A method and device for detecting abnormal internal resistance of an electric cell

Abstract

The application relates to the field of battery detection, and particularly relates to a detection method and device for abnormal internal resistance of a battery cell. The application divides continuous discharge conditions into corresponding electricity intervals based on an electricity-open circuit voltage curve, so that the relationship between voltage and internal resistance is more consistent in each interval, thereby providing a stable analysis window for accurate analysis. Secondly, the battery cell discharge condition data is associated to the corresponding electricity interval, feature data in the interval is obtained, and then valid electricity intervals are screened out, so that invalid data is eliminated. Then, the internal resistance deviation degree is calculated based on the decentralized voltage extreme value in the valid interval, the pressure difference deviation caused by the difference in the battery cell SOC is eliminated, the signal of the real internal resistance abnormality is amplified, and the instantaneous interference is avoided from being misjudged as abnormality. Finally, the abnormality is judged through the internal resistance deviation degree, the difference between the battery cells can be quantified, the real internal resistance abnormality and the instantaneous interference can be effectively distinguished, so that the internal resistance calculation error is reduced, the abnormality detection precision is improved, and the misjudgment and the missed detection are reduced.

Description

Technical Field

[0001] This invention relates to the field of battery cell testing, and specifically to a method and apparatus for detecting abnormal internal resistance in battery cells. Background Technology

[0002] The current method for detecting abnormal internal resistance of lithium-ion batteries mainly uses the pulse current method. This method collects voltage drop and current changes at the start or end of charging, and calculates the cell internal resistance using the ratio of voltage change to current change. This allows for the assessment of battery health, potential faults, and the impact on charging performance and vehicle power performance.

[0003] The pulse current method relies on short-time single-frame pulse data for calculation. There is a time delay in voltage and current acquisition, resulting in poor data synchronization. Furthermore, it is susceptible to noise and sampling frequency limitations, leading to low data quality. At the same time, this method is easily affected by dynamic operating conditions such as temperature, SOC, and current ratio. It does not effectively compensate for differences in temperature distribution and SOC, and uses a fixed threshold to judge anomalies, which cannot distinguish between real internal resistance anomalies and transient interference. This results in large internal resistance calculation errors, insufficient anomaly detection accuracy, and high false positive and false negative rates. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a method and apparatus for detecting abnormal internal resistance of battery cells, in order to solve the problems of being unable to distinguish between real internal resistance abnormalities and transient interference, resulting in large internal resistance calculation errors, insufficient abnormality detection accuracy, and high false positive and false negative rates.

[0005] In a first aspect, embodiments of the present invention provide a method for detecting abnormal internal resistance of a battery cell, the method comprising: The discharge condition data of each cell in the battery are obtained, and the capacity range of the battery in each open circuit voltage range is obtained, wherein the capacity range is obtained based on the capacity-open circuit voltage curve. The discharge condition data of each cell is correlated with the corresponding power range to obtain the characteristic data of each cell within the corresponding power range. Analyze the characteristic data of the battery cells within each of the stated power ranges to obtain the effective power ranges; Based on the decentralized voltage extreme value corresponding to each cell within the effective power range, the degree of deviation of the internal resistance of the cell is calculated, and the internal resistance of the cell is determined to be abnormal based on the degree of deviation.

[0006] Furthermore, obtaining the battery's capacity range corresponding to each open-circuit voltage range includes: Obtain the battery's charge-open-circuit voltage curve, wherein the charge-open-circuit voltage curve includes multiple basic voltage ranges; Calculate the rate of change of voltage for each base voltage range in the charge-open-circuit voltage curve; The width of the corresponding base voltage range is adjusted according to the voltage change rate, and the adjusted base voltage range is determined as the open circuit voltage range. Query the electrical quantity range corresponding to each of the aforementioned open-circuit voltage ranges.

[0007] Furthermore, the analysis of the characteristic data of the battery cells within each of the stated power ranges yields the effective power ranges, including: The decentralized voltage value of each cell is calculated using the measured voltage value and theoretical open-circuit voltage of the cells within each of the aforementioned power ranges. The voltage fluctuation of each cell is analyzed based on the decentralized voltage value, and the effective power range is determined based on the fluctuation.

[0008] Furthermore, the step of analyzing the voltage fluctuation of each cell based on the decentralized voltage value, and determining the effective capacity range based on the fluctuation, includes: Determine whether the decentralized voltage value of all cells within the power range falls within the preset voltage range; If all fall within the preset voltage range, then obtain the voltage variance of each cell within the power range at each time, and select the maximum voltage variance from multiple voltage variances. Compare the maximum voltage variance with the preset voltage variance; If the maximum voltage variance is greater than the preset voltage variance, then clustering is performed using the voltage standard deviation of each cell within the power range to identify outlier cells. If outlier cells exist, the range of battery capacity containing the outlier cells will be defined as the effective range of battery capacity.

[0009] Furthermore, the calculation of the internal resistance deviation of each cell based on the decentralized voltage extreme value corresponding to each cell within the effective charge range includes: The maximum and minimum decentralized voltage values ​​are obtained from the decentralized voltage values ​​of each cell within the effective power range. Based on the maximum decentralized voltage value and the maximum feedback current, calculate the first internal resistance deviation value in the charging direction, and based on the minimum decentralized voltage value and the maximum discharge current, calculate the second internal resistance deviation value in the discharge direction. Based on the first internal resistance deviation value and the second internal resistance deviation value, the initial internal resistance deviation degree is calculated; By utilizing the relationship between charge, temperature, and discharge DC internal resistance, the initial internal resistance deviation is corrected to obtain the corresponding internal resistance deviation of the battery cell.

[0010] Furthermore, the method of correcting the initial internal resistance deviation by utilizing the relationship between charge, temperature, and discharge DC internal resistance to obtain the corresponding internal resistance deviation of the battery cell includes: Obtain the temperature of each cell at each time point within the effective charge range; The temperature difference at each moment is calculated based on the temperature of each cell, and the moment with the largest temperature difference is determined from the temperature differences at each moment. Obtain the lowest and highest temperatures corresponding to the moment when the temperature difference is greatest; Using the lowest temperature and the corresponding charge of the battery cell, the first DC internal resistance is retrieved from the relationship between charge, temperature and discharge DC internal resistance; and using the highest temperature and the corresponding charge of the battery cell, the second DC internal resistance is retrieved from the relationship between charge, temperature and discharge DC internal resistance. The initial internal resistance deviation is corrected by using the difference between the second DC internal resistance and the first DC internal resistance to obtain the internal resistance deviation of the cell.

[0011] Furthermore, the method of correcting the initial internal resistance deviation by utilizing the relationship between charge, temperature, and discharge DC internal resistance to obtain the corresponding internal resistance deviation of the battery cell includes: The temperature of each cell within the effective power range at each time is obtained, and the temperature difference at each time is calculated based on the temperature of each cell. By utilizing the temperature difference and charge of each cell at various times within the effective charge range, the real-time DC internal resistance at each time is obtained by interpolation from the charge-temperature-discharge DC internal resistance relationship. The average DC internal resistance is calculated based on the real-time DC internal resistance, and the dynamic compensation value is calculated based on the average DC internal resistance and the preset reference DC internal resistance. The degree of internal resistance deviation of the cell is calculated based on the initial internal resistance deviation and the dynamic compensation value.

[0012] Furthermore, determining whether the internal resistance of the battery cell is abnormal based on the degree of deviation includes: Cells with internal resistance deviation greater than zero are selected as candidate abnormal cells, and the target effective charge range for triggering abnormalities in the candidate abnormal cells is determined. Query the number of times the candidate abnormal battery cell triggers an anomaly within the target effective power range, and query whether the candidate abnormal battery cell triggers an anomaly within the adjacent power range of the target effective power range; If the number of abnormal triggers is a preset number and no abnormality is triggered within the adjacent power range, then it is determined that the internal resistance of the candidate abnormal cell is not abnormal; or, if the number of abnormal triggers is greater than the preset number, and / or an abnormality is triggered within the adjacent power range, then it is determined that the internal resistance of the candidate abnormal cell is abnormal.

[0013] Furthermore, the method also includes: Candidate abnormal cells that trigger anomalies more than a preset number of times and / or trigger anomalies within the adjacent power range are designated as abnormal cells to be observed. The abnormality of the observed abnormal cell is continuously monitored over multiple time periods. If the abnormal situation is that the abnormality is triggered continuously within multiple time periods, then the abnormal cell to be observed is determined to be a real abnormal cell; or, if the abnormal situation is that the abnormality is not triggered continuously within multiple time periods, then the abnormal cell to be observed is determined to be a non-real abnormal cell.

[0014] Secondly, embodiments of the present invention provide a device for detecting abnormal internal resistance of a battery cell, the device comprising: The acquisition module is used to acquire discharge condition data of each cell in the battery and acquire the capacity range corresponding to each open circuit voltage range of the battery, wherein the capacity range is obtained based on the capacity-open circuit voltage curve. The processing module is used to associate the discharge condition data of each cell with the corresponding power range to obtain the characteristic data of each cell within the corresponding power range. The analysis module is used to analyze the characteristic data of the battery cells within each of the stated power ranges to obtain the effective power ranges; The calculation module is used to calculate the degree of internal resistance deviation of each cell based on the decentralized voltage extreme value corresponding to each cell within the effective power range, and to determine whether the internal resistance of the cell is abnormal based on the degree of internal resistance deviation.

[0015] Thirdly, embodiments of the present invention provide a computer device, including: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the method described in the first aspect or any corresponding embodiment thereof.

[0016] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing computer instructions that cause a computer to perform the method described in the first aspect or any of its corresponding embodiments.

[0017] This application divides the continuous discharge condition into corresponding charge intervals based on the charge-open-circuit voltage curve, making the relationship between voltage and internal resistance more consistent within each interval and providing a stable analysis window for accurate analysis. Secondly, by correlating the cell discharge condition data to the corresponding charge interval, characteristic data within the interval is obtained, and then the effective charge interval is selected, eliminating invalid data. Then, the degree of internal resistance deviation is calculated based on the decentralized voltage extreme value within the effective interval, eliminating the voltage difference offset caused by cell SOC differences, amplifying the signal of true internal resistance anomalies, and avoiding the misjudgment of transient interference as anomalies. Finally, judging anomalies by the degree of internal resistance deviation can quantify the differences between cells, effectively distinguish between true internal resistance anomalies and transient interference, thereby reducing internal resistance calculation errors, improving anomaly detection accuracy, and reducing misjudgments and missed detections. Attached Figure Description

[0018] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating a method for detecting abnormal internal resistance of a battery cell according to some embodiments of the present invention. Figure 2 This is a schematic diagram illustrating the relationship between charge, temperature, and discharge DC internal resistance according to some embodiments of the present invention; Figure 3 This is a flowchart illustrating a method for detecting abnormal internal resistance of a battery cell according to some embodiments of the present invention. Figure 4 This is a structural block diagram of a battery cell internal resistance abnormality detection device according to an embodiment of the present invention; Figure 5 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] According to embodiments of the present invention, a method and apparatus for detecting abnormal internal resistance of battery cells are provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0022] This embodiment provides a method for detecting abnormal internal resistance of battery cells. Figure 1 This is a flowchart of a method for detecting abnormal internal resistance of a battery cell according to an embodiment of the present invention, as shown below. Figure 1 As shown, the process includes the following steps: Step S101: Obtain the discharge condition data of each cell in the battery, and obtain the capacity range corresponding to each open circuit voltage range of the battery. The capacity range is obtained by dividing the capacity range based on the capacity-open circuit voltage curve.

[0023] In this embodiment, the battery is a whole composed of multiple cells connected in series and parallel. The cell is the smallest basic unit constituting the battery pack, and the voltage, capacity, temperature, and operating state of the battery pack are all determined by the state of each cell. The discharge condition data mainly includes information such as real-time individual cell voltage, total battery current, cell temperature, real-time state of charge (SOC), timestamp, and charge / discharge flag, which are used to reflect the electrical characteristics and operating state of the battery during the actual discharge process.

[0024] First, the battery discharge condition data is extracted daily based on the charge and discharge flag bits, and long-term low-current discharge data is removed based on the current change rate to complete the data cleaning. At the same time, the battery's corresponding capacity-open-circuit voltage curve is obtained, the open-circuit voltage is divided according to a fixed voltage interval, and the capacity interval corresponding to each open-circuit voltage interval is obtained by looking up a table.

[0025] It should be noted that the SOC-OCV curve, or SOC-Open Circuit Voltage curve, describes the one-to-one mapping between the cell's open circuit voltage (OCV) and the remaining capacity (SOC). At different capacity levels, the cell's equilibrium voltage is fixed and calibrable; therefore, this curve accurately reflects the voltage variation with capacity and is the core basis for capacity estimation, voltage decentralization, and range division.

[0026] Dividing the SOC-OCV curve according to fixed voltage intervals aims to segment the continuously changing discharge process into multiple intervals with clear characteristics and stable calculations, enabling refined analysis. By segmenting the voltage and then mapping it to the corresponding charge intervals, the algorithm can subdivide rapidly changing voltage regions and coarsely segment voltage plateau regions. This amplifies subtle voltage fluctuations caused by abnormal cell internal resistance while avoiding excessive invalid data that would waste computational resources, thus improving the sensitivity of anomaly detection and the efficiency of the algorithm.

[0027] Specifically, obtain the battery's capacity range corresponding to each open-circuit voltage range, including steps A1-A4: Step A1: Obtain the battery's charge-open-circuit voltage curve, which includes multiple base voltage ranges.

[0028] First, the open-circuit voltage-charge (SOC-OCV) curve corresponding to the current battery is selected as the basic data. This curve reflects the mapping relationship between the cell's open-circuit voltage and remaining charge. The voltage is segmented at fixed intervals to form multiple basic voltage ranges. For example, the value starts from 3.1V and is divided into ranges up to 4.4V in 0.1V increments, forming a series of basic voltage ranges such as 3.1V, 3.2V, 3.3V...4.4V. These basic ranges provide the initial division basis for subsequent dynamic width adjustment, while ensuring coverage of the battery's complete discharge voltage range, laying the foundation for subsequent range optimization and charge matching.

[0029] Step A2: Calculate the rate of change of voltage corresponding to each base voltage range in the charge-open-circuit voltage curve.

[0030] For each obtained baseline voltage range, its corresponding voltage change rate is calculated. Specifically, by obtaining the difference in charge between two adjacent baseline voltage points and combining it with the voltage range length, the voltage change amplitude per unit charge change is calculated, thus characterizing the voltage change rate of that range. In the linear region where voltage changes drastically, the voltage changes significantly with the charge, corresponding to a large voltage change rate; in the plateau region where voltage changes smoothly, the voltage changes only slightly with the charge, corresponding to a small voltage change rate, thereby distinguishing voltage ranges with different characteristics.

[0031] Step A3: Adjust the width of the corresponding base voltage range according to the voltage change rate, and determine the adjusted base voltage range as the open circuit voltage range.

[0032] Based on the calculated voltage change rate, the initial base voltage range is adaptively adjusted in width. For linear regions with large voltage change rates and significant voltage variations with charge, a smaller range width is used for subdivision to retain more subtle features; for plateau regions with small voltage change rates and gradual voltage changes, a larger range width is used for coarse subdivision to reduce computational load and improve algorithm efficiency. The range after width adjustment is used as the final open-circuit voltage range.

[0033] Understandably, a voltage change rate threshold is set. When the voltage change rate of a certain base voltage range is greater than the threshold, it is determined to be a linear voltage change region. The original 0.1V base range width is further refined to 0.01V. By increasing the number of sampling points, the subtle characteristics of cell voltage fluctuations are fully amplified, improving the sensitivity of internal resistance anomaly detection. When the voltage change rate is less than or equal to the threshold, it is determined to be a voltage plateau region. The original range width is retained or appropriately expanded to avoid redundant calculations and resource consumption. During the range adjustment process, it is ensured that each range is continuous and non-overlapping, completely covering the entire operating voltage range of the battery from 3.1V to 4.4V, ultimately forming an adaptive open-circuit voltage range that combines coarse and fine parameters and matches the characteristics of the SOC-OCV curve.

[0034] Step A4: Query the electrical quantity range corresponding to each open-circuit voltage range.

[0035] Based on the adjusted open-circuit voltage range, a segment-by-segment query and matching is performed in the pre-calibrated open-circuit voltage-electrical quantity correspondence table. For each determined upper and lower limits of the open-circuit voltage range, the corresponding electrical quantity value is found, thereby determining the electrical quantity range covered by that open-circuit voltage range.

[0036] For example, 3.1V corresponds to 0% SOC and 3.2V corresponds to 1.8% SOC. Therefore, the charge range corresponding to the open circuit voltage range of 3.1V-3.2V is 0%-1.8%. By completing the mapping of all ranges in sequence, a set of charge ranges for subsequent data grouping is obtained.

[0037] Step S102: Associate the discharge condition data of each cell with the corresponding power range to obtain the characteristic data of each cell in the corresponding power range.

[0038] In this embodiment, the real-time discharge data of each cleaned cell is matched to the corresponding power range based on the current power of the cell, thus completing the association and matching of data and range; within the same power range, the cell voltage is decentralized, that is, the real-time voltage of the cell is subtracted from the open circuit voltage corresponding to the current power to obtain a decentralized voltage sequence, forming the feature data of each cell for anomaly analysis within the corresponding power range.

[0039] For example, if the real-time discharge capacity of a battery cell is 25% at a certain moment, and this capacity falls within the pre-defined capacity range of 19.1% to 30.3%, the system will classify the discharge data such as voltage, current, and temperature at that moment into this capacity range. After the data of all moments within this capacity range is collected, a decentralized calculation is performed on each sampling point within the range. The decentralized voltage is obtained by subtracting the open-circuit voltage (OCV) corresponding to the current capacity from the real-time individual cell voltage. All decentralized voltages within the range are arranged in chronological order to form the decentralized voltage sequence of the battery cell within this capacity range.

[0040] Step S103: Analyze the characteristic data of the cells in each power range to obtain the effective power range.

[0041] In this embodiment of the application, a progressive condition judgment is performed on the decentralized voltage feature data in each power range. First, it is checked whether the voltage meets the preset positive and negative fluctuation range. Then, it is verified whether the cell voltage fluctuation variance reaches the effective threshold. Finally, a density clustering algorithm is used to identify whether there are outlier cells. Only the range that meets all the triggering conditions in sequence is determined to be an effective power range with anomaly analysis value.

[0042] Specifically, the characteristic data of the battery cells within each capacity range are analyzed to obtain the effective capacity range, including steps B1-B2: Step B1: Calculate the decentralized voltage value of each cell using the measured voltage value and theoretical open-circuit voltage of each cell within each power range.

[0043] The power ranges, defined based on the effective power range, are matched one by one with the discharge condition data after the day's cleaning, and the continuous discharge process is assigned to the corresponding range. For each cell and each moment of sampling data within each power range, the theoretical open-circuit voltage OCV_SOC_i at that power level is first obtained by querying the SOC-OCV curve using the cell's current power SOC. Then, the decentralized voltage C_cellv_i = cellv_i – OCV_SOC_i is calculated by subtracting the theoretical open-circuit voltage from the measured cell voltage cellv_i at that moment. This eliminates the overall voltage difference caused by the consistency of cell SOC, highlights the true voltage fluctuation, and forms the basic characteristic data for subsequent judgment.

[0044] Step B2: Analyze the voltage fluctuation of each cell based on the decentralized voltage value, and determine the effective power range based on the fluctuation.

[0045] The process involves analyzing the voltage fluctuations of each battery cell based on decentralized voltage values ​​and determining the effective power range based on these fluctuations. This includes: determining whether the decentralized voltage values ​​of all cells within the power range fall within a preset voltage range; if they do, obtaining the voltage variance of each cell at each time point within the power range and selecting the maximum voltage variance from multiple voltage variances; comparing the maximum voltage variance with the preset voltage variance; if the maximum voltage variance is greater than the preset voltage variance, clustering is performed using the voltage standard deviations of each cell within the power range to identify outlier cells; if outlier cells exist, the power range containing outlier cells is determined as the effective power range.

[0046] Specifically, firstly, a global statistical analysis is performed on the decentralized voltage values ​​of all cells at all times within the power range. The minimum value (min[C_cellv_i]) and maximum value (max[C_cellv_i]) of the decentralized voltage within this range are identified. It is then determined whether the voltage fluctuation range meets the preset conditions: the minimum value is less than -0.005V and the maximum value is greater than 0.005V. This confirms whether there are sufficiently significant positive and negative voltage fluctuations within the range. If the decentralized voltages of all cells are concentrated near zero and do not exceed this preset voltage range, it indicates insufficient data fluctuation and the conditions for anomaly analysis are not met. Only when this fluctuation range is met does the next step of volatility assessment proceed.

[0047] Secondly, after confirming that the decentralized voltage meets the preset fluctuation range, the variance std(C_cellv_i_t) of the decentralized voltage at all sampling times is calculated for each cell within the power range. This variance quantifies the stability and fluctuation intensity of the voltage of a single cell. A larger variance indicates more severe voltage fluctuations and a higher likelihood of internal resistance anomalies. After obtaining the voltage variances of all cells, the largest variance is selected to comprehensively assess the maximum fluctuation level of cells within the power range, providing a key indicator for subsequent threshold comparisons.

[0048] Then, the obtained maximum voltage variance is compared with the preset voltage variance threshold of 0.005. If the maximum voltage variance is greater than the threshold, it indicates that there are cells with significantly abnormal fluctuations in the interval, and outlier identification continues. If it is less than or equal to the threshold, it is determined that there are no valid anomalies, and the interval is directly excluded. When the variance condition is met, the standard deviation of voltage corresponding to each cell is used as the feature value, and the DBSCAN density clustering algorithm is used to perform cluster analysis on all cells. Low-density data points that are sparsely distributed and far from dense clusters are marked as outliers, and outlier cells with abnormal fluctuations are identified.

[0049] Finally, after completing the DBSCAN clustering identification, it is determined whether outlier cells were successfully detected within the current charge range. If cells marked as outliers are found in the clustering results, it indicates that there are abnormal cells in this range whose behavior is significantly inconsistent with other normal cells, making them valuable for internal resistance anomaly analysis. At this point, the charge range is officially determined as a valid charge range, which can be used for subsequent internal resistance deviation calculation, temperature compensation, and multi-dimensional anomaly determination. If no outlier cells are identified, it means that there are no valid abnormal signals in this range, and it will not be used as a basis for subsequent analysis.

[0050] For example, taking a battery pack containing 100 cells as an example, the standard deviation of the decentralized voltage of each cell within this range, std(C_cellv_i_t), is first calculated, resulting in 100 standard deviation values ​​(e.g., 0.003V for cell 1, 0.004V for cell 2, ..., 0.012V for cell 88, 0.013V for cell 89). These values ​​are then used as feature values ​​to construct a one-dimensional dataset. Next, the DBSCAN clustering algorithm is applied, with a neighborhood radius ε=0.002V and a minimum sample size min_samples=5. The algorithm uses each data point as the core and searches for the number of samples contained in its neighborhood: the standard deviation of most cells is concentrated in the 0.003-0.006V range, and the number of samples in their neighborhood is ≥5, forming a high-density core cluster; while the standard deviations of cells 88 and 89 are 0.012V and 0.013V respectively, and there are no other samples in their neighborhood (sample size <5), so they are marked as data points in a low-density region. Finally, cells 88 and 89, corresponding to these low-density data points, were selected and identified as outliers within this energy range.

[0051] This application's embodiments obtain a decentralized voltage by calculating the difference between the measured voltage and the theoretical OCV voltage of a single cell within the same effective charge range. This eliminates the overall voltage difference offset caused by differences in cell SOC consistency, effectively amplifies the true voltage fluctuations of cells with abnormal internal resistance, and prevents weak abnormal signals from being overwhelmed by normal data. Secondly, by performing a two-layer progressive verification of the decentralized voltage using preset positive and negative fluctuation ranges and voltage variance thresholds, charge ranges with effective fluctuation characteristics can be quickly screened out, ensuring the reliability of subsequent analysis data quality. Then, based on the voltage standard deviation, DBSCAN density clustering is used to automatically identify outlier cells, realizing the quantitative, objective, and automated judgment of abnormal cells, significantly improving detection accuracy and robustness. Finally, only charge ranges that meet all conditions are determined as effective charge ranges, ensuring that internal resistance anomalies can be clearly quantified and easily identified, while eliminating invalid data interference, providing a stable and reliable data foundation for subsequent internal resistance calculation, temperature compensation, and multi-layer anomaly judgment.

[0052] Step S104: Based on the decentralized voltage extreme value corresponding to each cell within the effective power range, calculate the degree of deviation of the internal resistance of the cell, and determine whether the internal resistance of the cell is abnormal based on the degree of deviation.

[0053] In this embodiment, the extreme values ​​of the decentralized voltage of each cell are extracted within the effective charge range. The initial internal resistance deviation value is calculated in combination with the charging and discharging current. Temperature coupling compensation is then performed using the charge-temperature-DC internal resistance relationship to obtain the corrected internal resistance deviation value. Through three-layer detection of static threshold, dynamic fluctuation ratio, and multi-cycle trend consistency, instantaneous interference is filtered out. Finally, based on the continuous abnormal state, it is accurately determined whether there is a real internal resistance abnormality in the cell.

[0054] Specifically, based on the decentralized voltage extreme values ​​corresponding to each cell within the effective charge range, the degree of deviation of the internal resistance of the cell is calculated, including steps C1-C4: Step C1: Obtain the maximum and minimum decentralized voltage values ​​from the decentralized voltage values ​​of each cell within the effective power range.

[0055] A global statistical analysis is performed on the decentralized voltage values ​​of all cells within the effective charge range at all sampling times. This involves iterating through the decentralized voltage data for each cell and each time point, and determining the maximum decentralized voltage value `max[C_cellv_i_t]` and the minimum decentralized voltage value `min[C_cellv_i_t]` within the entire range through numerical comparison. These two extreme values ​​comprehensively reflect the most extreme voltage deviation of the cell within this range, accurately representing the difference between the cell and a normal cell, and providing the most representative characteristic data for subsequent calculation of internal resistance deviation.

[0056] Step C2: Calculate the first internal resistance deviation value in the charging direction based on the maximum decentralization voltage value and the maximum feedback current, and calculate the second internal resistance deviation value in the discharging direction based on the minimum decentralization voltage value and the maximum discharge current.

[0057] Based on the maximum and minimum decentralized voltage values ​​within the effective charge range, the internal resistance deviation values ​​in the charging and discharging directions are calculated separately. Dividing the maximum decentralized voltage value by the maximum feedback current set by the battery system yields the first internal resistance deviation value IR_C in the charging direction; dividing the minimum decentralized voltage value by the maximum discharge current set by the battery system yields the second internal resistance deviation value IR_D in the discharging direction. By calculating the deviation values ​​in both charging and discharging directions separately, the degree of voltage anomaly of the cell under different operating conditions can be comprehensively reflected, improving the accuracy and comprehensiveness of internal resistance anomaly assessment.

[0058] Step C3: Calculate the initial internal resistance deviation based on the first internal resistance deviation value and the second internal resistance deviation value.

[0059] After obtaining the first internal resistance deviation value IR_C in the charging direction and the second internal resistance deviation value IR_D in the discharging direction, the two values ​​are arithmetically averaged. The initial internal resistance deviation degree IR is obtained using the formula IR=(IR_C+IR_D) / 2, with the unit being milliohms (mΩ). This initial internal resistance deviation degree integrates the voltage offset information during the charging and discharging process, and can preliminarily quantify the degree of deviation of the cell's internal resistance from that of a normal cell, serving as the basis for the result before temperature compensation.

[0060] Step C4: Using the relationship between charge, temperature, and discharge DC internal resistance, the initial internal resistance deviation is corrected to obtain the corresponding internal resistance deviation of the cell.

[0061] Temperature coupling compensation is performed on the initial internal resistance deviation by using the correlation between charge, temperature, and discharge DC internal resistance. First, temperature data of all cells within the effective charge range are collected at various times. Then, the temperature difference between the highest and lowest temperatures at each time is calculated. The time with the largest temperature difference is determined, and the lowest and highest temperatures at that time are extracted. Next, based on the charge corresponding to that time, the first DC internal resistance minDCR corresponding to the lowest temperature and the second DC internal resistance maxDCR corresponding to the highest temperature are obtained by querying and interpolating from a preset relationship table. Finally, the formula is used: IR_temp=IR-(maxDCR-minDCR) corrects the initial internal resistance deviation, eliminates the interference caused by temperature unevenness, and obtains the final internal resistance deviation that can truly reflect the cell's own degradation.

[0062] In this embodiment, the initial internal resistance deviation is corrected using the relationship between charge, temperature, and discharge DC internal resistance to obtain the corresponding internal resistance deviation of the cell, including: Step C401: Obtain the temperature of each cell at each time point within the effective charge range.

[0063] From the discharge condition data after cleaning, all sampling times covered by the effective charge range are extracted, and the real-time temperature data of all cells in the battery pack at each time point are read to form a temperature dataset indexed by time point and containing the temperature of each cell. This dataset completely records the temperature distribution and changes of the cells within the effective range, providing complete and accurate raw temperature information for subsequent temperature difference calculation, temperature extreme value extraction, and DC internal resistance query.

[0064] Step C402: Calculate the temperature difference at each moment based on the temperature of each cell, and determine the moment with the largest temperature difference from the temperature differences at each moment; obtain the lowest temperature and the highest temperature corresponding to the moment with the largest temperature difference.

[0065] The temperature data of all cells at each time point are statistically analyzed. The highest temperature, lowest temperature, and the difference between them are calculated for each time point. The temperature difference is obtained by iterating through all time points and the time with the largest temperature difference is determined. Then, the lowest temperature mint and the highest temperature maxt corresponding to that time point are extracted. The time with the largest temperature difference is selected to obtain the extreme state where temperature has the most significant impact on internal resistance, ensuring that subsequent compensation can cover the maximum possible temperature interference and improve the reliability of the correction results.

[0066] Step C403: Using the lowest temperature and the corresponding cell charge at the lowest temperature, look up the first DC internal resistance from the relationship between charge, temperature and discharge DC internal resistance; and using the highest temperature and the corresponding cell charge at the highest temperature, look up the second DC internal resistance from the relationship between charge, temperature and discharge DC internal resistance.

[0067] Using the lowest and highest temperatures corresponding to the moment of maximum temperature difference, as well as the actual charge of the cell at that moment, as query conditions, a search and linear interpolation calculation are performed in the pre-calibrated charge-temperature-discharge DC internal resistance (DCR) relationship table to obtain the first DC internal resistance minDCR corresponding to the lowest temperature and the second DC internal resistance maxDCR corresponding to the highest temperature. These two internal resistance values ​​reflect the difference in internal resistance of the battery at extreme temperatures under the current charge level, providing key parameters for temperature compensation.

[0068] Step C404: The difference between the second DC internal resistance and the first DC internal resistance is used to correct the initial internal resistance deviation, thereby obtaining the corresponding internal resistance deviation of the cell.

[0069] First, calculate the difference between the second DC internal resistance and the first DC internal resistance, maxDCR-minDCR. This difference represents the maximum internal resistance deviation that may be caused by uneven temperature distribution. Then, subtract this difference from the initial internal resistance deviation IR to obtain the final internal resistance deviation IR_temp after temperature compensation. This eliminates false anomalies caused by temperature factors and retains the real anomalies caused by the degradation of the cell's own internal resistance.

[0070] This application embodiment extracts the maximum and minimum decentralized voltages of all cells within the effective charge range at all times, calculates the internal resistance deviation values ​​in the charging and discharging directions respectively, and averages them to obtain the initial internal resistance deviation degree IR. This comprehensively quantifies the actual voltage deviation of the cells under all charging and discharging conditions, avoiding the one-sidedness of evaluation in a single direction. Secondly, by collecting temperature data for all time periods and all cells, the moment with the largest temperature difference can be located, accurately capturing the extreme state where temperature has the most significant impact on internal resistance, ensuring that subsequent compensation covers the maximum possible temperature interference. Then, based on the charge-temperature-DC internal resistance calibration table, interpolation is performed to obtain the highest and lowest DC internal resistance under this extreme state, which can accurately quantify the internal resistance deviation caused by uneven temperature distribution, providing a reliable correction basis for compensation. Finally, the formula IR_temp=IR (maxDCR The minDCR corrects the initial internal resistance deviation, effectively eliminating false anomalies caused by temperature gradients, preventing high-temperature or low-temperature cells from being misjudged as having internal resistance anomalies, and ensuring that the final internal resistance deviation truly reflects the cell's own degradation state.

[0071] For example, such as Figure 2 As shown, within a certain effective charge range, the lowest temperature corresponding to the moment of maximum temperature difference is -30℃, and the highest temperature is 25℃. The current cell charge is 50% SOC. First, find the intersection of the -30℃ row and the 50% SOC column in the graph to obtain minDCR = 0.132mΩ; then find the intersection of the 25℃ row and the 50% SOC column to obtain maxDCR = 0.032mΩ. If the initial internal resistance deviation IR calculated earlier is 0.11mΩ, then substitute it into the formula for temperature compensation: IR_temp = 0.11 - (0.032 - 0.132) = 0.21mΩ. Since IR_temp > 0, it indicates that after deducting the effect of uneven temperature distribution, the cell still has a significant internal resistance deviation, and can be identified as a cell with abnormal internal resistance.

[0072] This application's embodiments divide the open-circuit voltage into corresponding charge intervals based on curves, breaking down the continuous discharge process into segmented stable analysis units. This makes the correspondence between cell voltage and internal resistance more reliable, reducing calculation errors caused by operating condition fluctuations from the source. Secondly, the discharge operating condition data is matched to the corresponding charge intervals, and then the effective charge intervals are filtered out by analyzing feature data, eliminating invalid data interference and improving the quality of the basic data for anomaly judgment. Then, the degree of internal resistance deviation is calculated using decentralized voltage extreme values, which can eliminate the systematic voltage difference caused by the cell's own SOC offset, highlight the voltage change caused by the real internal resistance anomaly, and weaken the impact of instantaneous interference. Finally, anomaly judgment is performed based on the effective intervals and the stable features after decentralized processing, which can clearly distinguish between real internal resistance anomalies and instantaneous interference, significantly reducing internal resistance calculation errors, improving detection accuracy, and effectively reducing the false positive rate and false negative rate.

[0073] It should be noted that because the cell voltage changes non-linearly with the state of charge (SOC), directly calculating the voltage and internal resistance during the full discharge process would lead to an unstable relationship between voltage and internal resistance. However, dividing the charge range based on the charge-open-circuit voltage curve is equivalent to cutting the non-linear discharge process into multiple approximately linear segments. Within each segment, the relationship between voltage, SOC, and internal resistance is more stable and consistent, thereby reducing the interference caused by operating condition fluctuations from the source and making the internal resistance calculation more stable and with smaller errors.

[0074] Decentralized voltage is calculated by subtracting the theoretical OCV corresponding to the current SOC from the measured voltage. The overall voltage difference caused by the inconsistency in SOC between cells is a systematic offset and will be directly canceled out when calculating the difference. However, voltage fluctuations caused by internal resistance anomalies are local and relative differences, which will not be canceled out but will instead be highlighted. Transient interference is usually random and irregular small fluctuations, which will be weakened in extreme value statistics and interval averaging, thus highlighting the true internal resistance anomalies.

[0075] The effective power range is selected after fluctuation verification, variance judgment, and outlier identification. Only the range with continuous and obvious abnormal signals will be retained. Instantaneous interference does not meet the continuous fluctuation condition and will be directly excluded. In addition, decentralization has eliminated SOC deviation and only retains the relative differences between cells. The real internal resistance abnormality will stably show deviation within the range, while instantaneous interference only appears briefly and sporadically and cannot pass the effective range screening. Therefore, the two can be clearly distinguished.

[0076] Another embodiment of this application provides a method for detecting abnormal internal resistance of a battery cell. Figure 3 This is a flowchart of a method for detecting abnormal internal resistance of a battery cell according to another embodiment of the present invention, as shown below. Figure 3 As shown, the process includes the following steps: Step S201: Obtain the discharge condition data of each cell in the battery, and obtain the capacity range corresponding to each open circuit voltage range of the battery. The capacity range is obtained by dividing the capacity range based on the capacity-open circuit voltage curve.

[0077] In this embodiment, the battery is a whole composed of multiple cells connected in series and parallel. The cell is the smallest basic unit constituting the battery pack, and the voltage, capacity, temperature, and operating state of the battery pack are all determined by the state of each cell. The discharge condition data mainly includes information such as real-time individual cell voltage, total battery current, cell temperature, real-time state of charge (SOC), timestamp, and charge / discharge flag, which are used to reflect the electrical characteristics and operating state of the battery during the actual discharge process.

[0078] First, the battery discharge condition data is extracted daily based on the charge and discharge flag bits, and long-term low-current discharge data is removed based on the current change rate to complete the data cleaning. At the same time, the battery's corresponding capacity-open-circuit voltage curve is obtained, the open-circuit voltage is divided according to a fixed voltage interval, and the capacity interval corresponding to each open-circuit voltage interval is obtained by looking up a table.

[0079] It should be noted that the SOC-OCV curve, or SOC-Open Circuit Voltage curve, describes the one-to-one mapping between the cell's open circuit voltage (OCV) and the remaining capacity (SOC). At different capacity levels, the cell's equilibrium voltage is fixed and calibrable; therefore, this curve accurately reflects the voltage variation with capacity and is the core basis for capacity estimation, voltage decentralization, and range division.

[0080] Dividing the SOC-OCV curve according to fixed voltage intervals aims to segment the continuously changing discharge process into multiple intervals with clear characteristics and stable calculations, enabling refined analysis. By segmenting the voltage and then mapping it to the corresponding charge intervals, the algorithm can subdivide rapidly changing voltage regions and coarsely segment voltage plateau regions. This amplifies subtle voltage fluctuations caused by abnormal cell internal resistance while avoiding excessive invalid data that would waste computational resources, thus improving the sensitivity of anomaly detection and the efficiency of the algorithm.

[0081] Step S202: Associate the discharge condition data of each cell with the corresponding power range to obtain the characteristic data of each cell in the corresponding power range.

[0082] In this embodiment, the real-time discharge data of each cleaned cell is matched to the corresponding power range based on the current power of the cell, thus completing the association and matching of data and range; within the same power range, the cell voltage is decentralized, that is, the real-time voltage of the cell is subtracted from the open circuit voltage corresponding to the current power to obtain a decentralized voltage sequence, forming the feature data of each cell for anomaly analysis within the corresponding power range.

[0083] Step S203: Analyze the characteristic data of the cells in each power range to obtain the effective power range.

[0084] In this embodiment of the application, a progressive condition judgment is performed on the decentralized voltage feature data in each power range. First, it is checked whether the voltage meets the preset positive and negative fluctuation range. Then, it is verified whether the cell voltage fluctuation variance reaches the effective threshold. Finally, a density clustering algorithm is used to identify whether there are outlier cells. Only the range that meets all the triggering conditions in sequence is determined to be an effective power range with anomaly analysis value.

[0085] Step S204: Based on the decentralized voltage extreme value corresponding to each cell within the effective power range, calculate the degree of deviation of the internal resistance of the cell, and determine whether the internal resistance of the cell is abnormal based on the degree of deviation.

[0086] In this embodiment of the application, the degree of deviation of the internal resistance of each cell is calculated based on the decentralized voltage extreme value corresponding to each cell within the effective charge range, including steps D1-D4: Step D1: Obtain the maximum and minimum decentralized voltage values ​​from the decentralized voltage values ​​of each cell within the effective power range.

[0087] A global statistical analysis is performed on the decentralized voltage values ​​of all cells within the effective charge range at all sampling times. This involves iterating through the decentralized voltage data for each cell and each time point, and determining the maximum decentralized voltage value `max[C_cellv_i_t]` and the minimum decentralized voltage value `min[C_cellv_i_t]` within the entire range through numerical comparison. These two extreme values ​​comprehensively reflect the most extreme voltage deviation of the cell within this range, accurately representing the difference between the cell and a normal cell, and providing the most representative characteristic data for subsequent calculation of internal resistance deviation.

[0088] Step D2: Calculate the first internal resistance deviation value in the charging direction based on the maximum decentralized voltage value and the maximum feedback current, and calculate the second internal resistance deviation value in the discharging direction based on the minimum decentralized voltage value and the maximum discharge current.

[0089] Based on the maximum and minimum decentralized voltage values ​​within the effective charge range, the internal resistance deviation values ​​in the charging and discharging directions are calculated separately. Dividing the maximum decentralized voltage value by the maximum feedback current set by the battery system yields the first internal resistance deviation value IR_C in the charging direction; dividing the minimum decentralized voltage value by the maximum discharge current set by the battery system yields the second internal resistance deviation value IR_D in the discharging direction. By calculating the deviation values ​​in both charging and discharging directions separately, the degree of voltage anomaly of the cell under different operating conditions can be comprehensively reflected.

[0090] Step D3: Calculate the initial internal resistance deviation based on the first internal resistance deviation value and the second internal resistance deviation value.

[0091] After obtaining the first internal resistance deviation value IR_C in the charging direction and the second internal resistance deviation value IR_D in the discharging direction, the two values ​​are arithmetically averaged. The initial internal resistance deviation degree IR is obtained using the formula IR=(IR_C+IR_D) / 2, with the unit being milliohms (mΩ). This initial internal resistance deviation degree integrates the voltage offset information during the charging and discharging process, and can preliminarily quantify the degree of deviation of the cell's internal resistance from that of a normal cell, serving as the basis for the result before temperature compensation.

[0092] Step D4: Using the relationship between charge, temperature, and discharge DC internal resistance, the initial internal resistance deviation is corrected to obtain the corresponding internal resistance deviation of the cell.

[0093] Specifically, by utilizing the relationship between charge, temperature, and discharge DC internal resistance, the initial internal resistance deviation is corrected to obtain the corresponding internal resistance deviation of the cell, including: Step D401: Obtain the temperature of each cell at each time within the effective charge range, and calculate the temperature difference at each time based on the temperature of each cell.

[0094] First, real-time temperature data of all cells in the battery pack at each sampling time is extracted from the discharge condition data corresponding to the effective charge range, forming a complete three-dimensional dataset of time-cell-temperature. Then, for each sampling time, the temperature data of all cells at that time are statistically analyzed to calculate the highest and lowest temperatures at that time, and the difference between the two is the temperature difference at that time.

[0095] Step D402: Using the temperature difference and charge of each cell at each time within the effective charge range, interpolate from the charge-temperature-discharge DC internal resistance relationship to obtain the real-time DC internal resistance at each time.

[0096] Based on the calculated temperature differences at various times, and combined with the real-time battery cell charge at each time point within the effective charge range, a three-dimensional query parameter of time-temperature difference-charge is constructed. A pre-calibrated table of charge-temperature-discharge DC internal resistance (DCR) is called. For each time point, based on the temperature difference and battery cell charge at that time, a linear interpolation algorithm is used to accurately calculate the real-time DC internal resistance of each battery cell at that time. This process iterates through all times and all cells within the effective charge range to ensure that each data point corresponds to an accurate real-time DC internal resistance, avoiding deviations caused by data from a single time point.

[0097] Step D403: Calculate the average DC internal resistance based on the real-time DC internal resistance, and calculate the dynamic compensation value based on the average DC internal resistance and the preset reference DC internal resistance.

[0098] First, the real-time DC internal resistance of all cells at all times is summarized and statistically analyzed to calculate the average real-time DC internal resistance of all cells within the effective charge range. This average value is then used as the actual equivalent average DC internal resistance of the cells within that range. Next, a preset reference DC internal resistance (based on the DC internal resistance calibration of a normal cell of the same type under standard operating conditions) is retrieved. The dynamic compensation value is calculated by the difference between the average DC internal resistance and the preset reference DC internal resistance. This compensation value can match the changes in internal resistance under different temperatures and charge states in real time, avoiding deviations caused by fixed compensation.

[0099] Step D404: Calculate the degree of internal resistance deviation of the cell based on the initial internal resistance deviation and the dynamic compensation value.

[0100] First, the initial internal resistance deviation (IR) calculated in the previous stage is retrieved. Combined with the obtained dynamic compensation value, a correction formula is used to couple the initial internal resistance deviation with the dynamic compensation value for calculation, ultimately obtaining the precisely corrected internal resistance deviation of the cell. This process fully considers the nonlinear fluctuations in internal resistance caused by changes in temperature and charge over all time periods, abandoning the limitations of fixed difference compensation, ensuring that the corrected internal resistance deviation can truly reflect the cell's own degradation state, and improving the accuracy of internal resistance anomaly judgment.

[0101] It should be noted that, due to the significant nonlinear characteristics of cell internal resistance changes under localized high or low temperatures or sudden temperature rises / falls, compensation using a fixed difference (maxDCR-minDCR) can only cover the state at the moment of maximum temperature difference. This sometimes fails to adapt to the internal resistance fluctuations caused by dynamic temperature changes throughout the entire timeframe, easily leading to compensation distortion and misjudgments or omissions. This application's embodiment collects temperature data throughout the entire timeframe, calculates the real-time temperature difference, and combines this with energy interpolation to obtain the real-time DC internal resistance at each moment. Then, it calculates the dynamic compensation value using the average internal resistance and the reference internal resistance, achieving moment-by-moment adaptive dynamic compensation. This perfectly adapts to the nonlinear characteristics of internal resistance changes, effectively solves the distortion risk of fixed difference compensation, and significantly improves the accuracy of calculating the degree of deviation in cell internal resistance.

[0102] In this embodiment of the application, determining whether the internal resistance of the battery cell is abnormal based on the degree of internal resistance deviation includes: Step F1: Select cells with internal resistance deviation greater than zero as candidate abnormal cells, and determine the target effective charge range for triggering abnormality in the candidate abnormal cells.

[0103] First, the internal resistance deviation (IR_temp) of all battery cells after temperature compensation is extracted. Cells with IR_temp values ​​greater than 0 are selected as candidate abnormal cells. Then, the source of the abnormality determination for these candidate abnormal cells is traced back to pinpoint the effective charge range that triggered the internal resistance deviation calculation. This range is marked as the target effective charge range. For example, if a battery cell calculates IR_temp = 0.8mΩ (>0) in the SOC 30%-40% range, then this cell is listed as a candidate abnormal cell, and SOC 30%-40% is its target effective charge range, thus clarifying the core analysis scope for subsequent verification of the abnormality's validity.

[0104] Step F2: Query the number of abnormal triggers of the candidate abnormal cell within the target effective power range, and query whether the candidate abnormal cell triggers an abnormality within the adjacent power range of the target effective power range.

[0105] For each candidate abnormal cell, count the number of times its internal resistance deviation is greater than 0 within the target effective charge range, i.e., the number of abnormal triggers. Simultaneously, query the preceding (e.g., SOC 20%-30%) and following (e.g., SOC 40%-50%) adjacent ranges of the target effective charge range, and verify whether the candidate abnormal cell also triggered an IR_temp>0 abnormality determination in these two adjacent ranges. For example, if the target range is SOC 30%-40%, check whether the cell is also determined as a candidate abnormality in the 20%-30% and 40%-50% ranges, thus obtaining a complete spatial distribution characteristic of abnormal triggers.

[0106] Step F3: If the number of abnormal triggers is the preset number and no abnormality is triggered in the adjacent power range, then it is determined that the internal resistance of the candidate abnormal cell is not abnormal; or, if the number of abnormal triggers is greater than the preset number, and / or an abnormality is triggered in the adjacent power range, then it is determined that the internal resistance of the candidate abnormal cell is abnormal.

[0107] First, set a preset threshold for the number of abnormal triggers (e.g., 3 times). Compare the actual number of triggers of candidate abnormal cells with the threshold: if the number of triggers equals the preset number and no abnormality is triggered in adjacent charge intervals, the cell is determined to be a false abnormality caused by momentary fluctuations, and there is no real problem with the internal resistance; if the number of triggers is greater than the preset number, or an abnormality is triggered in adjacent charge intervals (meeting either condition), the cell is determined to have a persistent abnormality, and the internal resistance has become truly abnormal. For example, if a cell is triggered 4 times (>3 times), even if there is no abnormality in adjacent intervals, its internal resistance is still determined to be abnormal; if the number of triggers is 2 times but there is an abnormality in adjacent intervals, it is also determined to be an internal resistance abnormality.

[0108] In this embodiment of the application, the method further includes: selecting candidate abnormal cells that have triggered abnormalities more than a preset number of times and / or that trigger abnormalities within adjacent power ranges as abnormal cells to be observed; continuously detecting the abnormality of the abnormal cells to be observed within multiple time periods; if the abnormality is that the abnormality is triggered continuously within multiple time periods, then the abnormal cells to be observed are determined as real abnormal cells; or, if the abnormality is that the abnormality is not triggered continuously within multiple time periods, then the abnormal cells to be observed are determined as non-real abnormal cells.

[0109] First, candidate abnormal cells with abnormal internal resistance are listed as abnormal cells to be observed. Then, the abnormal triggering of the cell is continuously monitored in multiple time periods such as 7 days and 15 days, and it is recorded whether the abnormality of IR_temp>0 is triggered in the effective charge range in each period. If the monitoring results show that the cell triggers the abnormality in ≥3 consecutive time periods, it indicates that the abnormality has a consistent trend and is identified as a real abnormal cell. If it only triggers in 1-2 periods and there is no abnormality in the other periods, or the abnormal triggering has no continuous pattern, it is determined to be a non-real abnormal cell, and the misjudgment caused by instantaneous interference is excluded.

[0110] This application's embodiments utilize a rapid initial screening based on the temperature-compensated internal resistance deviation IR_temp>0, which efficiently identifies potential abnormal cells and locates their target effective charge range, clarifying the analysis scope for subsequent verification and significantly improving detection efficiency. Secondly, by statistically analyzing the number of abnormal triggers of candidate abnormal cells within the target range and verifying their abnormalities in adjacent charge ranges, it effectively eliminates transient interference with only a single trigger and no spatial continuity, avoiding false alarms caused by occasional factors such as current surges. Then, cells verified through the dynamic fluctuation layer are listed as objects to be observed, and their abnormal triggering within multiple time periods is continuously monitored, requiring the abnormal signal to appear stably over multiple consecutive periods to further eliminate non-continuous interference. Finally, through a three-layer progressive detection logic—from rapid initial screening to transient interference filtering, and then to continuous verification—the risk of misjudgment is gradually reduced, ensuring that the ultimately identified true abnormal cells possess spatial and temporal continuity, significantly improving the accuracy and reliability of internal resistance anomaly detection and effectively avoiding false alarms caused by transient interference or occasional fluctuations.

[0111] This embodiment also provides a detection device for abnormal internal resistance of battery cells. This device is used to implement the above embodiments and preferred embodiments, and will not be repeated as already described. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0112] This embodiment provides a device for detecting abnormal internal resistance of a battery cell, such as... Figure 4 As shown, it includes: The acquisition module 401 is used to acquire discharge condition data of each cell in the battery and acquire the capacity range corresponding to each open circuit voltage range of the battery. The capacity range is obtained by dividing the capacity range based on the capacity-open circuit voltage curve.

[0113] The processing module 402 is used to associate the discharge condition data of each cell with the corresponding power range to obtain the characteristic data of each cell in the corresponding power range. Analysis module 403 is used to analyze the characteristic data of the cells in each power range to obtain the effective power range; The calculation module 404 is used to calculate the degree of deviation of the internal resistance of each cell based on the decentralized voltage extreme value corresponding to each cell within the effective power range, and to determine whether the internal resistance of the cell is abnormal based on the degree of deviation.

[0114] In this embodiment of the application, the acquisition module 401 is used to acquire the battery's charge-open-circuit voltage curve, wherein the charge-open-circuit voltage curve includes multiple basic voltage intervals; calculate the voltage change rate corresponding to each basic voltage interval in the charge-open-circuit voltage curve; adjust the interval width of the corresponding basic voltage interval according to the voltage change rate, and determine the adjusted basic voltage interval as the open-circuit voltage interval; and query the charge interval corresponding to each open-circuit voltage interval.

[0115] In this embodiment of the application, the analysis module 403 is used to calculate the decentralized voltage value of each cell using the measured voltage value and theoretical open circuit voltage of each cell within each power range; analyze the voltage fluctuation of each cell based on the decentralized voltage value; and determine the effective power range based on the fluctuation.

[0116] In this embodiment, the analysis module 403 is used to determine whether the decentralized voltage values ​​of all cells within the power range fall within a preset voltage range; if they all fall within the preset voltage range, the voltage variance of each cell within the power range at each time is obtained, and the maximum voltage variance is selected from multiple voltage variances; the maximum voltage variance is compared with the preset voltage variance; if the maximum voltage variance is greater than the preset voltage variance, clustering is performed using the voltage standard deviation corresponding to each cell within the power range to identify outlier cells; if outlier cells exist, the power range containing outlier cells is determined as a valid power range.

[0117] In this embodiment, the calculation module 404 is used to obtain the maximum and minimum decentralized voltage values ​​from the decentralized voltage values ​​of each cell within the effective power range; calculate the first internal resistance deviation value in the charging direction based on the maximum decentralized voltage value and the maximum feedback current, and calculate the second internal resistance deviation value in the discharging direction based on the minimum decentralized voltage value and the maximum discharge current; calculate the initial internal resistance deviation degree based on the first and second internal resistance deviation values; and correct the initial internal resistance deviation degree using the relationship between power, temperature, and discharge DC internal resistance to obtain the internal resistance deviation degree corresponding to the cell.

[0118] In this embodiment, the calculation module 404 is used to obtain the temperature of each cell at each time within the effective charge range; calculate the temperature difference at each time based on the temperature of each cell, and determine the time with the largest temperature difference from the temperature differences at each time; obtain the lowest temperature and the highest temperature corresponding to the time with the largest temperature difference; use the lowest temperature and the charge of the cell corresponding to the lowest temperature to query the first DC internal resistance from the relationship between charge-temperature-discharge DC internal resistance, and use the highest temperature and the charge of the cell corresponding to the highest temperature to query the second DC internal resistance from the relationship between charge-temperature-discharge DC internal resistance; use the difference between the second DC internal resistance and the first DC internal resistance to correct the initial internal resistance deviation, and obtain the internal resistance deviation degree corresponding to the cell.

[0119] In this embodiment, the calculation module 404 is used to obtain the temperature of each cell at each time within the effective charge range, and calculate the temperature difference at each time based on the temperature of each cell; using the temperature difference and charge of each cell at each time within the effective charge range, interpolating from the charge-temperature-discharge DC internal resistance relationship to obtain the real-time DC internal resistance at each time; calculating the average DC internal resistance based on the real-time DC internal resistance, and calculating the dynamic compensation value based on the average DC internal resistance and the preset reference DC internal resistance; and calculating the internal resistance deviation degree corresponding to the cell based on the initial internal resistance deviation degree and the dynamic compensation value.

[0120] In this embodiment, the calculation module 404 is used to identify cells with internal resistance deviation greater than zero as candidate abnormal cells, and determine the target effective charge range for triggering abnormalities in the candidate abnormal cells; query the number of abnormal triggers of the candidate abnormal cells within the target effective charge range, and query whether the candidate abnormal cells trigger abnormalities in adjacent charge ranges within the target effective charge range; if the number of abnormal triggers is a preset number, and no abnormality is triggered in adjacent charge ranges, then it is determined that the internal resistance of the candidate abnormal cells is not abnormal; or, if the number of abnormal triggers is greater than the preset number, and / or, an abnormality is triggered in adjacent charge ranges, then it is determined that the internal resistance of the candidate abnormal cells is abnormal.

[0121] In this embodiment of the application, the device further includes: a continuous detection module, used to identify candidate abnormal cells that have triggered abnormalities more than a preset number of times and / or that trigger abnormalities within adjacent power ranges as abnormal cells to be observed; continuously detect the abnormality of the abnormal cells to be observed within multiple time periods; if the abnormality is that the abnormality is triggered continuously within multiple time periods, then the abnormal cell to be observed is determined as a real abnormal cell; or, if the abnormality is that the abnormality is not triggered continuously within multiple time periods, then the abnormal cell to be observed is determined as a non-real abnormal cell.

[0122] Please see Figure 5 , Figure 5 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 5 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system).

[0123] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GPA), or any combination thereof.

[0124] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.

[0125] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device as shown by a landing page for an app. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, which can be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0126] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0127] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0128] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0129] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for detecting abnormal internal resistance of a battery cell, characterized in that, The method includes: The discharge condition data of each cell in the battery are obtained, and the capacity range of the battery in each open circuit voltage range is obtained, wherein the capacity range is obtained based on the capacity-open circuit voltage curve. The discharge condition data of each cell is correlated with the corresponding power range to obtain the characteristic data of each cell within the corresponding power range. Analyze the characteristic data of the battery cells within each of the stated power ranges to obtain the effective power ranges; Based on the decentralized voltage extreme value corresponding to each cell within the effective power range, the degree of deviation of the internal resistance of the cell is calculated, and the internal resistance of the cell is determined to be abnormal based on the degree of deviation.

2. The method according to claim 1, characterized in that, The step of obtaining the battery's capacity range corresponding to each open-circuit voltage range includes: Obtain the battery's charge-open-circuit voltage curve, wherein the charge-open-circuit voltage curve includes multiple basic voltage ranges; Calculate the rate of change of voltage for each base voltage range in the charge-open-circuit voltage curve; The width of the corresponding base voltage range is adjusted according to the voltage change rate, and the adjusted base voltage range is determined as the open circuit voltage range. Query the electrical quantity range corresponding to each of the aforementioned open-circuit voltage ranges.

3. The method according to claim 1, characterized in that, The analysis of the characteristic data of the battery cells within each of the aforementioned power ranges yields the effective power ranges, including: The decentralized voltage value of each cell is calculated using the measured voltage value and theoretical open-circuit voltage of the cells within each of the aforementioned power ranges. The voltage fluctuation of each cell is analyzed based on the decentralized voltage value, and the effective power range is determined based on the fluctuation.

4. The method according to claim 3, characterized in that, The step of analyzing the voltage fluctuation of each cell based on the decentralized voltage value and determining the effective capacity range based on the fluctuation includes: Determine whether the decentralized voltage value of all cells within the power range falls within the preset voltage range; If all fall within the preset voltage range, then obtain the voltage variance of each cell within the power range at each time, and select the maximum voltage variance from multiple voltage variances. Compare the maximum voltage variance with the preset voltage variance; If the maximum voltage variance is greater than the preset voltage variance, then clustering is performed using the voltage standard deviation of each cell within the power range to identify outlier cells. If outlier cells exist, the range of battery capacity containing the outlier cells will be defined as the effective range of battery capacity.

5. The method according to claim 1, characterized in that, The calculation of the internal resistance deviation of each cell based on the decentralized voltage extreme value corresponding to each cell within the effective charge range includes: The maximum and minimum decentralized voltage values ​​are obtained from the decentralized voltage values ​​of each cell within the effective power range. Based on the maximum decentralized voltage value and the maximum feedback current, calculate the first internal resistance deviation value in the charging direction, and based on the minimum decentralized voltage value and the maximum discharge current, calculate the second internal resistance deviation value in the discharge direction. Based on the first internal resistance deviation value and the second internal resistance deviation value, the initial internal resistance deviation degree is calculated; By utilizing the relationship between charge, temperature, and discharge DC internal resistance, the initial internal resistance deviation is corrected to obtain the corresponding internal resistance deviation of the battery cell.

6. The method according to claim 5, characterized in that, The method of correcting the initial internal resistance deviation by utilizing the relationship between charge, temperature, and discharge DC internal resistance to obtain the corresponding internal resistance deviation of the battery cell includes: Obtain the temperature of each cell at each time point within the effective charge range; The temperature difference at each moment is calculated based on the temperature of each cell, and the moment with the largest temperature difference is determined from the temperature differences at each moment. Obtain the lowest and highest temperatures corresponding to the moment when the temperature difference is greatest; Using the lowest temperature and the corresponding charge of the battery cell, the first DC internal resistance is queried from the relationship between charge, temperature and discharge DC internal resistance; and using the highest temperature and the corresponding charge of the battery cell, the second DC internal resistance is queried from the relationship between charge, temperature and discharge DC internal resistance. The initial internal resistance deviation is corrected by using the difference between the second DC internal resistance and the first DC internal resistance to obtain the internal resistance deviation of the cell.

7. The method according to claim 5, characterized in that, The method of correcting the initial internal resistance deviation by utilizing the relationship between charge, temperature, and discharge DC internal resistance to obtain the corresponding internal resistance deviation of the battery cell includes: The temperature of each cell within the effective power range at each time is obtained, and the temperature difference at each time is calculated based on the temperature of each cell. By utilizing the temperature difference and charge of each cell at various times within the effective charge range, the real-time DC internal resistance at each time is obtained by interpolation from the charge-temperature-discharge DC internal resistance relationship. The average DC internal resistance is calculated based on the real-time DC internal resistance, and the dynamic compensation value is calculated based on the average DC internal resistance and the preset reference DC internal resistance. The degree of internal resistance deviation of the cell is calculated based on the initial internal resistance deviation and the dynamic compensation value.

8. The method according to claim 1, characterized in that, The process of determining whether the internal resistance of the battery cell is abnormal based on the degree of deviation includes: Cells with internal resistance deviation greater than zero are selected as candidate abnormal cells, and the target effective charge range for triggering abnormalities in the candidate abnormal cells is determined. Query the number of times the candidate abnormal battery cell triggers an anomaly within the target effective power range, and query whether the candidate abnormal battery cell triggers an anomaly within the adjacent power range of the target effective power range; If the number of abnormal triggers is a preset number and no abnormality is triggered within the adjacent power range, then it is determined that the internal resistance of the candidate abnormal cell is not abnormal; or, if the number of abnormal triggers is greater than the preset number, and / or an abnormality is triggered within the adjacent power range, then it is determined that the internal resistance of the candidate abnormal cell is abnormal.

9. The method according to claim 8, characterized in that, The method further includes: Candidate abnormal cells that trigger anomalies more than a preset number of times and / or trigger anomalies within the adjacent power range are designated as abnormal cells to be observed. The abnormality of the observed abnormal cell is continuously monitored over multiple time periods. If the abnormal situation is that the abnormality is triggered continuously within multiple time periods, then the abnormal cell to be observed is determined to be a real abnormal cell; or, if the abnormal situation is that the abnormality is not triggered continuously within multiple time periods, then the abnormal cell to be observed is determined to be a non-real abnormal cell.

10. A device for detecting abnormal internal resistance of a battery cell, characterized in that, The device includes: The acquisition module is used to acquire discharge condition data of each cell in the battery and acquire the capacity range corresponding to each open circuit voltage range of the battery. The processing module is used to associate the discharge condition data of each cell with the corresponding power range to obtain the characteristic data of each cell within the corresponding power range. The analysis module is used to analyze the characteristic data of the battery cells within each of the stated power ranges to obtain the effective power ranges; The calculation module is used to calculate the degree of internal resistance deviation of each cell based on the decentralized voltage extreme value corresponding to each cell within the effective power range, and to determine whether the internal resistance of the cell is abnormal based on the degree of internal resistance deviation.

Images