2". 70% of the data in the data set is stored as training data in the training data set (TrainData), and the remaining 30% of the data is stored as test data in the prediction data set (Test Data) to train the model; finally, simulation tests are performed on the training data set and the test data set, and the performance evaluation results (including the prediction result accuracy and confusion matrix) are output. Figure 11 This is a diagram showing the performance evaluation results of the training data set according to an embodiment of the present invention; wherein, Figure 11 (a) is the line chart of the prediction results. Figure 11 (b) is the confusion matrix; Figure 12 This is a performance evaluation result diagram of the test data set according to the embodiment of the present invention; wherein, Figure 12 (a) is the line chart of the prediction results. Figure 12 (b) is the confusion matrix.

[0127] The verification results of single lithium-ion batteries in different health states in the same environment are as follows:

[0128] The data was collected from 49 pilot-scale (healthy) battery data sets and 56 small-scale (poor) battery data sets, totaling 105 data sets, all in different health states. The classification results show that the model's prediction accuracy is high, reaching 94.5% for the training dataset. The model's accuracy in both missed and misclassified categories "1" and "2" is strong, judging by its precision, recall, and F1 score. The model's prediction accuracy in the test dataset was 87.5%. While the model exhibited some misclassifications for category "1" and some missed classifications for category "2," the overall performance was lower than that of the training dataset, but the difference was small, indicating that the model has good generalization capabilities and is relatively stable in predicting new data.

[0129] In summary, the technical solution of the present invention has high application value in the engineering field. The weak magnetic field monitoring system and method proposed in this invention, as a supplement to existing monitoring technologies, can achieve non-destructive in-situ monitoring of the complete charge and discharge cycle of single lithium-ion batteries. Combined with the feature value extraction method and the k-NN algorithm model, the battery health status is classified and evaluated. The accuracy rate of classifying and identifying lithium-ion batteries in different health states and before and after different operating conditions reaches 87.5%. In other words, the battery is finally classified as a pilot cell or a small-scale test cell, and the health status of the lithium-ion battery can be determined. It has room for application in fields such as electric vehicles and is of great significance for ensuring the safe operation of equipment.

[0130] The embodiments described above are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by persons skilled in the art should fall within the scope of protection defined by the claims of the present invention.

Claims

1. The lithium-ion single battery monitoring system based on weak magnetic detection technology is characterized by: include: A multi-channel magnetic field measurement module is used to measure the magnetic signal data of lithium-ion single batteries through a distributed fluxgate sensor array; A magnetic signal acquisition module is used to collect magnetic signal data generated by the lithium-ion single battery measured by the multi-channel magnetic field measurement module; The signal processing module is used to convert the pre-processed magnetic signal data into a curve graph, and complete the classification evaluation of the health status of the lithium-ion single battery in combination with the battery health status evaluation model.

2. The system according to claim 1, wherein: The distributed fluxgate sensor array used in the multi-channel magnetic field measurement module has a detection sensitivity direction perpendicular to the cylindrical end face, and the actual measured magnetic signal data is the external induced magnetic field signal at different positions on the surface of the lithium-ion single battery at the same time during the charge and discharge cycle.

3. The system according to claim 2, characterized in that The multi-channel magnetic field measurement module includes: An influencing factor acquisition unit is used to obtain influencing factors of the external induced magnetic field of the lithium-ion single cell based on Maxwell's equations, magnetic permeability, and the electrochemical reaction principle of the lithium-ion single cell; wherein the influencing factors include the internal current of the lithium-ion single cell and the magnetic susceptibility of the lithium-ion single cell material, and the magnetic susceptibility is an interference signal; A magnetic field change acquisition unit, configured to obtain a surface magnetic field change of the lithium-ion single cell based on the internal current of the lithium-ion single cell; The monitoring position acquisition unit is used to use a distributed fluxgate sensor array to scan the external induced magnetic field of the lithium-ion single battery during the charge and discharge cycle, and obtain the position where the magnetic field change on the surface of the lithium-ion single battery meets the preset conditions and is far away from the interference signal as the monitoring position.

4. The system according to claim 1, wherein: The signal processing module includes: The signal preprocessing unit is used to select 100 magnetic signal data within one second, remove the highest and lowest values, and calculate the average value of the remaining magnetic signal data as the magnetic field monitoring data value of this second; a curve generating unit, configured to generate a magnetic field change data curve based on the magnetic field monitoring data values ​​and corresponding time data; A feature extraction unit, configured to extract segmented time-domain magnetic feature values ​​of the magnetic field change data curve; The model building unit is used to build a battery health status assessment model by using the segmented time-domain magnetic eigenvalues ​​in combination with the k-NN algorithm.

5. The system according to claim 4, characterized in that The model building unit includes: A data set construction subunit is used to divide the segmented time-domain magnetic eigenvalues ​​into a training data set and a test data set; wherein the training data set is the segmented time-domain magnetic eigenvalues ​​that mark the health status categories of lithium-ion single batteries; and the test data set is the segmented time-domain magnetic eigenvalues ​​to be evaluated; The k value determination subunit is used to determine the k value of the k-NN algorithm through a cross-validation method; a Euclidean distance calculation subunit, configured to calculate the Euclidean distances of the segmented time-domain magnetic characteristic values ​​in the test data set and the training data set using a k-NN algorithm with a determined k value; A neighboring sample acquisition subunit, configured to obtain neighboring sample data points in the training dataset that meet a similarity threshold with the test dataset based on the Euclidean distance; The health status evaluation subunit is configured to obtain a health status evaluation result of the lithium-ion single battery of the test data set based on the health status category of the adjacent sample data points.

6. The system according to claim 4, characterized in that Improvements to the k-NN algorithm include introducing an automatic order-shuffling layer and a data normalization transposition layer, and adding a confusion matrix to evaluate the health status assessment results of lithium-ion single cells.

7. A lithium-ion single battery monitoring method based on weak magnetic field detection technology, using the system according to any one of claims 1 to 6, characterized in that: include: Measure the magnetic signal data of lithium-ion single batteries through a distributed fluxgate sensor array; Collect magnetic signal data generated by lithium-ion single batteries measured by a multi-channel magnetic field measurement module; The pre-processed magnetic signal data is converted into a curve graph, and combined with the battery health status assessment model to complete the classification assessment of the health status of the lithium-ion single battery.

8. The method according to claim 7, characterized in that The method for obtaining the monitoring position of the distributed fluxgate sensor array includes: Based on Maxwell's equations, magnetic permeability, and the electrochemical reaction principle of lithium-ion single cells, factors affecting the external induced magnetic field of the lithium-ion single cells are obtained; wherein the influencing factors include the internal current of the lithium-ion single cells and the magnetic susceptibility of the lithium-ion single cell material, and the magnetic susceptibility is an interference signal; Obtaining a change in the surface magnetic field of the lithium-ion single cell based on the internal current of the lithium-ion single cell; A distributed fluxgate sensor array is used to scan the external induced magnetic field of the lithium-ion single battery during the charge and discharge cycle, and the position where the magnetic field change on the surface of the lithium-ion single battery meets the preset conditions and is far away from the interference signal is obtained as the monitoring position.

Images