Fault detection method and device for battery pack

Abstract

The invention relates to the technical field of battery detection, and discloses a fault detection method and device for a battery pack, and the method comprises the steps: obtaining the multi-dimensional state data of the battery pack in a full operation cycle, and enabling the multi-dimensional state data to comprise the health state data of the battery pack and the accumulated operation data; performing cross-dimensional association processing on the multi-dimensional state data to obtain coupling features related to a fault; judging whether the battery pack has a fault or not based on normal references of the coupling characteristics, the health state data of the battery pack and the accumulative operation data, and obtaining a fault judgment result; and based on the fault judgment result, predicting the development trend of the battery pack fault to obtain fault early warning information.

Description

Technical Field

[0001] This invention relates to the field of battery testing technology, and more specifically to a method and apparatus for detecting faults in a battery pack. Background Technology

[0002] Currently, most battery packs are composed of multiple cells connected in series and parallel. During long-term charge and discharge cycles, environmental temperature and humidity fluctuations, and natural aging, cell polarization, aging failure, and inconsistency deviations are prone to occur. If these faults are not detected in time, they may cause safety accidents such as overheating, leakage, or even fire. Therefore, there is an urgent need for the accuracy and timeliness of battery pack fault detection.

[0003] Existing battery pack fault detection technologies generally rely on the core logic of "isolated parameter acquisition + static threshold comparison." For example, they collect the voltage and internal resistance of a single cell or the overall temperature and current of the battery pack and compare them with preset fixed thresholds. If the parameter exceeds the threshold, a fault is determined. Although some technologies distinguish communication connection status and conduct separate detection processes, they still do not fundamentally break through the limitations of "single parameter judgment." The core problem with these technologies is that they fail to establish a multi-dimensional parameter coupling and correlation model and lack a dynamic judgment benchmark adapted to the entire life cycle of the battery pack. This leads to distorted fault feature extraction, resulting in low detection accuracy and significant false positives and false negatives. For example, rising ambient temperature accelerates the time-series decay of cell internal resistance and amplifies the deviation of group internal resistance. Single parameter detection cannot capture this coupling effect. The normal parameter ranges of new and aged batteries differ significantly. Static benchmarks can lead to false alarms for aged batteries or false negatives for early faults in new batteries, ultimately failing to meet the needs of battery pack safety monitoring in complex scenarios. Summary of the Invention

[0004] In view of the shortcomings of the existing technology, the present invention proposes a battery pack fault detection method and device to solve the above-mentioned technical problems.

[0005] Firstly, a method for detecting faults in a battery pack is provided, including: Acquire multi-dimensional status data of the battery pack throughout its entire operating cycle, wherein the multi-dimensional status data includes battery pack health status data and cumulative operating data; Perform cross-dimensional correlation processing on the multi-dimensional state data to obtain fault-related coupling features; Based on the normal baseline of the coupling characteristics, battery pack health status data and cumulative operating data, it is determined whether the battery pack has a fault, and a fault judgment result is obtained. Based on the fault judgment results, the development trend of battery pack faults is predicted, and fault warning information is obtained.

[0006] Furthermore, acquire multi-dimensional status data throughout the entire operating cycle of the battery pack, including: Voltage acquisition units and internal resistance acquisition units are deployed at the cell measurement points of the battery pack; temperature acquisition units and current acquisition units are deployed at the battery pack shell; insulation resistance acquisition units are deployed at the battery pack insulation layer; and health status reading units and operating data statistics units are deployed in the battery pack management system. Voltage decay data and internal resistance cyclic change data of a single cell are collected by the voltage acquisition unit and the internal resistance acquisition unit to form cell-level operating data; The temperature field distribution data, charging and discharging current fluctuation data, and insulation performance change data of the battery pack are collected by the temperature acquisition unit, current acquisition unit, and insulation resistance acquisition unit, respectively, to form pack-level operation data. The real-time health status data of the battery pack is retrieved through the health status reading unit, and the cumulative charge and discharge cycle data and cumulative running time data of the battery pack are obtained through the running data statistics unit to form benchmark adaptation data. The cell-level operating data, package-level operating data, and benchmark adaptation data are associated and spliced ​​according to the collection timestamp to generate multi-dimensional status data for the entire operating cycle.

[0007] Furthermore, cross-dimensional correlation processing is performed on the multi-dimensional state data to obtain fault-related coupling features, including: A time interpolation algorithm is used to map the high-frequency sampled values ​​of the cell-level raw data and the low-frequency sampled values ​​of the package-level raw data to a time axis with the same time interval, thus completing the timing alignment. From the time-aligned cell data, the slope of parameter change of a single cell in continuous charge-discharge cycles is calculated to obtain the time-series characteristics; From the time-aligned packet-level data, the average parameter values ​​of cells of the same model are statistically analyzed, and the deviation ratio between the parameters of a single cell and the average parameter value is calculated to obtain the group characteristics. Environmental impact factors are used to weight and fuse time-series characteristics and population characteristics to obtain coupled characteristics related to faults.

[0008] Furthermore, environmental impact factors are used to weight and fuse time-series and population characteristics to obtain fault-related coupling characteristics, including: The actual temperature of the battery pack is read by the temperature acquisition unit, and the actual humidity of the environment in which the battery pack is located is read by the humidity sensor to determine the temperature factor and humidity factor. Retrieve the current health status data of the battery pack. If the health status data is in the low health range, increase the weighting coefficient of the temperature factor on the time-series characteristics. If the health status data is in the high health range, then reduce the weighting coefficient of the temperature factor on the time series features; If the ambient humidity is in the high humidity range, then increase the weighting coefficient of the humidity factor on the population characteristics. If the ambient humidity is in the low humidity range, then reduce the weighting coefficient of the humidity factor on the population characteristics. Weighting coefficients are applied to the corresponding time-series features and population features, and the two types of weighted features are fused to obtain the coupled features related to the fault.

[0009] Furthermore, based on the normal baseline of the coupling characteristics, battery pack health status data, and cumulative operating data, it is determined whether the battery pack has a fault, and a fault determination result is obtained, including: Obtain historical coupling characteristic data of the same model of battery pack throughout its entire life cycle, and group the historical data according to the battery pack health status data range and the cumulative number of operation range; Calculate the statistical fluctuation range of each set of historical data, obtain the normal upper limit and normal lower limit of the corresponding coupling characteristics for each set, and construct a normal benchmark library; Read the battery pack health status data and cumulative number of runs, and match the corresponding upper and lower normal limits from the normal benchmark library to obtain a dynamic normal benchmark. The current coupling characteristic is compared with the upper and lower limits of the dynamic normal reference. If the current coupling characteristic exceeds the upper and lower limits, the battery pack is determined to be faulty.

[0010] Furthermore, the statistical fluctuation range of each set of historical data is calculated to obtain the normal upper limit and normal lower limit of the corresponding coupling characteristics for each set, including: For each set of historical coupling feature data, a normal distribution is fitted to obtain the mean and standard deviation of each set of data; Based on the mean and standard deviation, a normal fluctuation range is defined; The lower limit of normal is the standard deviation of the mean minus a preset multiple, and the upper limit of normal is the standard deviation of the mean plus a preset multiple. The set upper and lower limits are verified to obtain the normal upper limit and normal lower limit of the coupling characteristics for each group.

[0011] Furthermore, based on the fault judgment results, the development trend of battery pack faults is predicted to obtain fault warning information, including: Based on the historical coupling characteristic change curves of the battery pack, a fault trend prediction model is constructed, wherein the current coupling characteristic deviation value, battery pack health status data and cumulative number of runs are used as input variables. Input the current detected coupling feature deviation value, battery pack health status data and cumulative number of runs into the prediction model to simulate and output the coupling feature change curve in the future operation phase; Determine the estimated operational stage where coupling characteristics reach the fault critical value from the change curve; Based on the estimated distance between the current and the previous operating phase, fault warning information is obtained.

[0012] Secondly, a fault detection device for a battery pack is provided, characterized in that, based on any one of the preceding claims, a fault detection method for a battery pack includes: The acquisition module is configured to acquire multi-dimensional status data of the battery pack throughout its entire operating cycle, wherein the multi-dimensional status data includes battery pack health status data and cumulative operating data; The correlation processing module is configured to perform cross-dimensional correlation processing on the multi-dimensional state data to obtain coupling features related to the fault. The judgment module is configured to determine whether the battery pack has a fault based on the normal baseline of the coupling characteristics, battery pack health status data and cumulative operating data, and obtain a fault judgment result. The prediction module is configured to predict the development trend of battery pack failures based on the fault judgment results, and obtain fault warning information.

[0013] Thirdly, a terminal is provided, characterized in that it includes a processor, an input device, an output device, and a memory, the processor, the input device, the output device, and the memory being interconnected, wherein the memory is used to store a computer program, the computer program including program instructions, and the processor is configured to invoke the program instructions to execute a battery pack fault detection method as described in any of the preceding claims.

[0014] Fourthly, a computer-readable storage medium is provided, characterized in that the computer-readable storage medium stores a computer program, the computer program including program instructions, which, when executed by a processor, cause the processor to perform a battery pack fault detection method as described in any of the preceding claims.

[0015] The invention employing the above technical solution has the following advantages: This invention overcomes the limitations of existing technologies that rely on isolated single-parameter data acquisition by acquiring multi-dimensional state data, including health status data and cumulative operating data, throughout the entire operating cycle of a battery pack. This provides a comprehensive data foundation for the accurate extraction of subsequent fault features. Furthermore, cross-dimensional correlation processing extracts fault coupling features, effectively capturing the coupling correlation effects between cell-level parameters, pack-level parameters, and environmental parameters. This addresses the problems of traditional single-parameter detection failing to reflect the true operating state of the battery pack and feature extraction distortion, thus improving the identification accuracy of fault features. Simultaneously, based on coupling features and a dynamically updated normal benchmark combining health status and cumulative operating data, fault judgment is achieved. This ensures the normal benchmark adapts to the entire lifecycle state of the battery pack, minimizing the drawbacks of static benchmarks such as false warnings for aging batteries and missed early faults in new batteries, thereby improving the accuracy and reliability of fault judgment. Finally, the fault judgment results are combined to predict fault development trends and generate early warning information. Attached Figure Description

[0016] To more clearly illustrate the specific embodiments of the present invention, the accompanying drawings used in the specific embodiments will be briefly described below. In all the drawings, the elements or parts are not necessarily drawn to scale.

[0017] Figure 1 This is a flowchart of a battery pack fault detection method according to the present invention; Figure 2 This is a flowchart of a battery pack fault detection device according to the present invention; Figure 3 This is a structural diagram of the terminal in the battery pack fault detection method and device of the present invention. Detailed Implementation

[0018] The embodiments of the technical solution of the present invention will now be described in detail with reference to the accompanying drawings. These embodiments are merely illustrative of the technical solution of the present invention and are therefore intended to limit the scope of protection of the present invention.

[0019] like Figures 1-3 As shown, a battery pack fault detection method of the present invention includes: Acquire multi-dimensional status data of the battery pack throughout its entire operating cycle, including battery pack health status data and cumulative operating data; Perform cross-dimensional correlation processing on multi-dimensional state data to obtain coupled features related to faults; Based on the normal baseline of coupling characteristics, battery pack health status data and cumulative operating data, it is determined whether the battery pack has a fault, and the fault judgment result is obtained. Based on the fault diagnosis results, the development trend of battery pack faults is predicted, and fault warning information is obtained.

[0020] In this embodiment, multi-dimensional state data of the battery pack throughout its entire operating cycle is obtained, including: Voltage acquisition units and internal resistance acquisition units are deployed at the cell measurement points of the battery pack; temperature acquisition units and current acquisition units are deployed at the battery pack shell; insulation resistance acquisition units are deployed at the battery pack insulation layer; and health status reading units and operating data statistics units are deployed in the battery pack management system. Voltage decay data and internal resistance cyclic change data of a single cell are collected by the voltage acquisition unit and the internal resistance acquisition unit to form cell-level operating data; The temperature field distribution data, charging and discharging current fluctuation data, and insulation performance change data of the battery pack are collected by the temperature acquisition unit, current acquisition unit, and insulation resistance acquisition unit, respectively, to form pack-level operation data. The real-time health status data of the battery pack is retrieved through the health status reading unit, and the cumulative charge and discharge cycle data and cumulative running time data of the battery pack are obtained through the running data statistics unit to form benchmark adaptation data. Cell-level operating data, package-level operating data, and benchmark adaptation data are correlated and spliced ​​according to the collection timestamp to generate multi-dimensional status data for the entire operating cycle.

[0021] Specifically, the core principle of this step is to achieve full coverage acquisition of cell-level, package-level, and reference-adaptation-level data through a multi-point heterogeneous acquisition unit, and to eliminate data timing deviations through timestamp association. The specific implementation is as follows: Data acquisition unit deployment: A voltage acquisition unit (accuracy ±1mV) and an internal resistance acquisition unit (accuracy ±1mΩ) are deployed at the tab of each cell in the battery pack; three temperature acquisition units (NTC thermistors, accuracy ±0.5℃) are evenly arranged inside the battery pack casing; a Hall current acquisition unit (accuracy ±0.1A) is deployed in the main positive and negative circuits; an insulation resistance acquisition unit is deployed at the gap between the battery pack insulation layer and the casing; and a health status reading unit and an operation data statistics unit are embedded in the battery pack management system (BMS).

[0022] Classified data acquisition: Voltage decay data of a single cell is acquired through a voltage / internal resistance acquisition unit at a sampling frequency of 100ms. ; Temperature field distribution data were collected using temperature / current / insulation resistance acquisition units at sampling frequencies of 500ms, 100ms, and 10s, respectively. (k is the temperature measurement point number), charging and discharging current fluctuation data Data on changes in insulation performance Integrate into package-level runtime data Real-time health status data of the battery pack is retrieved through the BMS built-in unit. (Based on capacity decay calculation,) =Current capacity / Rated capacity), Cumulative charge / discharge cycle data Cumulative runtime data Integrate into benchmark adaptation data .

[0023] Multi-dimensional data association and splicing: A timestamp interpolation algorithm is used to unify the data timing. For low-frequency sampled data (such as insulation resistance), linear interpolation is used to supplement it to a 100ms sampling frequency. The interpolation formula is as follows: in, < < For adjacent sampling times, To provide interpolation padding values, The previous sampling time The original parameter values, For the next sampling time The original parameter values, For the previous sampling time, For the next sampling time, This is the target time point where data needs to be supplemented. Ultimately, cell-level, package-level, and benchmark adaptation data are correlated using a unified timestamp to generate multi-dimensional status data for the entire operating cycle. .

[0024] In this embodiment, cross-dimensional correlation processing is performed on multi-dimensional state data to obtain fault-related coupling features, including: A time interpolation algorithm is used to map the high-frequency sampled values ​​of the cell-level raw data and the low-frequency sampled values ​​of the package-level raw data to a time axis with the same time interval, thus completing the timing alignment. From the time-aligned cell data, the slope of parameter change of a single cell in continuous charge-discharge cycles is calculated to obtain the time-series characteristics; From the time-aligned packet-level data, the average parameter values ​​of cells of the same model are statistically analyzed, and the deviation ratio between the parameters of a single cell and the average parameter value is calculated to obtain the group characteristics. Environmental impact factors are used to weight and fuse time-series characteristics and population characteristics to obtain coupled characteristics related to faults.

[0025] In this embodiment, environmental impact factors are used to weight and fuse time-series features and population features to obtain fault-related coupled features, including: The actual temperature of the battery pack is read by the temperature acquisition unit, and the actual humidity of the environment in which the battery pack is located is read by the humidity sensor to determine the temperature factor and humidity factor. Retrieve the current health status data of the battery pack. If the health status data is in the low health range, increase the weighting coefficient of the temperature factor on the time-series characteristics. If the health status data is in the high health range, then reduce the weighting coefficient of the temperature factor on the time series features; If the ambient humidity is in the high humidity range, then increase the weighting coefficient of the humidity factor on the population characteristics. If the ambient humidity is in the low humidity range, then reduce the weighting coefficient of the humidity factor on the population characteristics. Weighting coefficients are applied to the corresponding time-series features and population features, and the two types of weighted features are fused to obtain the coupled features related to the fault.

[0026] Specifically, the core principle of this step is to eliminate sampling frequency differences through temporal alignment, then extract temporal-population dual features, and combine environmental and health status factors to achieve feature weighted coupling. The specific implementation is as follows: Multi-dimensional data time series alignment: A linear time interpolation algorithm is used to map cell-level high-frequency data and package-level / reference adaptation low-frequency data to a 100ms time axis. The aligned data meets the requirements. (n is a positive integer) to ensure the temporal consistency of different types of data.

[0027] Extraction of temporal and group features: Timing feature extraction: The internal resistance decay slope of a single cell over 5 consecutive charge-discharge cycles is calculated as a timing feature. The formula is: in, This is the start time of the first loop. This is the end time of the 5th loop; Let be the internal resistance of the i-th cell at the beginning of the first cycle; ; ; ; The larger the absolute value, the faster the cell aging rate.

[0028] Population feature extraction: Statistical analysis of the average internal resistance of 18 cells in the same batch of battery packs. The deviation ratio of the internal resistance of a single cell from the mean is calculated as a group characteristic. The formula is: in, Let be the average internal resistance of all cells in the same batch of battery packs at a certain time t, where i is the cell number. Let be the measured internal resistance of the i-th cell at time t, where t is the uniform time for internal resistance acquisition. The larger the value, the worse the consistency of the battery cells.

[0029] Weighted fusion of environmental impact factors generates coupling features: This step quantifies the impact of the environment on battery parameters using temperature / humidity factors, and dynamically adjusts the weights based on SOH (State of Health). The core algorithm is as follows: Environmental factors were determined by taking the average of three temperature measurement points. As a temperature factor Ambient humidity is obtained through an external humidity sensor. As a humidity factor ; in, For a certain moment Below is the arithmetic mean of the measured temperatures at three temperature sampling points inside the battery pack casing. For the first Each temperature acquisition unit at time The measured temperature value.

[0030] Dynamic adjustment of weighting coefficients: SOH < 80% is set as the low health range; SOH ≥ 80% is set as the high health range. The physical meaning is the weight of the influence of temperature on the time-series characteristics (aging rate) of the cell. The electrode material activity of the low health cell (SOH < 80%) is reduced and it is more sensitive to temperature; the electrode activity of the high health cell (SOH ≥ 80%) is strong and the influence of temperature is weak.

[0031] When H is greater than 60%, RH is defined as the high humidity range; when H is less than or equal to 60%, RH is defined as the low humidity range. Coupling feature calculation: The weighted temporal features are fused with the population features to obtain the fault coupling feature values. The formula is: in, These are the weighting coefficients of the temperature factor on the time series characteristics. This represents the weighting coefficient of the humidity factor on the population characteristics. For the first Timing characteristics of energy-saving cells, For the first Group characteristic values ​​of energy-saving cells.

[0032] Coupling characteristics It comprehensively reflects the aging rate and consistency deviation of the battery cells, and incorporates the influence of environmental and health conditions, thus characterizing the potential risks of battery failure.

[0033] 150 samples (100 normal samples and 50 faulty samples) of the same battery pack model were selected to verify the effectiveness of the coupling features: coupling features of normal samples All fall into Inside, the fault sample All are outside this range, and the fault type (aging / consistency / polarization) is the same as... The bias pattern (time-series feature-dominated / group feature-dominated / dual feature-dominated) is a perfect match. It comprehensively reflects the aging rate and consistency deviation of the battery cells, and incorporates the influence of environmental and health conditions, so as to accurately characterize the potential risks of battery failure.

[0034] In this embodiment, based on the normal baseline of coupling characteristics, battery pack health status data, and cumulative operating data, it is determined whether the battery pack has a fault, and the fault determination result is obtained, including: Obtain historical coupling characteristic data of the same model of battery pack throughout its entire life cycle, and group the historical data according to the battery pack health status data range and the cumulative number of operation range; Calculate the statistical fluctuation range of each set of historical data, obtain the normal upper limit and normal lower limit of the corresponding coupling characteristics for each set, and construct a normal benchmark library; Read the battery pack health status data and cumulative number of runs, and match the corresponding upper and lower normal limits from the normal benchmark library to obtain a dynamic normal benchmark. The current coupling characteristic is compared with the upper and lower limits of the dynamic normal reference. If the current coupling characteristic exceeds the upper and lower limits, the battery pack is determined to be faulty.

[0035] In this embodiment, the statistical fluctuation range of each set of historical data is calculated to obtain the normal upper limit and normal lower limit of the corresponding coupling feature for each set, including: For each set of historical coupling feature data, a normal distribution is fitted to obtain the mean and standard deviation of each set of data; Based on the mean and standard deviation, a normal fluctuation range is defined; The lower limit of normal is the standard deviation of the mean minus a preset multiple, and the upper limit of normal is the standard deviation of the mean plus a preset multiple. The set upper and lower limits are verified to obtain the normal upper limit and normal lower limit of the coupling characteristics for each group.

[0036] Specifically, the core principle of this step is to build a dynamic benchmark library using historical data from the entire battery pack lifecycle, enabling accurate fault determination under different health states and operating cycles. The specific implementation is as follows: Dynamic normal benchmark library construction Collect historical coupling characteristic data of the same model of battery pack (Grouped by SOH interval and cumulative number of cycles N); For each group of historical data, a normal distribution is fitted, and the within-group mean is calculated. with standard deviation The formula is: , in, The number of samples in a certain set of historical coupled feature data. The arithmetic mean of a set of historical coupling feature data. Let the standard deviation be a set of historical coupling feature data. Historical sample number, For the first Coupled feature values ​​of historical samples; Define the dynamic normal fluctuation range of the coupling characteristics: normal lower limit value Normal upper limit ; Finally, a dynamic normal benchmark library is constructed. .

[0037] Fault diagnosis: Retrieve the current battery pack's SOH and Match the corresponding normal range from the benchmark library, if the current coupling feature is... satisfy If the duration is ≥30 seconds, the corresponding battery cell is determined to be faulty, and a fault diagnosis result is obtained. .

[0038] In this embodiment, based on the fault diagnosis results, the development trend of battery pack faults is predicted to obtain fault warning information, including: Based on the historical coupling characteristic change curves of the battery pack, a fault trend prediction model is constructed, wherein the current coupling characteristic deviation value, battery pack health status data and cumulative number of runs are used as input variables. Input the current detected coupling feature deviation value, battery pack health status data and cumulative number of runs into the prediction model to simulate and output the coupling feature change curve in the future operation phase; Determine the estimated operational stage where coupling characteristics reach the fault critical value from the change curve; Based on the estimated distance between the current and the previous operating phase, fault warning information is obtained.

[0039] Specifically, the core principle of this step is to construct a predictive model based on the historical coupling characteristic change patterns to achieve accurate prediction of the fault critical time. The specific implementation is as follows: Fault trend prediction model construction: A fault trend prediction model is constructed using a linear regression model, with the current coupling feature deviation value as the input variable. Battery pack health status (SOH) and cumulative number of runs. The output is that the coupling characteristics reach the fault threshold. Estimated number of cycles The model expression is: Where a, b, c, and d are model training coefficients, a is the coupling feature bias influence coefficient, b is the health status compensation coefficient, c is the cumulative runtime decay coefficient, d is the base offset coefficient, and e is the cumulative duration influence coefficient. for and The absolute deviation.

[0040] Graded fault warning: Based on the estimated number of cycles Generate Level 3 early warning information: like ≥100, generate a mild warning (indicating "cell performance degradation"); If 50≤ <100, generate a moderate warning (indicating "early cell failure"); like <50, generate a severe warning (indicating "high risk of cell failure").

[0041] Warning threshold ≥100 (mild), 50≤ <100 (moderate) <50 (severe), physically meaning the estimated number of charge-discharge cycles for the coupling characteristics to reach the fault threshold, reflects the urgency of the fault development. Its setting is based on fault tracking experiments of 200 groups of battery packs of the same model: when When the temperature is ≥100, follow-up revealed a failure rate of only 5% within 30 days, indicating that the battery pack could still operate normally and only required enhanced monitoring; when the temperature is ≤50, the failure rate was ≥100%. When the temperature is below 100°C, the failure rate reaches 40% within 30 days, requiring cell testing within a specified period (7 days). If the battery temperature is below 50°C, the failure rate will reach 95% within 30 days, requiring immediate shutdown and replacement of faulty battery cells to avoid safety accidents.

[0042] This embodiment, through the above process, enables end-to-end detection of power tool battery packs, from data acquisition to fault warning.

[0043] In other embodiments, a battery pack fault detection device is provided, and a battery pack fault detection method based on any of the preceding claims includes: The acquisition module is configured to acquire multi-dimensional status data of the battery pack throughout its entire operating cycle. The multi-dimensional status data includes battery pack health status data and cumulative operating data. The correlation processing module is configured to perform cross-dimensional correlation processing on multi-dimensional state data to obtain coupling features related to faults. The judgment module is configured to determine whether the battery pack has a fault based on the normal baseline of coupling characteristics, battery pack health status data and cumulative operating data, and obtain the fault judgment result. The prediction module is configured to predict the development trend of battery pack failures based on the fault judgment results and obtain fault warning information.

[0044] In other embodiments, a terminal is provided, including a processor, an input device, an output device, and a memory, which are interconnected. The memory is used to store a computer program, which includes program instructions. The processor is configured to invoke the program instructions to execute a battery pack fault detection method as described above.

[0045] In other embodiments, a computer-readable storage medium is provided that stores a computer program, the computer program including program instructions that, when executed by a processor, cause the processor to perform a battery pack fault detection method as described in any of the preceding embodiments.

[0046] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0047] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0048] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical or other forms.

[0049] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0050] Furthermore, the functional units in the various embodiments of the application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software program module.

[0051] If the integrated unit is implemented as a software program module and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0052] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage device, which may include: a flash drive, a read-only memory, a random access memory, a magnetic disk, or an optical disk, etc.

[0053] The embodiments of this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for detecting faults in a battery pack, characterized in that, include: Acquire multi-dimensional status data of the battery pack throughout its entire operating cycle, wherein the multi-dimensional status data includes battery pack health status data and cumulative operating data; Perform cross-dimensional correlation processing on the multi-dimensional state data to obtain fault-related coupling features; Based on the normal baseline of the coupling characteristics, battery pack health status data and cumulative operating data, it is determined whether the battery pack has a fault, and a fault judgment result is obtained. Based on the fault judgment results, the development trend of battery pack faults is predicted, and fault warning information is obtained.

2. The battery pack fault detection method according to claim 1, characterized in that, Acquire multi-dimensional status data of the battery pack throughout its entire operating cycle, including: Voltage acquisition units and internal resistance acquisition units are deployed at the cell measurement points of the battery pack; temperature acquisition units and current acquisition units are deployed at the battery pack shell; insulation resistance acquisition units are deployed at the battery pack insulation layer; and health status reading units and operating data statistics units are deployed in the battery pack management system. Voltage decay data and internal resistance cyclic change data of a single cell are collected by the voltage acquisition unit and the internal resistance acquisition unit to form cell-level operating data; The temperature field distribution data, charging and discharging current fluctuation data, and insulation performance change data of the battery pack are collected by the temperature acquisition unit, current acquisition unit, and insulation resistance acquisition unit, respectively, to form pack-level operation data. The real-time health status data of the battery pack is retrieved through the health status reading unit, and the cumulative charge and discharge cycle data and cumulative running time data of the battery pack are obtained through the running data statistics unit to form benchmark adaptation data. The cell-level operating data, package-level operating data, and benchmark adaptation data are associated and spliced ​​according to the collection timestamp to generate multi-dimensional status data for the entire operating cycle.

3. The battery pack fault detection method according to claim 2, characterized in that, Perform cross-dimensional correlation processing on the multi-dimensional state data to obtain fault-related coupling features, including: A time interpolation algorithm is used to map the high-frequency sampled values ​​of the cell-level raw data and the low-frequency sampled values ​​of the package-level raw data to a time axis with the same time interval, thus completing the timing alignment. From the time-aligned cell data, the slope of parameter change of a single cell in continuous charge-discharge cycles is calculated to obtain the time-series characteristics; From the time-aligned packet-level data, the average parameter values ​​of cells of the same model are statistically analyzed, and the deviation ratio between the parameters of a single cell and the average parameter value is calculated to obtain the group characteristics. Environmental impact factors are used to weight and fuse time-series characteristics and population characteristics to obtain coupled characteristics related to faults.

4. The battery pack fault detection method according to claim 3, characterized in that, Environmental impact factors are used to weight and fuse time-series and population characteristics to obtain fault-related coupled characteristics, including: The actual temperature of the battery pack is read by the temperature acquisition unit, and the actual humidity of the environment in which the battery pack is located is read by the humidity sensor to determine the temperature factor and humidity factor. Retrieve the current health status data of the battery pack. If the health status data is in the low health range, increase the weighting coefficient of the temperature factor on the time-series characteristics. If the health status data is in the high health range, then reduce the weighting coefficient of the temperature factor on the time series features; If the ambient humidity is in the high humidity range, then increase the weighting coefficient of the humidity factor on the population characteristics. If the ambient humidity is in the low humidity range, then reduce the weighting coefficient of the humidity factor on the population characteristics. Weighting coefficients are applied to the corresponding time-series features and population features, and the two types of weighted features are fused to obtain the coupled features related to the fault.

5. The battery pack fault detection method according to claim 1, characterized in that, Based on the coupling characteristics, battery pack health status data, and normal baseline of cumulative operating data, it is determined whether the battery pack has a fault, and a fault determination result is obtained, including: Obtain historical coupling characteristic data of the same model of battery pack throughout its entire life cycle, and group the historical data according to the battery pack health status data range and the cumulative number of operation range; Calculate the statistical fluctuation range of each set of historical data, obtain the normal upper limit and normal lower limit of the corresponding coupling characteristics for each set, and construct a normal benchmark library; Read the battery pack health status data and cumulative number of runs, and match the corresponding upper and lower normal limits from the normal benchmark library to obtain a dynamic normal benchmark. The current coupling characteristic is compared with the upper and lower limits of the dynamic normal reference. If the current coupling characteristic exceeds the upper and lower limits, the battery pack is determined to be faulty.

6. The battery pack fault detection method according to claim 5, characterized in that, Calculate the statistical fluctuation range of each set of historical data to obtain the normal upper limit and normal lower limit of the corresponding coupling characteristics for each set, including: For each set of historical coupling feature data, a normal distribution is fitted to obtain the mean and standard deviation of each set of data; Based on the mean and standard deviation, a normal fluctuation range is defined; The lower limit of normal is the standard deviation of the mean minus a preset multiple, and the upper limit of normal is the standard deviation of the mean plus a preset multiple. The set upper and lower limits are verified to obtain the normal upper limit and normal lower limit of the coupling characteristics for each group.

7. The battery pack fault detection method according to claim 1, characterized in that, Based on the fault diagnosis results, the development trend of battery pack faults is predicted to obtain fault warning information, including: Based on the historical coupling characteristic change curves of the battery pack, a fault trend prediction model is constructed, wherein the current coupling characteristic deviation value, battery pack health status data and cumulative number of runs are used as input variables. Input the current detected coupling feature deviation value, battery pack health status data and cumulative number of runs into the prediction model to simulate and output the coupling feature change curve in the future operation phase; Determine the estimated operational stage where coupling characteristics reach the fault critical value from the change curve; Based on the estimated distance between the current and the previous operating phase, fault warning information is obtained.

8. A fault detection device for a battery pack, characterized in that, A fault detection method for a battery pack according to any one of claims 1 to 7 includes: The acquisition module is configured to acquire multi-dimensional status data of the battery pack throughout its entire operating cycle, wherein the multi-dimensional status data includes battery pack health status data and cumulative operating data; The correlation processing module is configured to perform cross-dimensional correlation processing on the multi-dimensional state data to obtain coupling features related to the fault. The judgment module is configured to determine whether the battery pack has a fault based on the normal baseline of the coupling characteristics, battery pack health status data and cumulative operating data, and obtain a fault judgment result. The prediction module is configured to predict the development trend of battery pack failures based on the fault judgment results, and obtain fault warning information.

9. A terminal, characterized in that, The device includes a processor, an input device, an output device, and a memory, which are interconnected. The memory stores a computer program, which includes program instructions. The processor is configured to invoke the program instructions to execute a battery pack fault detection method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, the computer program including program instructions that, when executed by a processor, cause the processor to perform a battery pack fault detection method as described in any one of claims 1 to 7.

Images