Battery anomaly detection method, device, system and computer readable storage medium

Abstract

The application discloses a battery abnormality detection method, device, system and computer readable storage medium, the method comprises: determining the calibration voltage data curve and the to-be-detected voltage data curve of the to-be-detected battery; determining the window length and the moving step based on the sampling time interval; determining the voltage cumulative fluctuation value of the to-be-detected battery according to the window length, the moving step, the calibration voltage data curve and the to-be-detected voltage data curve; and determining whether the to-be-detected battery is abnormal according to the voltage cumulative fluctuation value and the threshold value. According to the calibration voltage data curve and the to-be-detected voltage data curve, the voltage fluctuation curve of the to-be-detected battery is determined, and whether the to-be-detected battery is abnormal is determined according to the voltage cumulative fluctuation value and the threshold value, so that the abnormality detection of single batteries and the abnormality detection of battery packs can be met, and the generalization ability and the robustness of the battery abnormality detection method are improved.

Description

Technical Field

[0001] This invention relates to the field of energy storage technology, and in particular to battery anomaly detection methods, devices, systems and computer-readable storage media. Background Technology

[0002] Lithium-ion batteries, as a new type of energy source, have advantages such as high energy density, large capacity, long cycle life, and no memory effect, and are widely used in electronic products and electromechanical equipment. As a result, the safety issues of lithium-ion batteries are particularly important. For the detection of abnormalities in lithium batteries, the detection method based on the voltage consistency of series battery packs is the most commonly used. Existing technologies detect abnormal batteries by judging whether there is a serious deviation in the voltage characteristic of a certain battery in the entire battery pack. This requires the assumption of the balance and consistency of the series battery pack to complete the abnormality detection of individual cells in the battery pack, which has limitations in the detection of individual cells.

[0003] Therefore, improving the generalization ability and robustness of battery anomaly detection methods is an urgent problem to be solved. Summary of the Invention

[0004] The main objective of this invention is to provide a battery anomaly detection method, apparatus, system, and computer-readable storage medium, aiming to solve the problem of how to improve the generalization ability and robustness of battery anomaly detection methods.

[0005] To achieve the above objectives, the present invention provides a battery anomaly detection method, which includes the following steps:

[0006] Determine the calibration voltage data curve and the test voltage data curve of the battery to be tested;

[0007] The window length and movement step size are determined based on the sampling time interval;

[0008] The cumulative voltage fluctuation value of the battery under test is determined based on the window length, the movement step size, the calibration voltage data curve, and the voltage data curve to be tested.

[0009] The presence of any abnormality in the battery under test is determined based on the cumulative voltage fluctuation value and the threshold.

[0010] Optionally, the steps of determining the calibration voltage data curve and the voltage data curve to be tested for the battery include:

[0011] The normal voltage data of the battery under test is sampled to obtain the calibration voltage data curve;

[0012] Obtain the test voltage data of the battery to be tested, and determine the test voltage data curve based on the time scale information corresponding to the calibration voltage data curve and the test voltage data.

[0013] Optionally, the steps of determining the window length and moving step size based on the sampling time interval include:

[0014] Obtain the number of sample points corresponding to the voltage data curve to be detected, and determine the window length and moving step size based on the number of sample points and the sampling time interval.

[0015] Optionally, the step of determining the cumulative voltage fluctuation value of the battery under test based on the window length, the moving step size, the calibration voltage data curve, and the voltage data curve to be tested includes:

[0016] Align the calibration voltage data curve and the voltage data curve to be detected, and collect data from the aligned calibration voltage data curve and the voltage data curve to be detected based on the acquisition window corresponding to the window length and the movement step size to obtain a window data set;

[0017] Calculate the cumulative voltage difference value corresponding to each window data in the window data set;

[0018] The cumulative voltage fluctuation value of the battery under test is calculated based on the cumulative voltage difference value of each window data in the window data set and the cumulative voltage difference value of the previous window data corresponding to each window data.

[0019] Optionally, the step of acquiring data from the aligned calibration voltage data curve and the voltage data curve to be detected based on the acquisition window corresponding to the window length and the movement step size to obtain the window data set includes:

[0020] At each interval of the specified moving step length, a data acquisition window corresponding to the specified window length is set on the aligned calibration voltage data curve and the voltage data curve to be detected.

[0021] Data is collected from the calibration voltage data curve segment and the voltage data curve segment to be detected in each acquisition window to obtain the window data corresponding to each acquisition window.

[0022] When data collection is completed for all the aforementioned collection windows, a window data set is obtained.

[0023] Optionally, the step of calculating the cumulative voltage difference value corresponding to each window data in the window data set includes:

[0024] Obtain the calibration charge / discharge voltage data set and the voltage to be detected data set corresponding to each window data set in the window data set;

[0025] Obtain the acquisition time point corresponding to each calibrated charge-discharge voltage data in the calibrated charge-discharge voltage data set and each voltage data to be detected in the voltage data set to be detected, and calculate the difference between the calibrated charge-discharge voltage data and the voltage data to be detected corresponding to the acquisition time point to obtain the difference set;

[0026] The differences in the set of differences are summed to obtain the cumulative voltage difference value corresponding to each window of data.

[0027] Optionally, before the step of determining whether the battery under test is abnormal based on the cumulative voltage fluctuation value and the threshold, the following steps are included:

[0028] Obtain the battery management system information, battery type, and charge / discharge condition information corresponding to the battery to be tested, and determine the fluctuation error and movement error based on the battery management system information, the battery type, and the charge / discharge condition information;

[0029] The threshold is calculated based on the fluctuation error, the movement error, the window length, and the movement step size.

[0030] Optionally, the step of determining the fluctuation error and the movement error based on the battery management system information, the battery type, and the charge / discharge condition information includes:

[0031] The fluctuation constant and positive and negative electrode materials of the battery to be tested are determined according to the battery type, and the fluctuation error is determined according to the fluctuation constant, the positive and negative electrode materials and the battery management system information.

[0032] The moving constant and internal resistance of the battery under test are determined according to the battery type, and the charging and discharging current of the battery under test is determined according to the charging and discharging condition information.

[0033] The movement error is determined based on the movement constant, the internal resistance value, and the charging / discharging condition information.

[0034] Optionally, the step of determining whether the battery under test is abnormal based on the cumulative voltage fluctuation value and the threshold includes:

[0035] The cumulative voltage fluctuation value is compared with the threshold.

[0036] If the cumulative voltage fluctuation value is greater than the threshold, an alarm is triggered, and the number of consecutive alarms is recorded. The number of consecutive alarms is then compared with a preset alarm count threshold.

[0037] If the number of consecutive alarms exceeds a preset alarm count threshold, then the battery under test is determined to be abnormal.

[0038] Furthermore, to achieve the above objectives, the present invention also provides a battery anomaly detection device, the battery anomaly detection device comprising:

[0039] The first determining module is used to determine the calibration voltage data curve and the voltage data curve to be tested of the battery under test;

[0040] The second determining module is used to determine the window length and the moving step size based on the sampling time interval;

[0041] The third determining module is used to determine the cumulative voltage fluctuation value of the battery to be tested based on the window length, the moving step size, the calibration voltage data curve, and the voltage data curve to be tested.

[0042] The judgment module is used to determine whether the battery under test is abnormal based on the cumulative voltage fluctuation value and the threshold.

[0043] Furthermore, the first determining module is also used for:

[0044] The normal voltage data of the battery under test is sampled to obtain the calibration voltage data curve;

[0045] Obtain the test voltage data of the battery to be tested, and determine the test voltage data curve based on the time scale information corresponding to the calibration voltage data curve and the test voltage data.

[0046] Furthermore, the second determining module is also used for:

[0047] Obtain the number of sample points corresponding to the voltage data curve to be detected, and determine the window length and moving step size based on the number of sample points and the sampling time interval.

[0048] Furthermore, the third determining module is also used for:

[0049] Align the calibration voltage data curve and the voltage data curve to be detected, and collect data from the aligned calibration voltage data curve and the voltage data curve to be detected based on the acquisition window corresponding to the window length and the movement step size to obtain a window data set;

[0050] Calculate the cumulative voltage difference value corresponding to each window data in the window data set;

[0051] The cumulative voltage fluctuation value of the battery under test is calculated based on the cumulative voltage difference value of each window data in the window data set and the cumulative voltage difference value of the previous window data corresponding to each window data.

[0052] Furthermore, the third determining module is also used for:

[0053] At each interval of the specified moving step length, a data acquisition window corresponding to the specified window length is set on the aligned calibration voltage data curve and the voltage data curve to be detected.

[0054] Data is collected from the calibration voltage data curve segment and the voltage data curve segment to be detected in each acquisition window to obtain the window data corresponding to each acquisition window.

[0055] When data collection is completed for all the aforementioned collection windows, a window data set is obtained.

[0056] Furthermore, the third determining module is also used for:

[0057] Obtain the calibration charge / discharge voltage data set and the voltage to be detected data set corresponding to each window data set in the window data set;

[0058] Obtain the acquisition time point corresponding to each calibrated charge-discharge voltage data in the calibrated charge-discharge voltage data set and each voltage data to be detected in the voltage data set to be detected, and calculate the difference between the calibrated charge-discharge voltage data and the voltage data to be detected corresponding to the acquisition time point to obtain the difference set;

[0059] The differences in the set of differences are summed to obtain the cumulative voltage difference value corresponding to each window of data.

[0060] Furthermore, the third determining module is also used for:

[0061] Obtain the battery management system information, battery type, and charge / discharge condition information corresponding to the battery to be tested, and determine the fluctuation error and movement error based on the battery management system information, the battery type, and the charge / discharge condition information;

[0062] The threshold is calculated based on the fluctuation error, the movement error, the window length, and the movement step size.

[0063] Furthermore, the third determining module is also used for:

[0064] The fluctuation constant and positive and negative electrode materials of the battery to be tested are determined according to the battery type, and the fluctuation error is determined according to the fluctuation constant, the positive and negative electrode materials and the battery management system information.

[0065] The moving constant and internal resistance of the battery under test are determined according to the battery type, and the charging and discharging current of the battery under test is determined according to the charging and discharging condition information.

[0066] The movement error is determined based on the movement constant, the internal resistance value, and the charging / discharging condition information.

[0067] Furthermore, the determination module is also used for:

[0068] The cumulative voltage fluctuation value is compared with the threshold.

[0069] If the cumulative voltage fluctuation value is greater than the threshold, an alarm is triggered, and the number of consecutive alarms is recorded. The number of consecutive alarms is then compared with a preset alarm count threshold.

[0070] If the number of consecutive alarms exceeds a preset alarm count threshold, then the battery under test is determined to be abnormal.

[0071] In addition, to achieve the above objectives, the present invention also provides a battery anomaly detection system, the battery anomaly detection system comprising: a memory, a processor, and a battery anomaly detection program stored in the memory and executable on the processor, wherein the battery anomaly detection program, when executed by the processor, implements the steps of the battery anomaly detection method as described above.

[0072] In addition, to achieve the above objectives, the present invention also provides a computer-readable storage medium storing a battery anomaly detection program, which, when executed by a processor, implements the steps of the battery anomaly detection method as described above.

[0073] The battery anomaly detection method proposed in this invention determines the calibration voltage data curve and the test voltage data curve of the battery under test; determines the window length and movement step size; based on the sampling time interval, and according to the window length, movement step size, calibration voltage data curve, and test voltage data curve, determines the cumulative voltage fluctuation value of the battery under test; and determines whether the battery under test has an anomaly based on the cumulative voltage fluctuation value and a threshold. This invention determines the voltage fluctuation curve of the battery under test based on the calibration voltage data curve and the test voltage data curve, and determines whether an anomaly exists based on the cumulative voltage fluctuation value and a threshold. This method can satisfy both single-cell anomaly detection and battery pack anomaly detection, improving the generalization ability and robustness of the battery anomaly detection method. Attached Figure Description

[0074] Figure 1 This is a flowchart illustrating the first embodiment of the battery anomaly detection method of the present invention;

[0075] Figure 2This is a flowchart illustrating the second embodiment of the battery anomaly detection method of the present invention;

[0076] Figure 3 This is a schematic diagram of the data acquisition process of the acquisition window in this invention;

[0077] Figure 4 This is a flowchart illustrating the third embodiment of the battery anomaly detection method of the present invention;

[0078] Figure 5 This is a flowchart illustrating the fourth embodiment of the battery anomaly detection method of the present invention;

[0079] Figure 6 This is a schematic diagram of the battery anomaly detection device of the present invention.

[0080] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0081] Reference Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the battery anomaly detection method of the present invention. The method includes:

[0082] Step S10: Determine the calibration voltage data curve and the voltage data curve to be tested for the battery under test;

[0083] Step S20: Determine the window length and movement step size based on the sampling time interval;

[0084] Step S30: Determine the cumulative voltage fluctuation value of the battery to be tested based on the window length, the movement step size, the calibration voltage data curve, and the voltage data curve to be tested.

[0085] Step S40: Determine whether the battery under test is abnormal based on the cumulative voltage fluctuation value and the threshold.

[0086] This embodiment of the battery anomaly detection method is applied to a battery anomaly detection system, which can be applied to smart devices such as terminal devices and PC terminals. For ease of description, battery anomaly detection is used as an example. The battery anomaly detection system samples the normal voltage data of the battery under test to obtain a calibration voltage data curve. During the operation of the battery under test, it acquires the voltage data to be tested and determines the voltage data curve to be tested based on the time scale information corresponding to the calibration voltage data curve and the voltage data to be tested. The battery anomaly detection system obtains the number of sample points corresponding to the calibration voltage data curve or the voltage data curve to be tested, and then... The number of sample points and the preset sampling time interval determine the window length and movement step size. The battery anomaly detection system aligns the calibration voltage data curve and the voltage data curve to be tested, and collects data from the aligned calibration voltage data curve and the voltage data curve to be tested based on the acquisition window corresponding to the window length and the movement step size, obtaining a window data set. The battery anomaly detection system calculates the cumulative voltage difference value corresponding to each window data in the window data set. Based on the cumulative voltage difference value corresponding to each window data in the window data set and the cumulative voltage difference value corresponding to the previous window data, the battery anomaly detection system calculates the cumulative voltage fluctuation value of the battery to be tested. The battery anomaly detection system determines whether the battery to be tested has an anomaly based on the cumulative voltage fluctuation value and a threshold. It should be noted that the battery anomaly detection system can be used to detect lithium batteries, as well as other types of batteries.

[0087] This embodiment of the battery anomaly detection method determines the calibration voltage data curve and the test voltage data curve of the battery under test based on a preset sampling time interval; determines the window length and movement step size; determines the cumulative voltage fluctuation value of the battery under test based on the window length, movement step size, calibration voltage data curve, and test voltage data curve; and determines whether the battery under test has an anomaly based on the cumulative voltage fluctuation value and a threshold. This invention determines the voltage fluctuation curve of the battery under test based on the calibration voltage data curve and the test voltage data curve, and determines whether an anomaly exists based on the cumulative voltage fluctuation value and a threshold. This satisfies both individual cell anomaly detection and battery pack anomaly detection, improving the generalization ability and robustness of the battery anomaly detection method.

[0088] The following will provide a detailed explanation of each step:

[0089] Step S10: Determine the calibration voltage data curve and the voltage data curve to be tested for the battery under test;

[0090] In this embodiment, the battery anomaly detection system obtains the calibration voltage data curve of the battery under test based on the sampling time interval and the normal voltage data of the battery under test; it also obtains the test voltage data curve of the battery under test based on the sampling time interval and the test voltage data of the battery under test. It should be noted that the sampling time interval is set according to the actual operating conditions of the battery under test, and the sampling time interval can be 1 second, 0.5 seconds, 0.1 seconds, etc., that is, a charge / discharge voltage data is sampled every sampling time interval; the calibration voltage data curve can be a calibration discharge voltage data curve, a calibration charge voltage curve, or a calibration charge / discharge voltage data curve; the test voltage data curve can be a test discharge voltage data curve, a test charge voltage curve, or a test charge / discharge voltage data curve.

[0091] Specifically, the steps for determining the calibration voltage data curve and the voltage data curve to be tested for the battery include:

[0092] Step S101: Sample the normal voltage data of the battery to be tested to obtain the calibration voltage data curve;

[0093] In this step, the battery anomaly detection system acquires the normal voltage data of the battery under test. This normal voltage data can be historical data or data obtained through computer simulation, representing the charge and discharge voltage data of the battery under test under the premise of no anomalies. The battery anomaly detection system samples the normal voltage data of the battery under test based on sampling time intervals to obtain a calibration voltage data curve. It should be noted that the methods for obtaining the calibration voltage data curve of the battery under test include: a) averaging the measured normal voltage data of the battery under test; if it is a battery pack (composed of multiple batteries connected in series), the maximum and minimum voltages in the charge and discharge voltage data can be removed before averaging; b) obtaining the normal voltage data of the battery under test through an equivalent circuit model to obtain the calibration voltage data curve; c) obtaining the normal voltage data of the battery under test through a battery degradation model to obtain the calibration voltage data curve; d) obtaining the normal voltage data of the battery under test through other simulation methods to obtain the calibration voltage data curve.

[0094] Specifically, method a can be as follows: If the battery under test is confirmed to be normal, the battery under test is measured according to a preset sampling time interval to determine multiple voltage data curves of the battery under test. The average of these multiple voltage data curves is then calculated to obtain the charge / discharge voltage curve of the calibration battery corresponding to the battery under test. Alternatively, method a can be as follows: The battery type corresponding to the battery under test is obtained. A preset number of batteries of the same type are obtained; these batteries are calibration batteries that have passed testing without abnormalities, resulting in a calibration battery set. Each calibration battery in the calibration battery set is measured, and voltage data corresponding to each calibration battery is obtained according to a preset sampling time interval to obtain the calibration voltage data curve of each calibration battery. The average of all calibration voltage data curves is then calculated to obtain the calibration voltage data curve corresponding to the battery under test.

[0095] Step S102: Obtain the voltage data to be tested of the battery to be tested, and determine the voltage data curve to be tested based on the time scale information corresponding to the calibration voltage data curve and the voltage data to be tested.

[0096] In this step, the battery anomaly detection system acquires the test voltage data of the battery under test during its operation, and determines the test voltage data curve based on the time scale information corresponding to the calibration voltage data curve and the test voltage data. It can be understood that the time scale information includes the time length of the calibration voltage data curve, the sampling time corresponding to the sample point in the calibration voltage data curve, etc., that is, the time length of the test voltage data curve and the sampling time corresponding to the sample point are the same as those of the calibration voltage data curve.

[0097] Step S20: Determine the window length and movement step size based on the sampling time interval;

[0098] In this embodiment, the battery anomaly detection system determines the window length and the movement step size based on the sampling time interval. The window length refers to the length of the acquisition window, which is the maximum number of voltage data that the acquisition window can acquire. The movement step size refers to the number of voltage data acquired using one acquisition window every time the charging and discharging voltage data is moved.

[0099] Specifically, the steps for determining the window length and movement step size based on the sampling time interval include:

[0100] Step S201: Obtain the number of sample points corresponding to the voltage data curve to be detected, and determine the window length and moving step size based on the number of sample points and the sampling time interval.

[0101] In this step, the battery anomaly detection system acquires the number of sample points corresponding to the voltage data curve to be detected, and determines the window length and movement step size based on the number of sample points and the sampling time interval. It can be understood that in order to ensure the accuracy and efficiency of anomaly detection, the acquisition window needs to acquire a certain number of charge and discharge voltage data. Therefore, when the sampling time interval is small, the number of sample points corresponding to the calibrated voltage data curve or the voltage data curve to be detected is large and the sample points are relatively dense. At this time, the window length and movement step size can be set relatively small. When the sampling time interval is large, the number of sample points corresponding to the calibrated voltage data curve or the voltage data curve to be detected is small and the sample points are relatively sparse. At this time, the window length and movement step size can be set relatively large.

[0102] Step S30: Determine the cumulative voltage fluctuation value of the battery to be tested based on the window length, the movement step size, the calibration voltage data curve, and the voltage data curve to be tested.

[0103] In this embodiment, after determining the window length and the movement step size, the battery anomaly detection system collects data from the calibration voltage data curve and the voltage data curve to be tested through the acquisition window corresponding to the window length and the movement step size, and then determines the cumulative voltage fluctuation value of the battery to be tested based on the data collected by the acquisition window.

[0104] Specifically, step S30 includes:

[0105] Step S301: Align the calibration voltage data curve and the voltage data curve to be detected, and collect data from the aligned calibration voltage data curve and the voltage data curve to be detected based on the acquisition window corresponding to the window length and the movement step size to obtain a window data set.

[0106] In this step, since the time length of the voltage data curve to be tested and the sampling time corresponding to the sample point are the same as those of the calibration voltage data curve, the battery anomaly detection system can align the calibration voltage data curve and the voltage data curve to be tested according to the time length and the sampling time corresponding to the sample point. The battery anomaly detection system collects data from the aligned calibration voltage data curve and the voltage data curve to be tested based on the acquisition window corresponding to the window length and the movement step size, and obtains a window data set. It can be understood that each acquisition window collects a portion of the voltage data on the calibration voltage data curve and the voltage data curve to be tested, and the voltage data collected by all acquisition windows constitute the window data set.

[0107] Step S302: Calculate the cumulative voltage difference value corresponding to each window data in the window data set;

[0108] In this step, the battery anomaly detection system calculates the cumulative voltage difference value corresponding to each window data in the window data set. It can be understood that the window data set includes the voltage data collected in each acquisition window. The voltage data on the calibration voltage data curve and the voltage data curve to be detected in each acquisition window are calculated by difference and summation to obtain the cumulative voltage difference value corresponding to each acquisition window, that is, the cumulative voltage difference value corresponding to each window data.

[0109] Step S303: Calculate the cumulative voltage fluctuation value of the battery to be tested based on the cumulative voltage difference value of each window data in the window data set and the cumulative voltage difference value of the previous window data corresponding to each window data.

[0110] In this step, each window data in the window data set is arranged according to the sampling time corresponding to the sample points in the calibration voltage data curve and the voltage data curve to be tested; that is, the window data collected earlier is sorted first, and the window data collected later is sorted last. The battery anomaly detection system obtains the cumulative voltage difference value of each window data in the window data set and the cumulative voltage difference value of the previous window data corresponding to each window data, and calculates the cumulative voltage fluctuation value of the battery to be tested. For example, the window data set includes the first window data, the second window data, the third window data, and the fourth window data, arranged according to the sampling time as the first window data, the second window data, the third window data, and the fourth window data. The battery anomaly detection system first calculates the voltage fluctuation value of the battery under test based on the cumulative voltage difference values ​​of the first and second window data. Then, it calculates the voltage fluctuation value of the battery under test based on the cumulative voltage difference values ​​of the second and third window data. Finally, it calculates the voltage fluctuation value of the battery under test based on the cumulative voltage difference values ​​of the third and fourth window data. The battery anomaly detection system sorts all the voltage fluctuation values ​​according to the order of calculation to obtain the cumulative voltage fluctuation value, which is used for subsequent anomaly detection.

[0111] Step S40: Determine whether the battery under test is abnormal based on the cumulative voltage fluctuation value and the threshold.

[0112] In this embodiment, the battery anomaly detection system compares each voltage fluctuation value in the cumulative voltage fluctuation value with a threshold value in chronological order to obtain a comparison result, and determines whether the battery under test is abnormal based on the comparison result. It should be noted that the threshold value can be preset or determined based on the battery under test, the window length, and the movement step size.

[0113] Specifically, step S40 includes:

[0114] Step S401: Compare the accumulated voltage fluctuation value with the threshold.

[0115] Step S402: If the cumulative voltage fluctuation value is greater than the threshold, an alarm is triggered, and the number of consecutive alarms is recorded. The number of consecutive alarms is then compared with a preset alarm count threshold.

[0116] Step S403: If the number of consecutive alarms is greater than the preset alarm count threshold, then it is determined that the battery under test is abnormal.

[0117] In steps S401 to S402, the battery anomaly detection system compares the cumulative voltage fluctuation value with a threshold value in chronological order. The system first compares the first voltage fluctuation value in the cumulative voltage fluctuation value with the threshold value. If the first voltage fluctuation value is not greater than the threshold, it continues to compare the second voltage fluctuation value with the threshold value. If all voltage fluctuation values ​​in the cumulative voltage fluctuation value are not greater than the threshold value, the battery under test is determined to be without an anomaly. The battery anomaly detection system compares the cumulative voltage fluctuation value with the threshold value in chronological order. If a certain voltage fluctuation value in the cumulative voltage fluctuation value is not greater than the threshold value, the system continues to compare the first voltage fluctuation value with the threshold value in chronological order. If the voltage fluctuation value exceeds a threshold, an alarm is triggered, and the consecutive alarm count is incremented. The system then compares the next voltage fluctuation value in the accumulated voltage fluctuation value with the threshold. If the next voltage fluctuation value exceeds the threshold, an alarm is triggered again, and the consecutive alarm count is incremented again, until the total number of consecutive alarms exceeds a preset alarm count threshold. At this point, the battery under test is considered abnormal. If, before the consecutive alarm count exceeds the preset alarm count threshold, the next voltage fluctuation value being compared is not greater than the threshold, the consecutive alarm count is reset to zero, and the remaining voltage fluctuation values ​​in the accumulated voltage fluctuation value are compared with the threshold. Essentially, the battery anomaly detection system only determines that the battery under test is abnormal when the voltage fluctuation value in the accumulated voltage fluctuation value exceeds the threshold and the total number of consecutive alarms exceeds the preset alarm count threshold. This avoids misjudging the battery as abnormal due to voltage fluctuations caused by non-abnormal reasons, thus improving the accuracy of battery anomaly detection.

[0118] The battery anomaly detection system in this embodiment samples the normal voltage data of the battery under test based on a preset sampling time interval to obtain a calibration voltage data curve. During the operation of the battery under test, it acquires the voltage data to be tested and determines the voltage data curve to be tested based on the time scale information corresponding to the calibration voltage data curve and the voltage data to be tested. The system acquires the number of sample points corresponding to either the calibration voltage data curve or the voltage data curve to be tested, and determines the window length and movement step size based on the number of sample points and the preset sampling time interval. The system aligns the calibration voltage data curve and the voltage data curve to be tested, and acquires data from the aligned calibration voltage data curve and the voltage data curve to be tested based on the acquisition window corresponding to the window length and the movement step size, obtaining a window data set. The system calculates the cumulative voltage difference value corresponding to each window data point in the window data set. Based on the cumulative voltage difference value corresponding to each window data point in the window data set and the cumulative voltage difference value corresponding to the previous window data point, the system calculates the cumulative voltage fluctuation value of the battery under test. The system then determines whether the battery under test is abnormal based on the cumulative voltage fluctuation value and a threshold. The voltage fluctuation curve of the battery under test is determined by calibrating the voltage data curve and the voltage data curve to be tested. The presence of anomalies is judged based on the cumulative voltage fluctuation value and the threshold. This method can satisfy both the anomaly detection of individual cells and the anomaly detection of battery packs, thus improving the generalization ability and robustness of the battery anomaly detection method.

[0119] Further, refer to Figure 2 The second embodiment of the present invention is proposed. The difference between the second embodiment and the first embodiment is that the step of acquiring data from the aligned calibration voltage data curve and the voltage data curve to be detected based on the acquisition window corresponding to the window length and the movement step size to obtain the window data set includes:

[0120] Step S3011: At each interval of the moving step length, a collection window corresponding to the window length is set on the aligned calibration voltage data curve and the voltage data curve to be detected.

[0121] Step S3012: Data is collected from the calibration voltage data curve segment and the voltage data curve segment to be detected in each acquisition window to obtain the window data corresponding to each acquisition window;

[0122] Step S3013: When data acquisition is completed for all the acquisition windows, a window data set is obtained.

[0123] In this embodiment, as Figure 3As shown, the battery anomaly detection system moves a step (stride) at intervals and sets a collection window (window) of the corresponding window length on the aligned calibration voltage data curve and the voltage data curve to be tested. Based on the running time, data is collected from the calibration voltage data curve segment and the voltage data curve to be tested through the collection window to obtain the window data corresponding to each collection window. When all collection windows on the aligned calibration voltage data curve and the voltage data curve to be tested have completed data collection, the window data corresponding to each collection window is sorted according to the running time to obtain the window data set.

[0124] The battery anomaly detection system in this embodiment collects data from the calibration voltage data curve and the voltage data curve to be tested by acquiring a collection window and a movement step size, and obtains a window data set, which helps to calculate the cumulative voltage fluctuation value of the battery to be tested. It can meet the anomaly detection of both individual cells and battery packs, and improves the generalization ability and robustness of the battery anomaly detection method.

[0125] Further, refer to Figure 4 The present invention proposes a third embodiment, which differs from the first and second embodiments in that the step of calculating the cumulative voltage difference value corresponding to each window data in the window data set includes:

[0126] Step S3021: Obtain the calibration charge / discharge voltage data set and the voltage to be detected data set corresponding to each window data in the window data set;

[0127] Step S3022: Obtain the acquisition time point corresponding to each calibration charge-discharge voltage data in the calibration charge-discharge voltage data set and each voltage data to be detected in the voltage data set to be detected, and calculate the difference between the calibration charge-discharge voltage data and the voltage data to be detected corresponding to the acquisition time point to obtain the difference set.

[0128] Step S3023: Sum the differences in the difference set to obtain the cumulative voltage difference value corresponding to each window data.

[0129] In this embodiment, the battery anomaly detection system acquires the calibration charge / discharge voltage data set and the voltage to be tested data set corresponding to each window data in the window data set. It acquires the acquisition time point corresponding to each calibration charge / discharge voltage data in the calibration charge / discharge voltage data set and each voltage to be tested data in the voltage to be tested data set, and calculates the difference between the calibration charge / discharge voltage data and the voltage to be tested data corresponding to the acquisition time point to obtain a difference set. The differences in the difference set are summed to obtain the cumulative voltage difference value corresponding to each window data. For example, the calibration charge / discharge voltage data set corresponding to each window of data includes 10 calibration charge / discharge voltage data points, and the acquisition time points corresponding to each calibration charge / discharge voltage data point are t1 to t10. Similarly, the voltage data set to be tested includes 10 voltage data points to be tested, and the acquisition time points corresponding to each voltage data point to be tested are also t1 to t10. The battery anomaly detection system calculates the difference between the calibration charge / discharge voltage data and the voltage data to be tested at acquisition time point t1, calculates the difference between the calibration charge / discharge voltage data and the voltage data to be tested at acquisition time point t2, and so on, to obtain the difference set. Then, each difference in the difference set is accumulated to obtain the cumulative voltage difference value corresponding to each window of data.

[0130] The battery anomaly detection system in this embodiment calculates the cumulative voltage difference value corresponding to each window of data, which helps to calculate the cumulative voltage fluctuation value of the battery under test in the subsequent calculation. It can meet the anomaly detection of both individual cells and battery packs, thus improving the generalization ability and robustness of the battery anomaly detection method.

[0131] Further, refer to Figure 5 The present invention proposes a fourth embodiment, which differs from the first to third embodiments in that, before the step of determining whether the battery under test is abnormal based on the cumulative voltage fluctuation value and the threshold, the following steps are included:

[0132] Step a: Obtain the battery management system information, battery type, and charge / discharge condition information corresponding to the battery to be tested, and determine the fluctuation error and movement error based on the battery management system information, the battery type, and the charge / discharge condition information;

[0133] In this step, the battery anomaly detection system acquires the battery management system information, battery type, and charge / discharge condition information corresponding to the battery under test. Based on the battery management system information, battery type, and charge / discharge condition information, it determines the fluctuation error and the shift error. It should be noted that the battery management system information and battery type include lithium batteries, such as lithium cobalt oxide and lithium iron phosphate. The charge / discharge condition information includes the impedance at the beginning and end of the charge / discharge cycle and the current during the charge / discharge process. The fluctuation error is used to describe the voltage error caused by observation or subtle differences in the battery itself. The shift error is used to describe the voltage error caused by impedance or other differences during battery operation, at the beginning and end of the charge / discharge cycle, or when the voltage changes significantly.

[0134] Step b: Calculate the threshold based on the fluctuation error, the movement error, the window length, and the movement step size.

[0135] In this step, after determining the fluctuation error and the movement error, the battery anomaly detection system calculates a threshold based on the fluctuation error, the movement error, the window length, and the movement step size. Specifically, the formula for calculating the threshold is as follows:

[0136] Threshold=wihdows*K+stride*P

[0137] Where Threshold is the threshold value, windows is the window length, K is the fluctuation error, stride is the step size, and P is the movement error. It is understandable that different batteries under test will have different data regarding capacity, voltage, current, internal resistance, etc. Furthermore, different window lengths and step sizes will result in different acquisition errors when collecting charge and discharge voltage data of the battery under test. Therefore, by adaptively adjusting the threshold value based on the battery under test, window length, and step size, the threshold value for different types of batteries under test can be determined, which helps improve the accuracy of battery anomaly detection.

[0138] Further, step a, which involves determining the fluctuation error and the movement error based on the battery management system information, the battery type, and the charge / discharge condition information, includes:

[0139] Step a1: Determine the fluctuation constant and positive and negative electrode materials of the battery to be tested according to the battery type, and determine the fluctuation error according to the fluctuation constant, the positive and negative electrode materials and the battery management system information;

[0140] In this step, the battery anomaly detection system determines the fluctuation constant and positive and negative electrode materials of the battery under test based on the battery type, and determines the fluctuation error based on the fluctuation constant, positive and negative electrode materials, and battery management system information; specifically, the formula for calculating the fluctuation error is as follows:

[0141] K = a + u

[0142] Where K is the fluctuation error, α is the fluctuation constant, and u is the observation deviation. The observation deviation can be adjusted and determined based on the positive and negative electrode materials of the battery under test and the battery management system information required for the battery under test.

[0143] Step a2: Determine the constant value and internal resistance value of the battery under test according to the battery type, and determine the charging and discharging current of the battery under test according to the charging and discharging condition information.

[0144] Step a3: Determine the movement error based on the movement constant, the internal resistance value, and the charging / discharging condition information.

[0145] In steps a2 to a3, the battery anomaly detection system determines the constant value of the battery under test, the difference between the maximum and minimum internal resistance values ​​measured after the battery has been left to stand for a long time, based on the battery type, and determines the charging and discharging current of the battery under test based on the charging and discharging condition information. The system also determines the shift error based on the constant value, maximum internal resistance, minimum internal resistance, and charging and discharging condition information. Specifically, the formula for calculating the shift error is as follows:

[0146] P = b + ΔR*I

[0147] ΔR=R max -R min

[0148] Where P is the movement error, ΔR is the difference between the maximum and minimum internal resistance values ​​measured after the battery has been left to stand for a long time, I is the magnitude of the charging and discharging current, Rmax is the maximum internal resistance value, and Rmin is the minimum internal resistance value.

[0149] The battery anomaly detection system in this embodiment determines the threshold by using relevant information of the battery under test, window length, and movement step size. This allows the threshold of voltage characteristic quantities to be adjusted according to actual conditions, which may lead to poor robustness. However, this system improves the robustness of battery anomaly detection, enabling it to meet the requirements of anomaly detection for both individual cells and battery packs. This enhances the generalization ability and robustness of the battery anomaly detection method.

[0150] like Figure 6 As shown, the present invention also provides a battery anomaly detection device. The battery anomaly detection device of the present invention includes:

[0151] The first determining module 101 is used to determine the calibration voltage data curve and the voltage data curve to be tested of the battery to be tested;

[0152] The second determining module 102 is used to determine the window length and the moving step size based on the sampling time interval;

[0153] The third determining module 103 is used to determine the cumulative voltage fluctuation value of the battery to be tested based on the window length, the moving step size, the calibration voltage data curve, and the voltage data curve to be tested.

[0154] The judgment module 104 is used to determine whether the battery under test is abnormal based on the cumulative voltage fluctuation value and the threshold.

[0155] Furthermore, the first determining module is also used for:

[0156] The normal voltage data of the battery under test is sampled to obtain the calibration voltage data curve;

[0157] Obtain the test voltage data of the battery to be tested, and determine the test voltage data curve based on the time scale information corresponding to the calibration voltage data curve and the test voltage data.

[0158] Furthermore, the first determining module is also used for:

[0159] Obtain the number of sample points corresponding to the voltage data curve to be detected, and determine the window length and moving step size based on the number of sample points and the sampling time interval.

[0160] Furthermore, the third determining module is also used for:

[0161] Align the calibration voltage data curve and the voltage data curve to be detected, and collect data from the aligned calibration voltage data curve and the voltage data curve to be detected based on the acquisition window corresponding to the window length and the movement step size to obtain a window data set;

[0162] Calculate the cumulative voltage difference value corresponding to each window data in the window data set;

[0163] The cumulative voltage fluctuation value of the battery under test is calculated based on the cumulative voltage difference value of each window data in the window data set and the cumulative voltage difference value of the previous window data corresponding to each window data.

[0164] Furthermore, the third determining module is also used for:

[0165] At each interval of the specified moving step length, a data acquisition window corresponding to the specified window length is set on the aligned calibration voltage data curve and the voltage data curve to be detected.

[0166] Data is collected from the calibration voltage data curve segment and the voltage data curve segment to be detected in each acquisition window to obtain the window data corresponding to each acquisition window.

[0167] When data collection is completed for all the aforementioned collection windows, a window data set is obtained.

[0168] Furthermore, the third determining module is also used for:

[0169] Obtain the calibration charge / discharge voltage data set and the voltage to be detected data set corresponding to each window data set in the window data set;

[0170] Obtain the acquisition time point corresponding to each calibrated charge-discharge voltage data in the calibrated charge-discharge voltage data set and each voltage data to be detected in the voltage data set to be detected, and calculate the difference between the calibrated charge-discharge voltage data and the voltage data to be detected corresponding to the acquisition time point to obtain the difference set;

[0171] The differences in the set of differences are summed to obtain the cumulative voltage difference value corresponding to each window of data.

[0172] Furthermore, the third determining module is also used for:

[0173] Obtain the battery management system information, battery type, and charge / discharge condition information corresponding to the battery to be tested, and determine the fluctuation error and movement error based on the battery management system information, the battery type, and the charge / discharge condition information;

[0174] The threshold is calculated based on the fluctuation error, the movement error, the window length, and the movement step size.

[0175] Furthermore, the third determining module is also used for:

[0176] The fluctuation constant and positive and negative electrode materials of the battery to be tested are determined according to the battery type, and the fluctuation error is determined according to the fluctuation constant, the positive and negative electrode materials and the battery management system information.

[0177] The moving constant and internal resistance of the battery under test are determined according to the battery type, and the charging and discharging current of the battery under test is determined according to the charging and discharging condition information.

[0178] The movement error is determined based on the movement constant, the internal resistance value, and the charging / discharging condition information.

[0179] Furthermore, the determination module is also used for:

[0180] The cumulative voltage fluctuation value is compared with the threshold.

[0181] If the cumulative voltage fluctuation value is greater than the threshold, an alarm is triggered, and the number of consecutive alarms is recorded. The number of consecutive alarms is then compared with a preset alarm count threshold.

[0182] If the number of consecutive alarms exceeds a preset alarm count threshold, then the battery under test is determined to be abnormal.

[0183] The present invention also provides a battery anomaly detection system.

[0184] The battery anomaly detection system includes: a memory, a processor, and a battery anomaly detection program stored in the memory and executable on the processor. When the battery anomaly detection program is executed by the processor, it implements the steps of the battery anomaly detection method as described above.

[0185] The method implemented when the battery anomaly detection program running on the processor is executed can be referred to in various embodiments of the battery anomaly detection method of the present invention, and will not be repeated here.

[0186] The present invention also provides a computer-readable storage medium.

[0187] The computer-readable storage medium stores a battery anomaly detection program, which, when executed by a processor, implements the steps of the battery anomaly detection method as described above.

[0188] The method 0 implemented when the battery anomaly detection program running on the processor is executed can be referred to in various embodiments of the battery anomaly detection method of the present invention, and will not be repeated here.

[0189] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that includes a list of elements is included.

[0190] The system includes not only those elements, but also other elements not explicitly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified,

[0191] An element specified by the phrase "includes a..." does not exclude the existence of other identical elements in the process, method, article, or system that includes that element.

[0192] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0193] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above-described embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0194] The above are merely preferred embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for detecting battery anomalies, characterized in that, The battery anomaly detection method includes the following steps: Determine the calibration voltage data curve and the test voltage data curve of the battery to be tested; The window length and movement step size are determined based on the sampling time interval; The cumulative voltage fluctuation value of the battery to be tested is determined based on the window length, the moving step size, the calibration voltage data curve, and the voltage data curve to be tested. The presence of an abnormality in the battery under test is determined based on the cumulative voltage fluctuation value and the threshold. The step of determining the cumulative voltage fluctuation value of the battery under test based on the window length, the moving step size, the calibration voltage data curve, and the voltage data curve to be tested includes: Align the calibration voltage data curve and the voltage data curve to be detected, and collect data from the aligned calibration voltage data curve and the voltage data curve to be detected based on the acquisition window corresponding to the window length and the movement step size to obtain a window data set; Calculate the cumulative voltage difference value corresponding to each window data in the window data set; The cumulative voltage fluctuation value of the battery under test is calculated based on the cumulative voltage difference value of each window data in the window data set and the cumulative voltage difference value of the previous window data corresponding to each window data.

2. The battery anomaly detection method as described in claim 1, characterized in that, The steps for determining the calibration voltage data curve and the voltage data curve to be tested of the battery include: The normal voltage data of the battery under test is sampled to obtain the calibration voltage data curve; Obtain the test voltage data of the battery to be tested, and determine the test voltage data curve based on the time scale information corresponding to the calibration voltage data curve and the test voltage data.

3. The battery anomaly detection method as described in claim 1, characterized in that, The steps for determining the window length and movement step size based on the sampling time interval include: Obtain the number of sample points corresponding to the voltage data curve to be detected, and determine the window length and moving step size based on the number of sample points and the sampling time interval.

4. The battery anomaly detection method as described in claim 1, characterized in that, The step of acquiring data from the aligned calibration voltage data curve and the voltage data curve to be detected based on the acquisition window corresponding to the window length and the movement step size to obtain a window data set includes: At each interval of the specified moving step length, a data acquisition window corresponding to the specified window length is set on the aligned calibration voltage data curve and the voltage data curve to be detected. Data is collected from the calibration voltage data curve segment and the voltage data curve segment to be detected in each acquisition window to obtain the window data corresponding to each acquisition window. When data collection is completed for all the aforementioned collection windows, a window data set is obtained.

5. The battery anomaly detection method as described in claim 1, characterized in that, The step of calculating the cumulative voltage difference value corresponding to each window data in the window data set includes: Obtain the calibration charge / discharge voltage data set and the voltage to be detected data set corresponding to each window data set in the window data set; Obtain the acquisition time point corresponding to each calibrated charge-discharge voltage data in the calibrated charge-discharge voltage data set and each voltage data to be detected in the voltage data set to be detected, and calculate the difference between the calibrated charge-discharge voltage data and the voltage data to be detected corresponding to the acquisition time point to obtain the difference set; The differences in the set of differences are summed to obtain the cumulative voltage difference value corresponding to each window of data.

6. The battery anomaly detection method as described in claim 1, characterized in that, Before the step of determining whether the battery under test is abnormal based on the cumulative voltage fluctuation value and the threshold, the following steps are included: Obtain the battery management system information, battery type, and charge / discharge condition information corresponding to the battery to be tested, and determine the fluctuation error and movement error based on the battery management system information, the battery type, and the charge / discharge condition information; The threshold is calculated based on the fluctuation error, the movement error, the window length, and the movement step size.

7. The battery anomaly detection method as described in claim 6, characterized in that, The step of determining the fluctuation error and the movement error based on the battery management system information, the battery type, and the charging and discharging condition information includes: The fluctuation constant and positive and negative electrode materials of the battery to be tested are determined according to the battery type, and the fluctuation error is determined according to the fluctuation constant, the positive and negative electrode materials and the battery management system information. The moving constant and internal resistance of the battery under test are determined according to the battery type, and the charging and discharging current of the battery under test is determined according to the charging and discharging condition information. The movement error is determined based on the movement constant, the internal resistance value, and the charging / discharging condition information.

8. The battery anomaly detection method according to any one of claims 1-7, characterized in that, The step of determining whether the battery under test is abnormal based on the cumulative voltage fluctuation value and the threshold includes: The cumulative voltage fluctuation value is compared with the threshold. If the cumulative voltage fluctuation value is greater than the threshold, an alarm is triggered, and the number of consecutive alarms is recorded. The number of consecutive alarms is then compared with a preset alarm count threshold. If the number of consecutive alarms exceeds a preset alarm count threshold, then the battery under test is determined to be abnormal.

9. A battery anomaly detection device, characterized in that, The battery anomaly detection device includes: The first determining module is used to determine the calibration voltage data curve and the voltage data curve to be tested of the battery under test; The second determining module is used to determine the window length and the moving step size based on the sampling time interval; The third determining module is used to determine the cumulative voltage fluctuation value of the battery to be tested based on the window length, the moving step size, the calibration voltage data curve, and the voltage data curve to be tested. The judgment module is used to determine whether the battery under test is abnormal based on the cumulative voltage fluctuation value and the threshold. The third determining module is also used for: Align the calibration voltage data curve and the voltage data curve to be detected, and collect data from the aligned calibration voltage data curve and the voltage data curve to be detected based on the acquisition window corresponding to the window length and the movement step size to obtain a window data set; Calculate the cumulative voltage difference value corresponding to each window data in the window data set; The cumulative voltage fluctuation value of the battery under test is calculated based on the cumulative voltage difference value of each window data in the window data set and the cumulative voltage difference value of the previous window data corresponding to each window data.

10. A battery anomaly detection system, characterized in that, The battery anomaly detection system includes: a memory, a processor, and a battery anomaly detection program stored in the memory and executable on the processor. When the battery anomaly detection program is executed by the processor, it implements the steps of the battery anomaly detection method as described in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a battery anomaly detection program, which, when executed by a processor, implements the steps of the battery anomaly detection method as described in any one of claims 1 to 8.

Images