Battery pack and health assessment method, system, device, apparatus, medium, and program

Abstract

The application provides a battery pack and a health evaluation method, system, device, apparatus, medium and program; the method comprises the following steps: obtaining a first vibration signal of the battery pack; the first vibration signal is collected by an acceleration sensor arranged on the battery pack during the operation of the battery pack; based on the first vibration signal, modal analysis is performed on the battery pack to obtain a first natural frequency of the battery pack; based on the first natural frequency, the structural health of the battery pack is evaluated to obtain a first evaluation result, so as to improve the accuracy of evaluating the structural health state of the battery pack.

Description

Technical Field

[0001] This application relates to battery technology, and more particularly to a battery pack and a health assessment method, system, device, apparatus, medium and procedure. Background Technology

[0002] In related technologies, when sensing the dynamic characteristics of a battery pack's structure, the structural health of the battery pack is generally assessed using preset evaluation criteria. However, the setting of these evaluation criteria does not take into account changes in the structural state of the battery pack. Therefore, when evaluating the structure of the battery pack using these evaluation criteria, it will affect the accuracy of the assessment of the structural health of the battery pack. Summary of the Invention

[0003] This application provides a battery pack and a health assessment method, system, device, apparatus, medium, and procedure to improve the accuracy of assessing the structural health status of the battery pack.

[0004] On the one hand, the technical solution of this application embodiment is implemented as follows: This application provides a method for assessing the health of a battery pack. The method includes: acquiring a first vibration signal of the battery pack; the first vibration signal is collected by an accelerometer sensor installed on the battery pack during the operation of the battery pack; performing modal analysis on the battery pack based on the first vibration signal to obtain a first natural frequency of the battery pack; and assessing the structural health of the battery pack based on the first natural frequency to obtain a first assessment result. Specifically, assessing the structural health of the battery pack based on the first natural frequency to obtain the first assessment result includes: acquiring a reference natural frequency of the battery pack; the reference natural frequency is determined based on historical monitoring data of the battery pack, the historical monitoring data being obtained by monitoring the vibration signal of the battery pack at historical moments; determining a first frequency offset rate between the first natural frequency and the reference natural frequency; and assessing the structural health of the battery pack based on the first frequency offset rate and a first offset threshold to obtain the first assessment result.

[0005] In the above embodiments, firstly, an accelerometer installed on the battery pack can collect the first vibration signal of the battery pack during operation. Then, based on the collected first vibration signal, modal analysis is performed on the battery pack to obtain the first natural frequency of the battery pack, thus realizing the monitoring of the natural frequency of the battery pack during operation. However, if the battery pack structure is damaged, the natural frequency of the battery pack will gradually shift. Since the reference natural frequency represents the healthy state of the battery pack structure, the structure of the battery pack corresponding to the reference natural frequency is used as the reference structure for evaluating the structural health of the battery pack. In this way, the first offset rate between the monitored first natural frequency of the battery pack and the reference natural frequency can be used to indirectly evaluate the deviation between the current structure of the battery pack and the reference structure. Then, the structural health of the battery pack can be evaluated based on the deviation and the corresponding first offset threshold. This method enables monitoring of the battery pack's structural condition without disassembling it or when it has not been subjected to invasive damage. This mitigates the interference of traditional detection methods on the integrity and reliability of the battery pack, allowing for long-term, continuous monitoring of the battery pack's structural health without increasing maintenance costs. This, in turn, enables early identification of structural anomalies, improving the battery pack's safety and reliability. Furthermore, predicting the battery pack's reference natural frequency based on historical monitoring data allows for real-time updates of the reference natural frequency. The historical monitoring data, obtained by monitoring the battery pack's vibration signals at historical moments, varies depending on the battery pack's structural state. This achieves a dynamic match between the determined reference natural frequency and the battery pack's structural state, significantly improving the accuracy and robustness of structural health assessments.

[0006] In some embodiments, the structural health of the battery pack is assessed based on a first frequency offset rate and a first offset threshold to obtain a first assessment result, including one of the following: if the absolute value of the first frequency offset rate is greater than or equal to the first offset threshold, the first assessment result is determined as the first result; the first result indicates that the structural health of the battery pack is at a first risk level; if the absolute value of the first frequency offset rate is less than the first offset threshold, for each of the multiple time intervals, a set of frequency offset rates corresponding to the time interval is determined based on multiple second vibration signals collected by an accelerometer within the time interval; the multiple time intervals are later than the acquisition time of the first vibration signals; the set of frequency offset rates corresponding to the time intervals includes the second frequency offset rate between the second natural frequency and the reference natural frequency corresponding to the multiple second vibration signals collected within the time interval; the structural health of the battery pack is assessed based on the set of frequency offset rates corresponding to the multiple time intervals to obtain the first assessment result.

[0007] In the above embodiments, the structural health of the battery pack is classified into different risk levels based on the relationship between the absolute value of the first frequency offset rate and the first offset threshold. The first offset threshold is the minimum value used to determine whether the battery pack's structural health is at the first risk level. Therefore, by determining the relationship between the absolute value of the first frequency offset rate and the first offset threshold, the accuracy of determining whether the battery pack's structural health is at the first risk level can be improved, thereby enhancing the safety and reliability of the battery pack.

[0008] In some embodiments, the structural health of the battery pack is evaluated based on a set of frequency offset rates corresponding to multiple time intervals to obtain a first evaluation result, including: for each set of frequency offset rates, the average of multiple second frequency offset rates in the set of frequency offset rates is calculated to obtain an average offset rate; if the absolute value of the average offset rate corresponding to the multiple sets of frequency offset rates is greater than or equal to a second offset threshold, the first evaluation result is determined as a second result; the second result indicates that the structural health of the battery pack is at a second risk level.

[0009] In the above embodiments, the structural health of the battery pack is classified into different risk levels based on the relationship between the absolute value of the second frequency offset rate and the second offset threshold. The second offset threshold is the minimum value used to determine whether the battery pack's structural health is at the second risk level. Therefore, by determining the relationship between the absolute value of the second frequency offset rate and the second offset threshold, the accuracy of determining whether the battery pack's structural health is at the second risk level can be improved, thereby enhancing the safety and reliability of the battery pack.

[0010] In some embodiments, the method further includes: if the first evaluation result is the second result, determining a third natural frequency of the battery pack based on the acquired third vibration signal; the third vibration signal is acquired by an accelerometer after acquiring multiple second vibration signals; determining a third frequency offset rate between the third natural frequency and a reference natural frequency; if the absolute value of the third frequency offset rate is greater than or equal to a third offset threshold, determining that the structural health of the battery pack is at a third risk level; the third offset threshold is not less than a second offset threshold.

[0011] In the above embodiments, after the structural health of the battery pack triggers the second risk level, the structural health of the battery pack is classified into different risk levels based on the relationship between the absolute value of the third frequency offset rate and the third offset threshold. The third offset threshold is the minimum value used to determine whether the structural health of the battery pack is at the third risk level. Therefore, by judging the relationship between the absolute value of the third frequency offset rate and the third offset threshold, the accuracy of determining whether the structural health of the battery pack is at the third risk level can be improved, thereby improving the safety and reliability of the battery pack.

[0012] In some embodiments, obtaining the reference inherent frequency of the battery pack includes: obtaining the initial calibration frequency of the battery pack; determining the reference inherent frequency based on historical monitoring data of the battery pack and the initial calibration frequency; the historical monitoring data includes at least one of the following: historical frequency offset rate, historical reference inherent frequency, and historical evaluation results.

[0013] In the above embodiments, based on the historical monitoring data and initial calibration frequency of the battery pack, the reference natural frequency of the battery pack is predicted, and the reference natural frequency is updated. This allows the determined reference natural frequency to dynamically adapt to the drift of the battery's reference natural frequency caused by material aging and changes in the usage environment. As a result, the updated reference natural frequency can dynamically match the structural state of the battery pack, significantly improving the accuracy and robustness of structural health assessment of the battery pack.

[0014] In some embodiments, modal analysis of the battery pack based on the first vibration signal is performed to obtain the first natural frequency of the battery pack, including: performing frequency domain analysis on the first vibration signal to obtain power spectral density distribution information of the battery pack; and performing modal parameter identification on the power spectral density distribution information to obtain the first natural frequency.

[0015] In the above embodiments, firstly, frequency domain analysis is performed on the first vibration signal, which can decompose the complex time domain signal into readable power spectral density distribution information. Then, modal parameter identification is performed on the power spectral density distribution information to obtain the first natural frequency of the battery pack, providing controllable data support for realizing non-invasive and global health assessment of the battery pack structure.

[0016] In some embodiments, the method further includes: determining and executing a target strategy corresponding to a first evaluation result; the target strategy is used to output at least alarm information corresponding to the risk level of the structure of the battery pack.

[0017] In the above embodiments, by determining and executing the target strategy corresponding to the first evaluation result, and the target strategy is used to output alarm information corresponding to the risk level of the battery pack structure, the operating parameters of the battery pack can be adjusted according to the output alarm information, thereby slowing down the damage rate of the battery pack structure, which not only improves the battery pack life, but also improves the safety of the battery pack.

[0018] In some embodiments, determining and executing the target strategy corresponding to the first evaluation result includes: if the first evaluation result is a first result, outputting a first alarm message; the first result indicates that the structural health of the battery pack is at a first risk level; if the first evaluation result is a second result, outputting a second alarm message and controlling the operating parameters of the battery pack not to exceed the target parameter threshold; the operating parameters include at least one of the following: temperature, charge / discharge rate; the second result indicates that the structural health of the battery pack is at a second risk level; if the first evaluation result is a third result, cutting off the high voltage of the battery pack, controlling the opening of the pressure relief valve of the battery pack, and outputting a third alarm message; the third result indicates that the structural health of the battery pack is at a third risk level.

[0019] In the above embodiments, different target strategies are set for different first evaluation results. These different target strategies are based on factors such as the degree of structural damage to the battery pack, the rate of performance degradation, and the safety of use. This graded processing strategy can not only maximize the effective service life of the battery pack, but also reduce the cost waste caused by excessive maintenance of the battery pack.

[0020] In some embodiments, the method further includes: acquiring multiple fourth vibration signals collected by an accelerometer within a second preset time period; for each fourth vibration signal, determining the acceleration generated by the vibration of the battery pack after being subjected to stress based on the fourth vibration signal; determining the target number corresponding to each of the multiple acceleration intervals; the target number corresponding to the acceleration interval is the number of accelerations within the acceleration interval among the accelerations corresponding to each of the fourth vibration signals; and evaluating the structural fatigue of the battery pack based on the target number corresponding to each of the multiple acceleration intervals to obtain a second evaluation result.

[0021] In the above embodiments, firstly, an accelerometer installed on the battery pack can collect multiple fourth vibration signals within a second preset time period during the battery pack's operation. Then, based on the acceleration corresponding to each of the multiple fourth vibration signals and multiple acceleration intervals, a target quantity corresponding to each acceleration interval can be obtained. Since the acceleration interval reflects the range of different stresses borne by the battery pack structure, and the target quantity reflects the number of times the battery pack bears different stress ranges, there is a clear correlation between the target quantity and the structural fatigue of the battery pack. Thus, the structural fatigue of the battery pack can be assessed by using the target quantity corresponding to each of the multiple acceleration intervals, thereby achieving a quantitative assessment of the progressive structural deterioration of the battery pack.

[0022] In some embodiments, the structural fatigue of the battery pack is evaluated based on the target quantity corresponding to each of the multiple acceleration intervals to obtain a second evaluation result, including: for each acceleration interval, determining the structural damage value corresponding to the acceleration interval based on the target quantity corresponding to the acceleration interval; summing the structural damage values ​​corresponding to the multiple acceleration intervals to obtain a cumulative damage value; and evaluating the structural fatigue of the battery pack based on the cumulative damage value to obtain a second evaluation result.

[0023] In the above embodiments, the cumulative damage value of the battery pack is determined based on the structural damage value of each acceleration interval, thereby realizing the assessment of the structural fatigue of the battery pack. This simplifies the complex and variable vibration and impact loads into quantifiable cumulative damage values, thereby enabling early warning and maintenance of the battery pack's fatigue life.

[0024] In some embodiments, the structural fatigue of the battery pack is evaluated based on the cumulative damage value to obtain a second evaluation result, including: if the cumulative damage value is greater than or equal to the damage threshold, the second evaluation result is determined as a fourth result; the fourth result is used to characterize the structural fatigue of the battery pack at a third risk level.

[0025] In the above embodiments, by setting a loss threshold, the fatigue failure critical point of the battery pack can be clearly quantified, enabling deterministic diagnosis and timely intervention of structural damage to the battery pack, thus improving the safety risks caused by fuzzy judgments or excessive maintenance of the battery pack.

[0026] In some embodiments, the method further includes: obtaining at least one state parameter of the battery pack; the state parameter includes one or more of the following: temperature, charging parameters, and discharging parameters; and updating multiple acceleration ranges based on the at least one state parameter to obtain updated multiple acceleration ranges.

[0027] In the above embodiments, since changes in the state parameters of the battery will cause changes in the mechanical response and fatigue damage accumulation rate of the battery pack structure under different acceleration ranges, the accuracy of the assessment of the structural fatigue of the battery pack can be improved by updating multiple acceleration ranges according to at least one state parameter.

[0028] In some embodiments, the first evaluation result includes the target fault type of the battery pack structure; the structural health of the battery pack is evaluated based on the first natural frequency to obtain the first evaluation result, including: using a target model to identify structural faults of the battery pack based on the first natural frequency to obtain the target fault type; the target model is trained based on at least one historical natural frequency of the battery pack and the fault type labels corresponding to each historical natural frequency.

[0029] In the above embodiments, the structural fault type of the battery pack is output through the target model, which can quickly locate specific structural faults such as cracks, loosening or deformation, providing accurate basis for targeted maintenance and safety protection of the battery pack, significantly improving the efficiency of fault handling of the battery pack and reducing the risk of secondary damage to the battery pack.

[0030] On the other hand, embodiments of this application provide a battery management system for implementing the methods provided in embodiments of this application.

[0031] In another aspect, embodiments of this application provide a battery pack including at least one battery cell and the aforementioned battery management system.

[0032] In another aspect, embodiments of this application provide an electrical device including the aforementioned battery pack.

[0033] On the other hand, embodiments of this application provide a battery pack health assessment device, the device comprising: a first acquisition module, configured to acquire a first vibration signal of the battery pack; the first vibration signal is collected by an accelerometer sensor mounted on the battery pack during the operation of the battery pack; an analysis module, configured to perform modal analysis on the battery pack based on the first vibration signal to obtain a first natural frequency of the battery pack; and a first assessment module, configured to assess the structural health of the battery pack based on the first natural frequency to obtain a first assessment result; wherein the first assessment module comprises: a first acquisition unit, configured to acquire a reference natural frequency of the battery pack; the reference natural frequency is determined based on historical monitoring data of the battery pack, the historical monitoring data being obtained by monitoring the vibration signal of the battery pack at historical moments; a first determination unit, configured to determine a first frequency offset rate between the first natural frequency and the reference natural frequency; and a first assessment unit, configured to assess the structural health of the battery pack based on the first frequency offset rate and a first offset threshold to obtain the first assessment result.

[0034] In another aspect, embodiments of this application provide a computer device, which includes: a memory for storing computer-executable instructions or computer programs; and a processor for executing the computer-executable instructions or computer programs stored in the memory to implement the method provided in embodiments of this application.

[0035] In another aspect, embodiments of this application provide a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the method provided in embodiments of this application.

[0036] In another aspect, embodiments of this application provide a computer program product, including a computer program or instructions, which, when executed by a processor, implement the method provided in embodiments of this application. Attached Figure Description

[0037] Figure 1 This is a schematic diagram of the implementation process of a battery pack health assessment method provided in an embodiment of this application. Figure 1 ; Figure 2 This is a schematic diagram of an acceleration sensor mounted on a battery pack according to an embodiment of this application; Figure 3 This is a schematic diagram of a battery pack health assessment system provided in an embodiment of this application. Figure 1 ; Figure 4 This is a schematic diagram of the implementation process of a battery pack health assessment method provided in an embodiment of this application. Figure 2 ; Figure 5 This is a schematic diagram of a battery pack health assessment system provided in an embodiment of this application. Figure 2 ; Figure 6 This is a schematic diagram of the implementation process of a battery pack health assessment method provided in an embodiment of this application. Figure 3 ; Figure 7 This is a schematic diagram of the composition structure of a battery pack health assessment device provided in an embodiment of this application.

[0038] It should be noted that the terms "first" and "second" mentioned above are only used to distinguish between different options and do not represent the degree of superiority or inferiority of the options or their priority in the implementation process. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0040] In this embodiment, the battery pack can be made from battery cells and / or battery modules. A battery cell refers to a single battery unit, which is the basic unit capable of converting chemical energy into electrical energy. It can be used to manufacture battery modules or battery packs to supply power to electrical devices. A single battery cell can be a primary battery or a secondary battery. A secondary battery is a battery cell that can be recharged after discharge to reactivate its active materials and continue to be used. Battery cells can be lithium-ion batteries, sodium-ion batteries, sodium-lithium-ion batteries, lithium metal batteries, sodium metal batteries, lithium-sulfur batteries, magnesium-ion batteries, nickel-metal hydride batteries, nickel-cadmium batteries, or lead-acid batteries, etc., and this embodiment is not limited to these types. A single battery cell can be cylindrical, cuboid, or other shapes. A battery cell includes an electrode assembly, which comprises a positive electrode, a negative electrode, and a separator. During the charging and discharging process of the battery cell, active ions (such as lithium ions) repeatedly insert and extract between the positive and negative electrodes. The separator is positioned between the positive and negative electrodes to mitigate short circuits while allowing active ions to pass through.

[0041] In some embodiments, the separator is a separator membrane. This application does not impose any particular limitation on the type of separator membrane; any known porous separator membrane with good chemical and mechanical stability can be selected. As an example, the main material of the separator can be selected from at least one of glass fiber, non-woven fabric, polyethylene, polypropylene, polyvinylidene fluoride, and ceramic. The separator can be a single-layer film or a multi-layer composite film, without particular limitation. When the separator is a multi-layer composite film, the materials of each layer can be the same or different, without particular limitation. The separator can be a separate component located between the positive and negative electrodes, or it can be attached to the surfaces of the positive and negative electrodes. In some embodiments, the separator is a solid electrolyte. The solid electrolyte is disposed between the positive and negative electrodes, serving both to transport ions and to isolate the positive and negative electrodes.

[0042] In some embodiments, the battery cell also includes an electrolyte, which acts as a conductor of ions between the positive and negative electrodes. This application does not impose specific limitations on the type of electrolyte; it can be selected according to requirements. The electrolyte can be liquid, gel, or solid. Liquid electrolytes include electrolyte salts and solvents. In some embodiments, the electrolyte salt may be selected from at least one of lithium hexafluorophosphate, lithium tetrafluoroborate, lithium perchlorate, lithium hexafluoroarsenate, lithium bis(fluorosulfonyl)imide, lithium bis(trifluoromethanesulfonyl)imide, lithium trifluoromethanesulfonate, lithium difluorophosphate, lithium difluorooxalate borate, lithium dioxalate borate, lithium difluorodioxalate phosphate, and lithium tetrafluorooxalate phosphate.

[0043] In some embodiments, the solvent may be selected from at least one of ethylene carbonate, propylene carbonate, methyl ethyl carbonate, diethyl carbonate, dimethyl carbonate, dipropyl carbonate, methyl propyl carbonate, ethyl propyl carbonate, butyl carbonate, fluoroethylene carbonate, methyl formate, methyl acetate, ethyl acetate, propyl acetate, methyl propionate, ethyl propionate, propyl propionate, methyl butyrate, ethyl butyrate, 1,4-butyrolactone, sulfolane, dimethyl sulfone, methyl ethyl sulfone, and diethyl sulfone. The solvent may also be an ether solvent. Ether solvents may include one or more of ethylene glycol dimethyl ether, ethylene glycol diethyl ether, diethylene glycol dimethyl ether, triethylene glycol dimethyl ether, tetraethylene glycol dimethyl ether, 1,3-dioxolane, tetrahydrofuran, methyl tetrahydrofuran, diphenyl ether, and crown ethers. Among them, the gel electrolyte includes a polymer as the electrolyte backbone network, combined with an ionic liquid - lithium salt. Solid electrolytes include polymer solid electrolytes, inorganic solid electrolytes, and composite solid electrolytes. In some implementations, the electrode assembly is a wound structure. The positive and negative electrode sheets are wound into a wound structure. This application provides a method for assessing the health of a battery pack, such as... Figure 1 As shown, the method includes steps S100 to S120: Step S100: Acquire the first vibration signal of the battery pack; the first vibration signal is collected by an accelerometer installed on the battery pack during the operation of the battery pack; Here, an accelerometer can refer to a device that can measure the acceleration of an object.

[0044] In some embodiments, the accelerometer is disposed at a target location within the battery pack. This target location may include, but is not limited to, one or more of the following: bolt points on the support plate, connection points between various structures within the battery pack, points at a first distance from the structural adhesive within the battery pack, the amplitude center point of the battery pack, and the housing, etc. The connection points may include, but are not limited to, welding points, riveting points, bolt connections, and wiring harness connections between components.

[0045] like Figure 2 As shown, six acceleration sensors are installed on the battery pack. These six acceleration sensors are respectively located at the following locations: the bolt point 21 on the support plate, the weld point 22 on the battery pack, the side wall point 23 on the battery pack housing, the amplitude center point 26 on the battery pack (e.g., the cell pressure plate, cell bottom plate, or cell side plate), the structural adhesive point 25 on the battery pack cell structure, and the weak point 24 on the support plate of the battery pack.

[0046] In some implementations, the target location can be set by the user based on historical experience, or it can be determined through battery pack simulation. For example, after simulating the battery pack to obtain a stress contour map, accelerometers are placed in areas of the stress contour map that are greater than a first stress threshold; and / or, after simulating the battery pack to obtain a stiffness contour map, accelerometers are placed in areas of the stiffness contour map that are less than a first stiffness threshold. The stress contour map uses color visualization to visually represent the magnitude of internal stresses borne by each structure within the battery pack under specific operating conditions (such as vibration, impact, and thermal expansion) on the battery pack's geometric model. The stiffness contour map uses color distribution to show the ability of each structure within the battery pack to resist deformation under load.

[0047] In some implementations, the number of acceleration sensors disposed on the battery pack can be single or multiple. It is understood that when there are multiple acceleration sensors, the first vibration signal includes a first sub-vibration signal collected by each of the multiple acceleration sensors.

[0048] In one example, when there are three accelerometers, these three accelerometers can be respectively positioned at target locations in the front, rear, and middle regions of the battery module within the battery pack. The target location in the front region can refer to the edge connection point in the front area of ​​the module, the target location in the rear region can refer to the edge connection point in the rear area of ​​the module, and the target location in the middle region can refer to a point in the middle region of the module with poor heat dissipation. In this scenario, the first vibration signal includes the first sub-vibration signals collected by each of the three accelerometers; that is, the first vibration signal includes the first sub-vibration signal collected by the accelerometer in the front region, the first sub-vibration signal collected by the accelerometer in the rear region, and the first sub-vibration signal collected by the accelerometer in the middle region.

[0049] In some implementations, the accelerometer may refer to a micro-electro-mechanical system (MEMS) accelerometer, which is characterized by its small size and low power consumption. It is understood that by integrating a MEMS accelerometer into the battery pack, the battery pack does not need to provide additional power to the MEMS accelerometer, reducing the wiring complexity of the battery pack and thus reducing its hardware cost.

[0050] In some embodiments, the accelerometer can be a multi-axis accelerometer or a single-axis accelerometer. It is understood that when the accelerometer is a multi-axis accelerometer and there is only one accelerometer, the first vibration signal includes a first sub-signal corresponding to each direction. When the accelerometer is a multi-axis accelerometer and there are multiple accelerometers, the first vibration signal includes a first sub-vibration signal collected by each accelerometer, and each first sub-vibration signal includes a second sub-signal corresponding to each direction.

[0051] It is understandable that the various structures within the battery pack will vibrate during operation. Thus, the vibration signals of the battery pack can be collected by an accelerometer installed on the battery pack to obtain the first vibration signal.

[0052] Step S110: Based on the first vibration signal, perform modal analysis on the battery pack to obtain the first natural frequency of the battery pack; Here, modal analysis is a core analytical method in structural dynamics, used to determine the inherent vibration characteristics of a battery pack under dynamic loads. A natural frequency is a specific frequency exhibited by an object or system in free vibration caused solely by its own initial disturbances (such as initial displacement or velocity) without sustained external force.

[0053] In some implementations, the first vibration signal is preprocessed to obtain a fifth vibration signal; based on the fifth vibration signal, modal analysis is performed on the battery pack to obtain the first natural frequency of the battery pack.

[0054] In some implementations, when the first vibration signal includes a first sub-signal corresponding to each direction, modal analysis is performed on each first sub-signal in the first vibration signal to obtain a first sub-frequency corresponding to each first sub-signal. It can be understood that in this scenario, the first natural frequency includes the first sub-frequency corresponding to all the first sub-signals respectively.

[0055] In some implementations, when the first vibration signal includes multiple first sub-vibration signals, and each first sub-vibration signal includes a second sub-signal corresponding to each direction, modal analysis is performed on each second sub-signal in the first vibration signal to obtain the second sub-frequency corresponding to each second sub-signal. It can be understood that in this scenario, the first natural frequency includes the second sub-frequency corresponding to all the second sub-signals respectively.

[0056] Step S120: Based on the first inherent frequency, assess the structural health of the battery pack to obtain the first assessment result.

[0057] In some implementations, the first assessment result may include one or more of the following: risk level information of the structural health of the battery pack, health information of the battery pack, and fault type of the battery pack.

[0058] It is understandable that the natural frequency of a battery pack characterizes the intrinsic properties of its structural stiffness and mass distribution. Therefore, the initial natural frequency of a battery pack is usually set relatively high to avoid external excitation frequencies and thus improve the resonance of the battery pack. However, as the battery pack is used, when the structure of the battery pack is damaged, the natural frequency of the battery pack will gradually shift. Therefore, by detecting the natural frequency of the battery pack, the structural health of the battery pack can be monitored.

[0059] In some implementations, the first vibration signal is preprocessed to obtain a fifth vibration signal; based on the fifth vibration signal, modal analysis is performed on the battery pack to obtain the first natural frequency of the battery pack.

[0060] In some implementations, where the first inherent frequency includes the first sub-frequency corresponding to each of the first sub-signals, the structural health of the battery pack is evaluated based on at least one first sub-frequency to obtain a first evaluation result.

[0061] In some implementations, where the first inherent frequency includes the second sub-frequency corresponding to each of the second sub-signals, the structural health of the battery pack is evaluated based on at least one second sub-frequency to obtain a first evaluation result.

[0062] Understandably, after determining the first assessment result, a target strategy corresponding to the first assessment result can be determined. By implementing the target strategy, the rate of damage to the battery pack structure can be slowed down, effectively improving the battery pack's service life.

[0063] In this embodiment, firstly, an accelerometer installed on the battery pack can collect a first vibration signal of the battery pack during operation. Then, based on the collected first vibration signal, modal analysis is performed on the battery pack to obtain its first natural frequency, thus enabling monitoring of the battery pack's natural frequency during operation. Damage to the battery pack's structure will cause its natural frequency to gradually shift. Since the reference natural frequency characterizes the battery pack's structural health, the structure of the battery pack corresponding to the reference natural frequency is used as the reference structure for evaluating the battery pack's structural health. Thus, the first offset rate between the monitored first natural frequency and the reference natural frequency can indirectly assess the deviation between the battery pack's current structural state and the reference structure. Then, based on this deviation and the corresponding first offset threshold, the structural health of the battery pack can be evaluated. This method enables monitoring of the battery pack's structural condition without disassembling it or when it is not subjected to invasive damage. This reduces the interference of traditional detection methods on the integrity and reliability of the battery pack, allowing for long-term and continuous monitoring of the battery pack's structure without increasing maintenance costs. This, in turn, enables early identification of structural anomalies in the battery pack, improving its safety and reliability.

[0064] In some embodiments, step S120 may include steps S121 to S123: Step S121: Obtain the reference natural frequency of the battery pack; the reference natural frequency is determined based on the historical monitoring data of the battery pack, and the historical monitoring data is obtained by monitoring the vibration signal of the battery pack at historical moments; Here, the reference natural frequency is used as a benchmark for assessing the structural health of the battery.

[0065] In some implementations, the reference natural frequency can also be predicted based on historical monitoring data of the battery pack. This historical monitoring data can refer to frequency parameter data related to the frequency of the battery pack, such as historical reference natural frequency or historical natural frequency.

[0066] In some implementations, historical monitoring data is obtained by monitoring the vibration signal of the battery pack at a historical time. It is understood that the historical monitoring data can be the vibration signal of the battery pack at a historical time; the historical monitoring data can also be frequency parameter data related to the frequency of the battery pack obtained based on the vibration signal of the battery pack at a historical time, such as historical reference natural frequency or historical natural frequency; the historical monitoring data can also be structural state data related to the structural health status of the battery pack obtained based on the vibration signal of the battery pack at a historical time, such as historical assessment results.

[0067] In this application, historical monitoring data is obtained by monitoring the vibration signals of the battery pack at historical moments. The structural state of the battery pack is constantly changing, and this state can vary due to the battery pack's own characteristics (e.g., the gradual aging of materials within the battery pack over time) and / or the influence of the usage environment. The vibration signals of the battery pack differ under different structural states. Therefore, by monitoring the vibration signals of the battery pack at historical moments, the reference natural frequency of the battery pack can be matched with its actual structural state. Furthermore, by setting the reference natural frequency as a reference standard for assessing the structural health of the battery pack, dynamic updates to the reference standard are achieved, enabling dynamic matching between the reference standard and the structural state of the battery pack. This significantly improves the accuracy and robustness of the structural health assessment of the battery pack.

[0068] Step S122: Determine the first frequency offset rate between the first natural frequency and the reference natural frequency; Here, frequency offset rate refers to the degree of deviation of the natural frequency from the reference natural frequency.

[0069] In some implementations, a first difference between a first natural frequency and a reference natural frequency is obtained; the first difference is divided by the reference natural frequency to obtain a first frequency offset rate.

[0070] It is understandable that the reference natural frequency serves as an assessment benchmark for the structural health of a battery. The battery structure corresponding to the reference natural frequency is generally in a healthy state. The first frequency offset rate can obtain the degree of deviation between the natural frequency and the reference natural frequency. Therefore, the degree of damage to the battery structure can also be indirectly assessed through the first frequency offset rate.

[0071] In some implementations, where the first inherent frequency includes the first sub-frequency corresponding to each of the first sub-signals, a first sub-frequency offset rate between the reference inherent frequency and the first sub-frequency is determined for each first sub-frequency. It is understood that in this scenario, the first frequency offset rate includes the first sub-frequency offset rate corresponding to each of the first sub-signals.

[0072] In some implementations, where the first intrinsic frequency includes the second sub-frequency corresponding to each of the second sub-signals, a second sub-frequency offset rate between the reference intrinsic frequency and the second sub-frequency is determined for each second sub-frequency. It is understood that in this scenario, the first frequency offset rate includes the second sub-frequency offset rates corresponding to each of the second sub-signals.

[0073] Step S123: Based on the first frequency offset rate and the first offset threshold, evaluate the structural health of the battery pack to obtain the first evaluation result.

[0074] In some implementations, a first offset threshold can be obtained through a first mapping table, wherein the first mapping table stores different offset thresholds and the evaluation results corresponding to each offset threshold.

[0075] In one example, the first offset threshold could be 5%.

[0076] In some implementations, where the first frequency offset rate includes the first sub-frequency offset rates corresponding to all the first sub-signals respectively, the structural health of the battery pack is evaluated based on at least one first sub-frequency offset rate and a first offset threshold to obtain a first evaluation result.

[0077] In some implementations, where the first frequency offset rate includes the second sub-frequency offset rates corresponding to all the second sub-signals respectively, the structural health of the battery pack is evaluated based on at least one second sub-frequency offset rate and a first offset threshold to obtain a first evaluation result.

[0078] In this embodiment, firstly, a first frequency offset rate between a reference natural frequency and a first natural frequency is determined. Since the reference natural frequency characterizes the structural health of the battery pack, the battery structure corresponding to the reference natural frequency can be used as a benchmark structure for evaluating the structural health of the battery. Thus, the first offset rate can indirectly assess the deviation between the current structural state of the battery and the benchmark structure. Then, based on this deviation and the corresponding first offset threshold, the structural health of the battery pack can be evaluated. This method enables the monitoring of the structural state of the battery pack, achieving early identification of structural anomalies and improving the safety and reliability of the battery pack. Furthermore, predicting the reference natural frequency of the battery pack based on historical monitoring data allows for real-time updates of the reference natural frequency. The historical monitoring data is obtained by monitoring the vibration signal of the battery pack at historical moments. Since the vibration signal changes with the structural state of the battery pack, this achieves dynamic matching between the determined reference natural frequency and the structural state of the battery pack, significantly improving the accuracy and robustness of the structural health assessment of the battery pack.

[0079] In some embodiments, step S123 may include steps S1231 and S1232: Step S1231: If the absolute value of the first frequency offset rate is greater than or equal to the first offset threshold, determine the first evaluation result as the first result; the first result indicates that the structural health of the battery pack is at the first risk level. In some implementations, if the absolute value of the first offset rate is greater than a first offset threshold, the first evaluation result is determined as the first result.

[0080] In some implementations, the first evaluation result is determined as the first result if the absolute value of the first offset rate is equal to the first offset threshold.

[0081] In some implementations, when the first frequency offset rate includes the first sub-frequency offset rates corresponding to all the first sub-signals respectively, the first evaluation result can be determined as the first result when the absolute value of at least one first sub-frequency offset rate is greater than or equal to the first offset threshold.

[0082] In some implementations, when the first inherent frequency includes the second sub-frequency offset rates corresponding to all the second sub-signals respectively, the first evaluation result can be determined as the first result when the absolute value of at least one second sub-frequency offset rate is greater than or equal to the first offset threshold.

[0083] In this application, the first risk level may refer to a low risk level, which means that the damage to the battery pack structure is minimal at this risk level.

[0084] Step S1232: If the absolute value of the first frequency offset rate is less than the first offset threshold, for each of the multiple time intervals, based on the multiple second vibration signals collected by the accelerometer within the time interval, determine the frequency offset rate set corresponding to the time interval; the multiple time intervals are later than the acquisition time of the first vibration signal; the frequency offset rate set corresponding to the time interval includes the second frequency offset rate between the second natural frequency and the reference natural frequency corresponding to the multiple second vibration signals collected within the time interval; based on the frequency offset rate sets corresponding to the multiple time intervals, evaluate the structural health of the battery pack to obtain the first evaluation result.

[0085] In some implementations, when the first frequency offset rate includes the first sub-frequency offset rates corresponding to all the first sub-signals respectively, the absolute value of the first frequency offset rate being less than the first offset threshold may mean that all the first sub-frequency offset rates are less than the first offset threshold.

[0086] In some implementations, when the first frequency offset rate includes the second sub-frequency offset rates corresponding to all the second sub-signals respectively, the absolute value of the first frequency offset rate being less than the first offset threshold may mean that all the second sub-frequency offset rates are less than the first offset threshold.

[0087] In some implementations, multiple time intervals can refer to at least two time intervals, such as two time intervals, four time intervals, five time intervals, etc.

[0088] In this application, multiple time intervals are sequential in time. For example, if the length of a time interval is 2 minutes and there are 3 time intervals, then the 3 time intervals can refer to [0,2), [2,4), [6,8).

[0089] In some implementations, for each time interval, the accelerometer is controlled to collect data at a predicted acquisition frequency to obtain multiple second vibration signals. These multiple second vibration signals can refer to at least two second vibration signals, such as two second vibration signals, five second vibration signals, etc.

[0090] In this application, the frequency offset rate set includes multiple second frequency domain offset rates, and there is a one-to-one correspondence between the multiple second frequency domain offset rates and multiple second vibration signals.

[0091] In some implementations, for each second vibration signal, modal analysis is performed on the battery pack based on the second vibration signal to obtain the second natural frequency of the battery pack; and a second frequency offset rate between the second natural frequency and the reference natural frequency is determined.

[0092] In implementation, when the second vibration signal includes multiple third sub-signals, a third sub-frequency corresponding to each third sub-signal is determined; for each third sub-frequency, a second sub-frequency offset rate between each third sub-frequency and the reference natural frequency is determined. It is understood that in this scenario, the second frequency offset rate includes the second sub-frequency offset rates corresponding to all the third sub-signals. In this application, the inclusion of multiple third sub-signals in the second vibration signal may be due to the accelerometer being a multi-axis sensor, and / or multiple accelerometers being installed on the battery pack.

[0093] In this application, the determination of the second frequency offset rate can be found in the specific implementation description of the first frequency offset rate described above, and will not be repeated here.

[0094] In this application, there is a one-to-one correspondence between multiple time intervals and multiple frequency offset sets.

[0095] In some implementations, the structural health of the battery pack is evaluated based on a set of frequency offset rates corresponding to multiple time intervals and a second offset threshold, to obtain a first evaluation result.

[0096] In this embodiment, the structural health of the battery pack is classified into different risk levels based on the relationship between the absolute value of the first frequency offset rate and the first offset threshold. The first offset threshold is the minimum value used to determine whether the battery pack's structural health is at the first risk level. Therefore, by determining the relationship between the absolute value of the first frequency offset rate and the first offset threshold, the accuracy of determining whether the battery pack's structural health is at the first risk level can be improved, thereby enhancing the safety and reliability of the battery pack.

[0097] In some embodiments, step S1322 may include steps S200 and S201: Step S200: For each set of frequency offset rates, calculate the average of multiple second frequency offset rates in the set of frequency offset rates to obtain the average offset rate; In some implementations, multiple second frequency offset rates may refer to at least two second frequency offset rates, such as two second frequency offset rates, three second frequency offset rates, etc.

[0098] In some implementations, when the second frequency offset rate includes the second sub-offset rates corresponding to all the third sub-signals, the average offset rate is obtained by averaging all the second sub-offset rates corresponding to the multiple second frequency offset rates. For example, when there are two second frequency offset rates and three second sub-offset rates corresponding to each second frequency offset rate, the average offset rate is obtained by averaging the six second sub-offset rates.

[0099] Step S201: If the absolute value of the average offset rate corresponding to the multiple frequency offset rate sets is greater than or equal to the second offset threshold, the first evaluation result is determined as the second result; the second result indicates that the structural health of the battery pack is at the second risk level.

[0100] In this application, the relationship between the second offset threshold and the first offset threshold is not limited. For example, the second offset threshold can be equal to the first offset threshold, the second offset threshold can be greater than the first offset threshold, or the second offset threshold can be less than the first offset threshold.

[0101] In some implementations, if the absolute value of the average offset rate corresponding to multiple frequency offset rate sets is greater than the second offset threshold, the first evaluation result is determined as the second result.

[0102] In some implementations, if the absolute value of the average offset rate corresponding to each of the multiple frequency offset rate sets is equal to the second offset threshold, the first evaluation result is determined as the second result.

[0103] In some implementations, if a portion of the average offset rate corresponding to each of the multiple frequency offset rate sets is greater than the second offset threshold, and another portion of the average offset rate corresponding to each of the multiple frequency offset rate sets is equal to the second offset threshold, the first evaluation result is determined as the second result.

[0104] In this application, the second risk level can refer to a medium risk level. It is understood that the degree of damage to the battery pack structure under this risk level is greater than the degree of damage to the battery pack structure under the first risk level.

[0105] In this embodiment, the structural health of the battery pack is classified into different risk levels based on the relationship between the absolute value of the second frequency offset rate and the second offset threshold. The second offset threshold is the minimum value used to determine whether the battery pack's structural health is at the second risk level. Therefore, by determining the relationship between the absolute value of the second frequency offset rate and the second offset threshold, the accuracy of determining whether the battery pack's structural health is at the second risk level can be improved, thereby enhancing the safety and reliability of the battery pack.

[0106] In some embodiments, after step S201, steps S202 to S204 may be included: Step S202: If the first evaluation result is the second result, determine the third natural frequency of the battery pack based on the acquired third vibration signal; the third vibration signal is acquired by the accelerometer after acquiring multiple second vibration signals; In this application, when the structural health of the battery pack is determined to be at the second risk level, the acceleration sensor is controlled to collect a third vibration signal. It can be seen that the second risk level is determined based on multiple second vibration signals. Therefore, the collection time of the third vibration signal is later than the collection time corresponding to each of the multiple second vibration signals.

[0107] In some implementations, the structural health of the battery pack is assessed based on the acquired third vibration signal to determine the third natural frequency of the battery pack.

[0108] In implementation, when the third vibration signal includes multiple fourth sub-signals, the fourth sub-frequency corresponding to each fourth sub-signal is determined. It is understood that in this scenario, the third natural frequency includes the fourth sub-frequency corresponding to all the fourth sub-signals. In this application, the inclusion of multiple fourth sub-signals in the third vibration signal may be due to the accelerometer being a multi-axis sensor, and / or multiple accelerometers being installed on the battery pack.

[0109] Step S203: Determine the third frequency offset rate between the third natural frequency and the reference natural frequency; In some implementations, a third difference between the third natural frequency and the reference natural frequency is obtained; the third difference is divided by the reference natural frequency to obtain the third frequency offset rate.

[0110] In implementation, when the third inherent frequency includes all the fourth sub-frequencys corresponding to the fourth sub-signals, a third sub-offset rate is determined for each fourth sub-frequency relative to the reference inherent frequency. It can be understood that in this scenario, the third frequency offset rate includes the third sub-frequency offset rates corresponding to all the fourth sub-signals.

[0111] Step S204: If the absolute value of the third frequency offset rate is greater than or equal to the third offset threshold, determine that the structural health of the battery pack is at the third risk level; the third offset threshold is not less than the second offset threshold.

[0112] In some implementations, if the absolute value of the third offset rate is greater than the third offset threshold, the first evaluation result is determined as the third result.

[0113] During implementation, if at least one of the absolute values ​​of the third sub-frequency offset rate is greater than the third offset threshold, the first evaluation result is determined as the third result.

[0114] In some implementations, the first evaluation result is determined as the third result if the absolute value of the third offset rate is equal to the third offset threshold.

[0115] In this application, the third offset threshold is not less than the second offset threshold. Thus, when the absolute value of the third frequency offset rate is greater than or equal to the third offset threshold, it can be seen that the degree of damage to the battery structure under the third frequency offset rate is more severe than the degree of damage to the battery structure under the second frequency offset rate. In other words, the risk level of the third risk level is higher than that of the second risk level. For example, the third risk level can refer to a high risk level.

[0116] In some implementations, the third offset threshold is not less than the second offset threshold, which can mean that the second offset threshold is less than the third offset threshold, or that the third offset threshold is greater than the second offset threshold.

[0117] In this embodiment, after the structural health of the battery pack triggers the second risk level, the structural health of the battery pack is classified into different risk levels based on the relationship between the absolute value of the third frequency offset rate and the third offset threshold. The third offset threshold is the minimum value used to determine whether the structural health of the battery pack is at the third risk level. Therefore, by judging the relationship between the absolute value of the third frequency offset rate and the third offset threshold, the accuracy of determining whether the structural health of the battery pack is at the third risk level can be improved, thereby enhancing the safety and reliability of the battery pack.

[0118] In some embodiments, step S121 may further include one of steps S1211 and S1212: Step S1211: Determine the initial calibration frequency of the battery pack as the reference natural frequency; Here, the initial calibration frequency can be the inherent frequency calibrated by the battery pack at the factory. It is understood that this initial calibration frequency is obtained by measuring the battery pack under laboratory conditions.

[0119] In some implementations, the initial calibration frequency can be obtained by experimental modal analysis or high-fidelity simulation calculations when the battery pack is in a healthy state (i.e., without damage, loosening, or fatigue cracks).

[0120] In this embodiment of the application, the initial calibration frequency is always determined as the reference natural frequency, that is, in this scenario, the reference natural frequency of the battery pack is a fixed value.

[0121] Step S1212: Based on historical monitoring data of the battery pack, predict the reference natural frequency of the battery pack to obtain the predicted reference natural frequency; the historical monitoring data includes at least one of the following: historical frequency offset rate, historical reference natural frequency, and historical evaluation results. The historical evaluation results characterize the risk level of the battery pack structure, which can be one of a first risk level, a second risk level, and a third risk level, with the risk level increasing sequentially from the first risk level to the third risk level.

[0122] In some implementations, the historical monitoring data of the battery pack is processed using a sliding window averaging method to obtain the predicted baseline natural frequency.

[0123] In some implementations, the historical monitoring data is slid segment by segment with a first window length to obtain the arithmetic mean of all data points within each window; based on the arithmetic mean corresponding to each window, the predicted baseline natural frequency is determined.

[0124] For example, if the historical monitoring data includes a historical baseline natural frequency, and the data volume of the historical baseline natural frequency is 5, and the length of the first window is 3, then by controlling the length of the first window to move over the historical baseline natural frequency with a step size of 1, the first arithmetic mean corresponding to each of the three windows can be obtained; the mean of the first arithmetic mean corresponding to each of the three windows is calculated to obtain the first mean; the first mean is multiplied by the first window length (i.e., 3) to obtain the first sum; the difference between the first sum and the last two data points of the historical baseline natural frequency is calculated to obtain the predicted baseline natural frequency.

[0125] For example, if the historical monitoring data includes historical frequency offset rates, and the data volume of the historical frequency offset rate is 5, and the length of the first window is 3, then by controlling the length of the first window to move along the historical frequency offset rate with a step size of 1, the second arithmetic mean corresponding to each of the three windows can be obtained; the mean of the second arithmetic mean corresponding to each of the three windows is calculated to obtain the second mean; the second mean is multiplied by the length of the first window (i.e., 3) to obtain the second sum; the difference between the second sum and the last two data points of the historical frequency offset rate is calculated to obtain the first difference; since the first difference represents the frequency offset between the natural frequency of the currently acquired vibration signal and the predicted reference natural frequency, the predicted reference natural frequency can be determined based on the first difference and the natural frequency of the currently acquired vibration signal.

[0126] In some implementations, a first model is used to predict the reference natural frequency of the battery pack based on historical monitoring data, thereby obtaining the predicted reference natural frequency. The first model can be any model capable of performing this function; for example, it can be a long short-term memory neural network model or a support vector machine model.

[0127] In some implementations, when conducting the first assessment of the structural health of the battery pack, the initial calibration frequency of the battery pack can be determined as the reference natural frequency. Then, when conducting the second assessment of the structural health of the battery pack, the reference natural frequency used in the first health assessment of the battery pack can be predicted based on historical monitoring data, and so on.

[0128] In some implementations, when conducting the first assessment of the structural health of the battery pack, the initial calibration frequency of the battery pack can be predicted based on the historical vibration signals of the battery pack to obtain the predicted reference natural frequency. Then, when conducting the second assessment of the structural health of the battery pack, the reference natural frequency of the battery pack used in the first health assessment can be predicted based on historical monitoring data, and so on.

[0129] It should be noted that the frequency offset rate is determined based on the battery pack's natural frequency and reference natural frequency. The battery pack's natural frequency is determined based on the battery pack's vibration signal. Therefore, the historical frequency offset rate is obtained by monitoring the battery pack's vibration signal at historical moments. There is a corresponding relationship between the historical evaluation results and the historical frequency offset rate. As explained above, the historical frequency offset rate is obtained by monitoring the battery pack's vibration signal at historical moments; that is, the historical evaluation results are obtained by monitoring the battery pack's vibration signal at historical moments. Furthermore, the reference natural frequency is obtained based on the battery pack's historical monitoring data, and the historical monitoring data includes at least one of the historical frequency offset rate, the historical reference natural frequency, and the historical evaluation results. In other words, the historical frequency offset rate, the historical reference natural frequency, and the historical evaluation results are all related to the battery pack's vibration signal at historical moments. Therefore, in the embodiments of this application, the historical monitoring data is obtained by monitoring the battery pack's vibration signal at historical moments.

[0130] In this embodiment, on the one hand, the initial calibration frequency of the battery pack is determined as the reference natural frequency. The battery pack structure corresponding to the initial calibration frequency is in an optimal health state. By using the optimal health state of the battery pack as a benchmark, an accurate assessment of the structural health of the battery pack can be achieved. On the other hand, the reference natural frequency of the battery pack is predicted based on historical monitoring data, realizing the updating of the reference natural frequency. This allows the reference natural frequency to dynamically adapt to the drift of the battery's reference natural frequency caused by material aging and changes in the usage environment. This ensures that the structural health assessment of the battery is always based on a reference value that matches the development trend of the battery pack for offset rate calculation, which can significantly improve the accuracy and robustness of the structural health assessment of the battery pack.

[0131] In some embodiments, step S121 may further include steps S1213 and S1214: Step S1213: Obtain the initial calibration frequency of the battery pack; Step S1214: Determine the reference natural frequency based on the historical monitoring data and initial calibration frequency of the battery pack; the historical monitoring data includes at least one of the following: historical frequency offset rate, historical reference natural frequency, and historical evaluation results.

[0132] In some implementations, the initial calibration frequency of the battery pack is determined as the reference natural frequency when the structural health of the battery pack is initially assessed.

[0133] In some implementations, when performing non-first-time structural health assessments of the battery pack, a second model is used to predict the baseline inherent frequency to be used in each subsequent assessment of the battery pack's structural health, based on historical monitoring data and the baseline inherent frequency predicted in the previous assessment. This predictive baseline inherent frequency is then obtained. Non-first-time structural health assessments of the battery pack can refer to second, third, or fifth assessments, etc. The second model can be any model capable of performing this function; for example, it could be a long short-term memory neural network model or a linear regression model.

[0134] For example, during a second assessment of the battery pack's structural health, a second model is used to update the initial calibration frequency based on the historical frequency offset rate, resulting in a baseline intrinsic frequency for the second assessment. It is understood that in this scenario, the historical frequency offset rate can be the frequency offset rate determined during the initial assessment of the battery pack.

[0135] For example, when conducting a third assessment of the battery pack's structural health, the second model is used to update the baseline natural frequency used in the second assessment based on historical assessment results and historical baseline natural frequencies, resulting in the baseline natural frequency used for the third assessment. It is understandable that in this scenario, the historical assessment result can be the first assessment result determined during the second assessment of the battery pack, or it can be the first assessment result determined during both the second and first assessments. The historical baseline natural frequency can be the baseline natural frequency determined during the second assessment of the battery pack, or it can be the baseline natural frequency determined during both the second and first assessments.

[0136] For example, when conducting the fourth assessment of the structural health of the battery pack, the second model is used to update the reference natural frequency used in the third assessment based on the historical assessment results, historical frequency offset rate, and historical reference natural frequency, so as to obtain the reference natural frequency used for the fourth assessment.

[0137] For example, during the fourth assessment of the battery pack's structural health, the second model is used to update the baseline natural frequency used in the third assessment based on the historical baseline natural frequency, resulting in the baseline natural frequency used for the fourth assessment. It is understood that in this scenario, the historical baseline natural frequency can be the baseline natural frequency determined during the first, second, and third assessments of the battery pack.

[0138] In some implementations, when conducting a non-first assessment of the structural health of the battery pack, a third model is used to determine the adjustment amount for the baseline natural frequency based on historical monitoring data of the battery pack. The baseline natural frequency predicted in the previous assessment is then updated based on this adjustment amount to obtain the baseline natural frequency corresponding to the current assessment. The third model can be any model capable of performing this function; for example, it could be an autoregressive integral moving average model.

[0139] For example, during the second assessment of the structural health of the battery pack, a third model is used to determine a first adjustment based on historical monitoring data; the initial calibration frequency is then updated based on the first adjustment to obtain a reference natural frequency for the second assessment. It is understood that in this scenario, the historical monitoring data can be at least one of the frequency offset rate determined during the first assessment of the battery pack, the first assessment result, and the reference natural frequency.

[0140] For example, during the third assessment of the battery pack's structural health, a third model is used to determine a second adjustment based on historical monitoring data. This second adjustment is then used to update the reference natural frequency used in the second assessment, resulting in the reference natural frequency for the third assessment. It is understood that in this scenario, the historical monitoring data can be at least one of the frequency offset rate determined during the second assessment, the first assessment result, and the reference natural frequency; alternatively, it can be at least one of the frequency offset rate, the first assessment result, and the reference natural frequency determined during the second and first assessments, respectively.

[0141] It is understood that in the embodiments of this application, when the structural health of the battery pack is evaluated each time, the reference natural frequency used in each evaluation of the structural health of the battery pack is predicted based on historical monitoring data and the reference natural frequency predicted in the previous evaluation process, so that the updated reference natural frequency can be dynamically matched with the structural state of the battery pack.

[0142] In the above embodiments, based on the historical monitoring data and initial calibration frequency of the battery pack, the reference natural frequency of the battery pack is predicted, and the reference natural frequency is updated. This allows the determined reference natural frequency to dynamically adapt to changes in the battery pack due to material aging and usage environment. As a result, the updated reference natural frequency can dynamically match the structural state of the battery pack, significantly improving the accuracy and robustness of structural health assessment of the battery pack.

[0143] In some embodiments, step S100 above includes steps S101 and S102: Step S101: Perform frequency domain analysis on the first vibration signal to obtain the power spectral density distribution information of the battery pack; Here, frequency domain analysis is a processing method that converts time-domain signals into frequency-domain representations. Power spectral density distribution information is a key parameter describing the energy distribution of a vibration signal at different frequencies.

[0144] In some implementations, the first vibration signal is subjected to Fast Fourier Transform processing to obtain power spectral density distribution information.

[0145] In some implementations, the first vibration signal is subjected to wavelet packet transform processing to obtain power spectral density distribution information.

[0146] In this application, before performing frequency domain analysis on the first vibration signal, data preprocessing can be performed on the first vibration signal to obtain a sixth vibration signal; frequency domain analysis is then performed on the sixth vibration signal to obtain the power spectral density distribution information of the battery pack. This data preprocessing may include one or more of the following: filtering processing, noise reduction processing. It is understood that data preprocessing of the first vibration signal can improve the reliability of the data.

[0147] In practice, the first vibration signal can be filtered using a bandpass filter to obtain the sixth vibration signal. The filtering frequency of the bandpass filter can be set from 5 Hz to 500 Hz.

[0148] Step S102: Modal parameter identification is performed on the power spectral density distribution information to obtain the first natural frequency.

[0149] In some implementations, the power spectral density distribution information is processed using the peak method to obtain the first natural frequency.

[0150] In some implementations, the power spectral density distribution information is processed using a random subspace identification method to obtain the first natural frequency.

[0151] In some implementations, the power spectral density distribution information is processed using singular value decomposition to obtain the first natural frequency.

[0152] In this embodiment, firstly, frequency domain analysis is performed on the first vibration signal, which can decompose the complex time domain signal into readable power spectral density distribution information. Then, modal parameter identification is performed on the power spectral density distribution information to obtain the first natural frequency of the battery pack, providing controllable data support for realizing non-invasive and global health assessment of the battery pack structure.

[0153] In some embodiments, step S120 may be followed by step S130: Step S130: Determine and execute the target strategy corresponding to the first evaluation result; the target strategy is used to output alarm information corresponding to at least the risk level of the battery pack structure.

[0154] In some implementations, the target strategy corresponding to the first evaluation result can be obtained through a second mapping table. This second mapping table stores the processing strategies corresponding to different evaluation results.

[0155] In this embodiment of the application, by determining and executing the target strategy corresponding to the first evaluation result, and the target strategy is used to output at least alarm information corresponding to the risk level of the battery pack structure, the operating parameters of the battery pack can be adjusted according to the output alarm information, thereby slowing down the damage rate of the battery pack structure, which not only improves the battery pack life, but also improves the safety of the battery pack.

[0156] In some embodiments, step S130 may further include steps S131 to S133: Step S131: If the first evaluation result is the first result, output the first alarm message; the first result indicates that the structural health of the battery pack is at the first risk level. Here, the first warning message is used to indicate that the current structural health of the battery pack is at the first risk level.

[0157] It should be noted that if the first assessment result is the first result, the damage to the battery pack structure is relatively minor. In this scenario, only the first alarm message can be output to warn the user about the current structural health of the battery pack.

[0158] In some implementations, the first alarm message is output through the human-machine interface of a device installed in the battery pack.

[0159] In some implementations, the first alarm information is pushed to the user's terminal (e.g., mobile phone) so that the first alarm information is output through the terminal.

[0160] Step S132: If the first evaluation result is the second result, output the second alarm information and control the operating parameters of the battery pack to not exceed the target parameter threshold; the operating parameters include at least one of the following: temperature, charge / discharge rate; the second result indicates that the structural health of the battery pack is at the second risk level; Here, the charge / discharge rate is a core parameter describing the battery's charge / discharge speed, used to measure the relationship between current magnitude and the battery's rated capacity. The second warning message indicates that the battery pack's current structural health is at the second risk level.

[0161] It should be noted that when the first evaluation result is the second result, the battery pack structure is more severely damaged than in the first result. In this scenario, if only alarm information is output without controlling the operation of the battery pack, the damage rate of the battery pack structure will be accelerated. Therefore, it is necessary to control the operating parameters of the battery pack to slow down the damage rate of the battery pack structure.

[0162] In this application, different operating parameters correspond to different target parameter thresholds. For example, when the operating parameters include temperature, the corresponding target parameter threshold is the temperature threshold, and when the operating parameters include charge / discharge rate, the corresponding target parameter threshold is the rate threshold.

[0163] It should be noted that if the battery pack temperature remains high, it will cause the electrode and casing materials to expand or contract. Furthermore, the battery pack is made of components made of different materials, and these different materials (such as aluminum casing, copper busbars, and battery cells) have different coefficients of thermal expansion. This difference can generate significant thermal stress within the battery pack. When this thermal stress exceeds the material's tolerance limits, it can lead to plastic deformation, connection failure, or even casing cracking. Therefore, it is necessary to control the battery pack temperature to keep it below a certain threshold.

[0164] If the battery pack is continuously charged and discharged at a high rate, the volume of the electrode material will expand dramatically during the charging and discharging process, resulting in greater expansion force. This will directly compress the battery pack's casing, module frame, and connectors, leading to structural deformation or loosening of the connections. At the same time, operating at a high rate will also be accompanied by a significant temperature rise, which will cause thermal expansion and thermal stress, exacerbating cyclic fatigue damage to critical parts such as casing welds and electrode tab welds. Ultimately, this will reduce the mechanical integrity and service life of the battery pack. Therefore, it is necessary to control the charge and discharge rate of the battery pack to not exceed the rate threshold.

[0165] In some implementations, controlling the operating parameters of the battery pack to be no higher than the target parameter threshold can be achieved by controlling the temperature of the battery pack to be no higher than the temperature threshold; it can also be achieved by controlling the charge / discharge rate of the battery pack to be no higher than the rate threshold; or it can be achieved by controlling both the temperature of the battery pack and the charge / discharge rate of the battery pack to be no higher than the rate threshold.

[0166] Step S133: If the first evaluation result is the third result, cut off the high voltage of the battery pack, control the opening of the pressure relief valve of the battery pack, and output the third alarm information; the third result indicates that the structural health of the battery pack is at the third risk level.

[0167] Here, the charge / discharge rate is a core parameter describing the battery's charge / discharge speed, used to measure the relationship between current magnitude and battery rated capacity. The third warning message indicates that the battery pack's current structural health is at the third risk level. High voltage refers to AC or DC voltage levels exceeding a specific safety threshold, which can be a voltage posing a significant danger to humans or equipment; for example, this specific safety threshold could be 60 volts (V). A pressure relief valve (PRV) is an automatic safety device used to protect pressure vessels, pipelines, or systems from damage or explosion risks caused by excessive internal pressure.

[0168] In some implementations, the high voltage of the battery pack can be cut off by disconnecting the high voltage relay of the battery pack.

[0169] In some implementations, the high voltage of the battery pack can be cut off by disconnecting the high voltage interlock circuit.

[0170] Understandably, in this application, cutting off the high-voltage electrical energy of the battery pack stops the current flow rate within the battery pack. This halts the intense volume expansion and Joule heating of the electrode materials during charging and discharging, effectively eliminating cyclic mechanical and thermal stresses acting on the casing, welds, and connectors. This alleviates fatigue cracks, bolt loosening, and material softening caused by repeated expansion and contraction of the battery pack. Furthermore, this operation can mitigate irreversible damage to the battery pack structure caused by high-temperature, high-pressure gases and electric arcs, such as melting and perforation. Controlling the opening of the battery pack's pressure relief valve allows for the rapid discharge of high-pressure, high-temperature gases accumulated inside the battery pack, effectively mitigating plastic deformation or even rupture of the battery pack's outer casing due to pressure buildup.

[0171] In this embodiment of the application, different target strategies are set for different first evaluation results. The different target strategies are set based on the degree of structural damage to the battery pack, the rate of performance degradation, and the safety of use. This graded processing strategy can not only maximize the effective service life of the battery pack, but also alleviate the cost waste caused by over-maintenance of the battery pack.

[0172] In some embodiments, the above-described battery pack health assessment method may further include steps S300 to S330: Step S300: Acquire multiple fourth vibration signals collected by the accelerometer within a second preset time period; In some implementations, where the accelerometer is a multi-axis sensor and / or multiple accelerometers are provided on the battery pack, the fourth vibration signal may include multiple fifth sub-signals.

[0173] Step S310: For each fourth vibration signal, determine the acceleration generated by the vibration of the battery pack after it is subjected to stress based on the fourth vibration signal; In some implementations, where the fourth vibration signal includes multiple fifth sub-signals, for each fifth sub-signal, a sub-acceleration generated by the vibration of the battery pack after being subjected to stress is determined based on the fifth sub-signal. It is understood that in this scenario, the acceleration corresponding to the fourth vibration signal includes the sub-accelerations corresponding to the multiple fifth sub-signals respectively.

[0174] In some implementations, each fourth vibration signal is calibrated and converted according to the sensitivity of its corresponding accelerometer to obtain the corresponding acceleration.

[0175] In practice, the acceleration is obtained by subtracting the zero-point bias of the battery pack from the output voltage corresponding to the fourth vibration signal and dividing the first voltage difference by the sensitivity.

[0176] In some implementations, a fifth vibration signal is obtained by performing a fast Fourier transform on the fourth vibration signal; the frequency information of the fifth vibration signal is determined; and the acceleration of the frequency information is determined.

[0177] Step S320: Determine the number of targets corresponding to each of the multiple acceleration intervals; the number of targets corresponding to each acceleration interval is the number of accelerations within the acceleration interval among the accelerations corresponding to each of the fourth vibration signals; In some implementations, after obtaining the acceleration of each fourth vibration signal, the acceleration corresponding to each fourth vibration signal can be matched with multiple acceleration intervals and assigned to the matched acceleration interval.

[0178] In practice, when the acceleration includes multiple sub-accelerations, for each sub-acceleration corresponding to the fourth vibration signal, each sub-acceleration is matched with multiple acceleration intervals and assigned to the corresponding acceleration intervals.

[0179] Understandably, after dividing the acceleration corresponding to all the fourth vibration signals, the number of accelerations in each acceleration interval can be counted, and the data obtained is the target quantity corresponding to that acceleration interval. Alternatively, during the division of the fourth vibration signals, the number of accelerations in each acceleration interval can be updated in real time. In this way, after completing the division of the acceleration corresponding to all the fourth vibration signals, the target quantity corresponding to each acceleration interval can be obtained.

[0180] In some implementations, the multiple acceleration ranges can be preset by the user or obtained by updating the multiple acceleration ranges based on at least one state parameter of the battery pack.

[0181] Step S330: Based on the target quantity corresponding to multiple acceleration intervals, evaluate the structural fatigue of the battery pack to obtain the second evaluation result.

[0182] In some implementations, the second assessment result may refer to the risk level information of the structural fatigue of the battery pack.

[0183] It is understandable that the structural fatigue of a battery pack is essentially the process by which tiny defects (such as dislocations and slip bands) in the battery pack's structure gradually expand into macroscopic cracks under alternating stress, ultimately leading to fracture. The acceleration range reflects the range of different stresses the battery pack's structure endures, while the target number reflects the number of times the battery pack experiences different stress ranges. It can be seen that there is a clear correlation between the target number and the structural fatigue degree of the battery pack; increasing the target number significantly increases the structural fatigue degree, and the larger the acceleration range, the faster the fatigue accumulation rate of the battery pack under the same target number. Therefore, by using the target number corresponding to multiple acceleration ranges, the structural fatigue degree of the battery pack can be assessed.

[0184] In some implementations, for each acceleration range, the structural fatigue of the battery pack is evaluated based on the target quantity corresponding to the acceleration range and a preset quantity threshold, to obtain a second evaluation result.

[0185] During implementation, if the number of targets corresponding to the acceleration interval within the first time period is greater than or equal to the first quantity threshold, the structural fatigue of the battery pack is determined to be at the first risk level; if the number of targets corresponding to the acceleration interval within the second time period is greater than or equal to the second quantity threshold, the structural fatigue of the battery pack is determined to be at the second risk level; and if the number of targets corresponding to the acceleration interval within the third time period is greater than or equal to the third quantity threshold, the structural fatigue of the battery pack is determined to be at the third risk level.

[0186] For example, if the target quantity corresponding to the acceleration interval within 7 days is greater than or equal to 50, the structural fatigue of the battery pack is determined to be at the first risk level; if the target quantity corresponding to the acceleration interval within 30 days is greater than or equal to 200, the structural fatigue of the battery pack is determined to be at the second risk level; and if the target quantity corresponding to the acceleration interval within 14 days is greater than or equal to 500, the structural fatigue of the battery pack is determined to be at the third risk level.

[0187] In some implementations, the cumulative damage value of the battery pack is determined based on the target quantity corresponding to multiple acceleration intervals; based on the cumulative damage value and a preset damage threshold, the structural fatigue of the battery pack is evaluated to obtain a second evaluation result.

[0188] Understandably, after determining the first evaluation result, a corresponding treatment strategy can be determined for the second evaluation result. By implementing the treatment strategy, the structural fatigue of the battery pack can be reduced, effectively improving the service life of the battery pack.

[0189] In this embodiment, firstly, an accelerometer installed on the battery pack can collect multiple fourth vibration signals within a second preset time period during the battery pack's operation. Then, based on the acceleration corresponding to each of the multiple fourth vibration signals and multiple acceleration intervals, a target quantity corresponding to each acceleration interval can be obtained. Since the acceleration interval reflects the range of different stresses borne by the battery pack structure, and the target quantity reflects the number of times the battery pack bears different stress ranges, there is a clear correlation between the target quantity and the structural fatigue of the battery pack. Thus, the structural fatigue of the battery pack can be assessed by using the target quantity corresponding to each of the multiple acceleration intervals, thereby achieving a quantitative assessment of the progressive structural deterioration of the battery pack.

[0190] In some embodiments, step S330 may include steps S331 to S333: Step S331: For each acceleration interval, determine the structural damage value corresponding to the acceleration interval based on the number of targets corresponding to the acceleration interval; In some implementations, a reference quantity corresponding to the acceleration range is obtained; this reference quantity characterizes the amount the battery pack structure can withstand before failure under a constant stress level. The quotient obtained by dividing the reference quantity corresponding to the acceleration range by the target quantity is determined as the structural damage value corresponding to the acceleration range.

[0191] Step S332: Sum the structural damage values ​​corresponding to multiple acceleration intervals to obtain the cumulative damage value; Step S333: Based on the cumulative damage value, evaluate the structural fatigue of the battery pack to obtain the second evaluation result.

[0192] It should be noted that the number of times a battery pack structure can withstand different stresses varies. By statistically analyzing the number of targets experienced in each acceleration range, the cumulative damage value of these battery packs can be calculated, thus enabling the assessment of the structural fatigue of the battery pack.

[0193] In this embodiment, the cumulative damage value of the battery pack is determined based on the structural damage value in each acceleration interval, thereby realizing the assessment of the structural fatigue of the battery pack. This simplifies the complex and variable vibration and impact loads into quantifiable cumulative damage values, thus enabling early warning and maintenance of the battery pack's fatigue life.

[0194] In some embodiments, step S333 above may include step S3331: Step S3331: If the cumulative damage value is greater than or equal to the damage threshold, the second evaluation result is determined as the fourth result; the fourth result is used to characterize the structural fatigue of the battery pack as being at the third risk level.

[0195] In some implementations, the damage threshold can be set to 1.

[0196] In some implementations, if the second evaluation result is the fourth result, the high voltage of the battery pack is cut off, the pressure relief valve of the battery pack is opened, and a fourth alarm message is output.

[0197] In this embodiment of the application, by setting a loss threshold, the fatigue failure critical point of the battery pack can be clearly quantified, enabling deterministic diagnosis and timely intervention of structural damage to the battery pack, thereby improving the safety risks caused by fuzzy judgments or excessive maintenance of the battery pack.

[0198] In some embodiments, the above-described battery pack health assessment method may further include steps S340 and S350: Step S340: Obtain at least one state parameter of the battery pack; the state parameter includes one or more of the following: temperature, charging parameters, and discharging parameters; Here, charging parameters are key indicators describing the characteristics of a battery pack during the charging process, directly affecting charging efficiency, safety, lifespan, and device compatibility. These charging parameters may include, but are not limited to, one or more of the following: charging voltage, charging current, and charging time. Discharging parameters are key indicators describing the performance of a battery during the discharge process, directly affecting the device's range, output stability, and battery lifespan. These discharging parameters may include, but are not limited to, one or more of the following: discharging voltage, discharging current, and discharging capacity.

[0199] Step S350: Based on at least one state parameter, update multiple acceleration intervals to obtain updated multiple acceleration intervals.

[0200] In this application, the state parameters of the battery directly affect the material stiffness, expansion stress and vibration transmission characteristics of the battery pack, which will cause the mechanical response and fatigue damage accumulation rate of the battery pack structure to change under different acceleration ranges. Therefore, it is necessary to update multiple acceleration ranges according to at least one state parameter.

[0201] In some implementations, a first mapping relationship between temperature and adjustment amount is obtained; a first target adjustment amount is determined from the first mapping relationship based on the battery temperature; and multiple acceleration ranges of the battery are updated based on the first target adjustment amount.

[0202] In some implementations, a second mapping relationship between charging parameters and adjustment amounts is obtained, and a second target adjustment amount is determined from the second mapping relationship based on the battery's charging parameters; multiple acceleration ranges of the battery are updated based on the second target adjustment amount.

[0203] In some implementations, a third mapping relationship between discharge parameters and adjustment amounts is obtained, and a third target adjustment amount is determined from the third mapping relationship based on the battery's discharge parameters; multiple acceleration ranges of the battery are updated based on the third target adjustment amount.

[0204] In this embodiment, since changes in the state parameters of the battery cause changes in the mechanical response and fatigue damage accumulation rate of the battery pack structure under different acceleration ranges, the accuracy of the assessment of the structural fatigue of the battery pack can be improved by updating multiple acceleration ranges according to at least one state parameter.

[0205] In some embodiments, the first evaluation result includes the target fault type of the battery pack structure; step S120 above may further include step S124: Step S124: Using the target model, based on the first natural frequency, identify the structural faults of the battery pack and obtain the target fault type; the target model is trained based on at least one historical natural frequency of the battery pack and the fault type labels corresponding to each historical natural frequency.

[0206] In some implementations, the failure type may include, but is not limited to, one or more of the following: loose bolts, delamination of structural adhesive, etc.

[0207] In some implementations, the target model can be any model capable of achieving the target fault type, such as a long short-term memory neural network model or a random forest model.

[0208] In some implementations, the target model can also output a health score for the battery pack.

[0209] In this embodiment, the structural fault type of the battery pack is output through the target model, which can quickly locate specific structural faults such as cracks, loosening or deformation, providing accurate basis for targeted maintenance and safety protection of the battery pack, significantly improving the efficiency of fault handling of the battery pack and reducing the risk of secondary damage to the battery pack.

[0210] With the rapid development of new energy vehicles and energy storage systems, the structural safety and operational stability of power battery packs, as core components, are receiving increasing attention. During vehicle operation, battery packs are subject to various external disturbances such as road surface excitation, motor vibration, and temperature changes, leading to periodic or random vibrations. When the frequency of the external excitation approaches the natural frequency of the battery pack, resonance may occur, causing serious safety hazards such as structural fatigue, loosening of connectors, electrical connection failure, and even thermal runaway.

[0211] Traditional battery pack safety monitoring primarily focuses on monitoring electrical parameters such as temperature, voltage, and current, lacking real-time sensing of the structural dynamic characteristics. Although some studies have used finite element analysis or offline modal testing methods to obtain the natural frequencies of the battery pack, these methods cannot reflect the dynamic changes of the battery pack under actual operating conditions and lack real-time capability. Furthermore, during use, the natural frequencies of the battery pack gradually shift due to factors such as aging, delamination, bolt loosening, and structural deformation, making it difficult for existing technologies to achieve online monitoring of this critical structural parameter.

[0212] This application provides a battery pack health assessment system, such as... Figure 3 As shown, the system includes: an accelerometer sensor array 1, a data acquisition module 2, a signal processing and analysis module 3, a central control and decision-making module 4, and a communication and human-machine interface 5. Among these, An acceleration sensor array 1 is mounted on the battery pack body to collect vibration signals; Data acquisition module 2 is used to receive vibration signals acquired by acceleration sensor array 1; Signal processing and analysis module 3 is used to perform signal filtering, spectrum analysis and mode recognition functions; Central control and decision-making module 4 is used for load inherent frequency comparison, anomaly judgment and early warning generation; Communication and human-machine interaction interface 5 is used to realize communication with the Battery Management System (BMS), cloud platform and information interaction between humans and machines.

[0213] In some embodiments, the accelerometer array includes at least three accelerometers positioned at structurally critical points in the battery pack (e.g., the surface of the battery pack casing, critical connection points). These critical points can be determined based on theoretical experience or simulation analysis of the battery pack. During simulation analysis, stiffness contour maps and / or stress contour maps of the battery pack can be obtained. For the stiffness contour map, the structurally critical point may be located in a weak-stiff region; for the stress contour map, the structurally critical point may be located in a high-stress region.

[0214] like Figure 3 As shown, the accelerometer array includes six accelerometers. The first accelerometer 11 is located near a weak point on the battery pack's support plate; the second accelerometer 11 is located at the battery mounting and cell structural adhesive points; the third accelerometer 11 is located near the amplitude center of the battery pack (e.g., at the cell pressure plate, bottom plate, or side plate); the fourth accelerometer 11 is located near the support plate bolts; the fifth accelerometer 11 is located near the weld seam; and the sixth accelerometer 11 is located on the side wall of the housing. The accelerometers used in this application are triaxial sensors.

[0215] In some implementations, the accelerometer is mounted on the battery pack body using structural adhesive, which ensures that the accelerometer does not slip or fall off during its lifespan.

[0216] In some implementations, each accelerometer is connected to the data acquisition module wirelessly or via a wired connection to achieve synchronous acquisition of multi-degree-of-freedom vibration signals of the battery pack. These multi-degree-of-freedom vibration signals can refer to the vibration signals of the battery pack in different directions. For example, when the accelerometer is a triaxial accelerometer, it can simultaneously measure the vibration signals of the battery pack in three-dimensional space (X-axis, Y-axis, and Z-axis), meaning the multi-degree-of-freedom vibration signals can refer to vibration signals in the X-direction, Y-direction, and Z-direction.

[0217] In some implementations, when the accelerometer array includes three accelerometers, these three accelerometers can be positioned at the front (the connection point at the edge of the module), the middle (the area with poor heat dissipation), and the rear (the connection point at the edge of the module) of the module in the battery pack, which can basically cover the area of ​​the module that needs to be monitored.

[0218] In some implementations, the data sampling module can perform the following functions: multi-channel synchronous sampling, signal conditioning, and a wireless / wired communication interface. The wireless / wired communication interface can be implemented using one or more of the following methods: Controller Area Network (CAN) bus, 4G (4th Generation Mobile Communication Technology), and Ethernet. The sampling frequency corresponding to multi-channel synchronous sampling can be set to less than or equal to 1000Hz, and signal conditioning of the sampled signal can be achieved using one or more of the following methods: filtering and amplification.

[0219] In some implementations, during normal operation of the battery pack, the data acquisition module continuously acquires signals output by each acceleration sensor. The sampling frequency of the data acquisition module is not less than 1000Hz, and the acquisition time window is 5 to 30 seconds. This setting ensures that the vibration signals acquired by the data sampling module contain sufficient dynamic information.

[0220] In some implementations, the signal processing and analysis module performs the following functions: real-time filtering, FFT / WPT spectrum analysis, coherence analysis, intrinsic frequency identification, and modal parameter extraction.

[0221] In some implementations, the data sampling module transmits multiple raw vibration signals obtained after signal conditioning to the signal processing and analysis module wirelessly or via wired connection. The real-time filtering function in this module filters each raw vibration signal individually to remove noise interference. Then, a Fast Fourier Transform (FFT) or Wavelet Packet Transform (WPT) is used to extract the frequency domain features of each vibration signal output from the real-time filtering function, obtaining the power spectral density distribution of each vibration signal. Next, coherence analysis and natural frequency identification are performed on the power spectral density distributions corresponding to each accelerometer. Finally, modal parameter extraction is used to obtain the modal parameters of each vibration signal. The natural frequency identification can be based on the frequency domain peak method or the random subspace method, and the modal parameters of each vibration signal can include frequency, damping ratio, and mode shape.

[0222] In some implementations, the central control and decision-making module performs the following functions: inherent frequency offset calculation, anomaly detection logic, early warning generation, and adaptive update mechanism. The inherent frequency offset calculation can be achieved by comparing it with a reference value (i.e., the reference inherent frequency); the anomaly detection logic can be achieved through threshold judgment and trend analysis; early warning generation can be achieved through audio-visual / data reporting; and the adaptive update mechanism can be achieved through a sliding window / machine learning model.

[0223] In some implementations, after acquiring the natural frequency of each vibration signal, the central control and decision module identifies the deviation between the natural frequency and the reference natural frequency through the natural frequency offset calculation function. Then, it determines whether there is a structural abnormality in the battery pack through the anomaly judgment logic function. When it is determined that there is a structural abnormality in the battery pack, it triggers an early warning signal through the early warning generation function and sends the early warning information to the BMS or cloud monitoring platform through the CAN bus or wireless communication.

[0224] During implementation, the inherent frequency offset information between the inherent frequency and the reference inherent frequency is obtained; the inherent frequency offset information (i.e., the first offset rate mentioned above) is compared with the inherent frequency offset threshold (i.e., the first offset threshold mentioned above) (e.g., ±5%), and if the inherent frequency offset information is greater than or equal to the inherent frequency offset threshold, a warning signal is triggered.

[0225] In other implementations, the warning generation function can also determine whether abnormal features such as new frequency peaks or spectral broadening appear in the power spectral density corresponding to each vibration signal, and trigger a warning signal if abnormal features are identified. It should be noted that the warning signal can be based on audio-visual signals; for example, a voice signal can be used to trigger the warning signal, or different colored light signals can be used to trigger the warning signal.

[0226] In some implementations, based on historical monitoring data, a sliding window averaging method or machine learning models (such as long short-term memory networks or support vector machines) are used to model the evolution trend of the baseline natural frequency in order to update the baseline natural frequency. This method can adapt to long-term changes such as battery pack aging and temperature drift. The historical monitoring data may include one or more of the following: historical baseline natural frequency, historical natural frequency offset information, and whether the historical natural frequency offset information corresponds to structural risks.

[0227] In some implementations, the health assessment system can interact with the BMS / Electronic Control Unit (ECU) via a CAN bus in the communication and human-machine interface. Wireless communication methods such as 4G, 5G, and WiFi can facilitate information exchange between the health assessment system and a cloud monitoring platform. The health assessment system connects to a display screen, such as a Liquid Crystal Display (LCD) or a Light-Emitting Diode (LED), via wired or wireless means, allowing real-time monitoring of the battery pack's frequency curve and health status. The health assessment system can also communicate with mobile applications (APPs) and web platforms, enabling remote health and alarm notifications for the battery pack through these platforms.

[0228] This application provides a method for assessing the health of a battery pack, such as... Figure 4As shown, the method may include steps S401 to S413: Step S401: Acquire the multi-channel vibration time-domain signal (i.e., the first vibration signal mentioned above).

[0229] Here, the data acquisition module can obtain multi-channel vibration time-domain signals. It can be understood that these multi-channel vibration time-domain signals include sub-signals from different directions acquired by each accelerometer.

[0230] Step S402: Signal preprocessing.

[0231] Here, the signal processing and analysis unit performs signal preprocessing operations on the multi-channel vibration time-domain signal, including filtering and noise reduction.

[0232] Step S403: Frequency domain analysis.

[0233] Here, the signal processing and analysis unit performs frequency domain analysis on the signal obtained from S402 to obtain the corresponding power spectral density (i.e., the power spectral density information mentioned above). This frequency analysis can be implemented using FFT or WPT.

[0234] Step S404: Modal parameter identification.

[0235] Here, modal parameters are identified using the peak method / random subspace method based on the extracted power spectral density.

[0236] Step S405: Extract the current natural frequency and obtain the reference natural frequency.

[0237] Here, the natural frequencies corresponding to the multi-channel vibration time-domain signals can be obtained through step S404.

[0238] In some implementations, the reference natural frequency can be the factory calibration value, or the evolution trend of the reference natural frequency can be modeled using a sliding window averaging method or a machine learning model to update the reference natural frequency.

[0239] Step S406: Determine whether the cumulative attenuation of the natural frequency is greater than or equal to 5%.

[0240] Here, if yes, proceed to step S407; otherwise, proceed to step S408.

[0241] In some implementations, when the offset rate between the natural frequency and the reference natural frequency (i.e., the first offset rate mentioned above) is greater than or equal to 5%, an S1 warning (i.e., the first risk level mentioned above) is triggered.

[0242] Step S407: Trigger S1 level warning.

[0243] It should be noted that when an S1 level warning is triggered, the battery pack's operating parameters are maintained, and this warning information is only used for early alerts.

[0244] Step S408: Determine if the offset continues.

[0245] Here, if yes, proceed to step S409; otherwise, proceed to step S414.

[0246] Step S409: Initiate trend analysis and calculate the average offset using a 7-day sliding window.

[0247] In some implementations, multiple offset rates over 7 days are obtained; the average offset rate is calculated by averaging these multiple offset rates.

[0248] Step S410: The attenuation rate is greater than or equal to 3% for 3 consecutive times.

[0249] Here, if it is true, proceed to step S411; otherwise, proceed to step S414.

[0250] In some implementations, an S2 warning (i.e., the second risk level mentioned above) is triggered when the average offset rate of three consecutive times is greater than or equal to 3%.

[0251] Step S411: Trigger S2 level warning.

[0252] In some implementations, while triggering the S2 level warning, the temperature of the battery pack should also be controlled to be no higher than a first temperature threshold, and / or the charge / discharge rate of the battery pack should be controlled to be no higher than a first rate threshold.

[0253] Step S412: The cumulative attenuation rate is greater than or equal to 50%.

[0254] Here, if it is true, proceed to step S413; otherwise, proceed to step S414.

[0255] Step S413: Trigger S3 level warning (third risk level).

[0256] In some implementations, while triggering the S3 level warning, the high voltage of the battery pack should also be cut off, and the pressure relief valve of the battery pack should be opened.

[0257] Step S414: Continuously monitor and update the health score.

[0258] In this application, the battery pack's health is scored when an S1 or S2 level warning is triggered. However, the battery's health is no longer scored after an S3 level warning is triggered.

[0259] In some implementations, a battery pack health score of 8 is determined when the offset rate is 5%. A health score of 6 is determined when there are three consecutive offset rates greater than 3%. A health score of less than 4 is determined when the cumulative offset rate of the battery pack is greater than 50%.

[0260] The battery pack health assessment method provided in this application embodiment can bring the following beneficial effects: 1. Possesses proactive early warning capabilities for system structural anomalies. This application deploys a miniature accelerometer at the bottom of the power battery pack to collect vibration signals of the battery pack in real time during operation. Combined with a Battery Management System (BMS) for Fast Fourier Transform analysis, the natural frequency of the battery pack is accurately extracted. When a drop in the natural frequency exceeds a preset threshold (e.g., >5%), a structural anomaly warning is immediately triggered. This enables early proactive identification and warning of potential mechanical faults such as loose connectors, bracket deformation, and module displacement, significantly improving the foresight of battery pack safety protection.

[0261] 2. Achieve non-invasive, non-destructive mechanical condition monitoring. This application eliminates the need to disassemble the battery pack or damage the original structure (it can detect damage even when the battery has not been subjected to invasive procedures). Structural health assessment can be completed solely by collecting vibration data using an accelerometer, making it a typical non-invasive testing technology. This approach mitigates the interference with battery pack integrity and reliability caused by traditional testing methods, ensuring long-term, continuous monitoring of the battery pack's structural condition without increasing maintenance costs.

[0262] 3. Supports real-time and continuous dynamic monitoring This application leverages the real-time data processing capabilities of the BMS to continuously collect and analyze acceleration signals during vehicle operation, enabling full-lifecycle, all-weather, and high-frequency dynamic monitoring of the battery pack's structural status. Compared to periodic manual inspections or offline testing, this application offers superior response speed and data continuity, providing reliable support for structural health diagnosis.

[0263] 4. Effectively reduces the risk of thermal runaway and fire. Abnormalities in battery pack structure (such as loose bolts, structural fractures, and delamination) can easily lead to localized cell compression, electrode breakage, or internal short circuits, which are significant causes of thermal runaway. This application, by identifying structural degradation trends in advance, can issue early warnings before potential hazards escalate into serious accidents, thereby fundamentally blocking the chain reaction path of structural failure → electrical short circuit → thermal runaway and significantly improving overall vehicle safety.

[0264] 5. Controllable costs and easy to scale up deployment The MEMS accelerometer used in this application is a mature, mass-produced device. This sensor is small in size, has low power consumption, and requires no additional power supply or complex wiring. Combined with the existing BMS architecture, the overall hardware cost of the system is low, and it has strong replicability and scalability, making it suitable for applications across all scenarios, from passenger cars to commercial vehicles and special vehicles.

[0265] 6. Compatible with existing BMS architecture, enabling seamless integration. This application uses a standard communication interface (such as integrated circuit bus / serial peripheral interface / CAN) to interface with the BMS main control unit, without requiring replacement of the existing BMS hardware platform; only a spectrum analysis and diagnostic algorithm module needs to be added at the software layer. The system has good backward compatibility and scalability, and can be quickly deployed to existing production lines, reducing technology migration costs.

[0266] 7. Supports the integration of artificial intelligence (AI) technology to achieve intelligent feedback and closed-loop management. This application uploads historical vibration data and fault tags (such as loose bolts and delamination of structural adhesive) to a cloud platform, and constructs a battery pack structural health evolution model (i.e., the first model mentioned above) by combining machine learning algorithms (such as LSTM and random forest). Through an AI self-learning mechanism, it achieves fault mode recognition, remaining life prediction, and intelligent early warning push, ultimately achieving intelligent closed-loop management of "perception-analysis-early warning-feedback", and promoting the evolution of battery systems towards smart energy devices with self-diagnosis and autonomous early warning capabilities.

[0267] 8. Small size and lightweight, without increasing space or weight burden. The MEMS accelerometers used in this application are typically smaller than 3×3×1mm and weigh less than 1 gram. They can be embedded in the bottom tray of the battery pack or at the module connection point, achieving "invisible deployment." The system deployment process does not change the original structural layout, nor does it increase the volume, weight, or installation complexity of the battery pack, fully meeting the stringent requirements of new energy vehicles for lightweighting and space utilization.

[0268] Furthermore, this application proposes a health assessment method for a battery pack. By statistically analyzing the number of acceleration events exceeding a specific threshold (e.g., the number of accelerations >1.5g) experienced by the battery pack during operation, and combining this with time windows, frequency distribution, and structural fatigue models, a quantitative assessment of the degree of structural fatigue is achieved, thereby triggering different levels of safety responses.

[0269] First, let's introduce the health assessment system for the battery pack used to implement this method, such as... Figure 5As shown, the system includes: an acceleration sensor array 1, a data acquisition module 2, a fatigue acceleration event detection module 6, a sliding time window counter 7, a graded trigger judgment and decision module 8, a BMS collaborative control and response module 9, and a cloud platform and data management 10. Among these, An acceleration sensor array 1 is mounted on the battery pack body to collect vibration signals; Data acquisition module 2 is used to receive vibration signals acquired by acceleration sensor array 1; The fatigue acceleration event detection module 6 is used to detect fatigue acceleration events. The sliding time window counter 7 is used to count the number of different fatigue acceleration events; The graded trigger judgment and decision module 8 is used to identify the risk level based on the number of different acceleration events; BMS Collaborative Control and Response Module 9 is used to collaborate with the BMS to perform battery pack safety management. Cloud platform and data management 10 for remote monitoring and management of battery packs.

[0270] In some implementations, the fatigue acceleration event detection unit performs real-time filtering on the input vibration signal, and then processes the filtered vibration signal according to preset judgment conditions and defined thresholds to obtain event information of the vibration signal. This event information includes event time count, channel number, peak value, and duration. The real-time filtering can be implemented using a bandpass filter, which can be configured to range from 5Hz to 500Hz. The defined threshold can be preset by the user, for example, from 1.5g to 3.0g. The judgment condition can be that the vibration signal acquisition time is greater than 50ms, and the peak value of the acquired signal is greater than or equal to the preset defined threshold.

[0271] In some implementations, the fatigue acceleration event detection module sends its output signal to a sliding time window counter, which then counts event information within a preset window length and obtains the current cumulative value and trend change rate. The window length of this sliding event window counter can be pre-configured by the user, for example, to 7 days. The statistical indicators of the sliding window counter may include cumulative time, event frequency, and EWMA-weighted average.

[0272] In some implementations, the graded triggering judgment and decision module is connected to a sliding time window counter, allowing the module to determine the risk level of fatigue damage to the battery pack based on the counter's output. For example, if the cumulative value within the current window is greater than or equal to 50 times within 7 days, an S1 level warning is triggered. If the cumulative value within the current window is greater than or equal to 200 times within 30 days, an S2 level warning is triggered. If the cumulative value within the current window is greater than or equal to 500 times within 14 days, an S3 level warning is triggered.

[0273] In some implementations, after triggering an S1 level warning, a warning log is generated and pushed to the app. After triggering an S2 level warning, the BMS can be powered off and an alarm can be triggered. After triggering an S3 level warning, emergency response measures of the battery pack can be triggered, such as controlling the BMS to power off, triggering an alarm, and shutting off the high-pressure valve.

[0274] In some implementations, cumulative damage to the battery pack is calculated using the Miner linear model to obtain a cumulative loss value; if the cumulative loss value is greater than or equal to 1, an S3 level warning is triggered.

[0275] In some implementations, the BMS collaborative control and response module can send the triggered risk level to the BMS via flexible data rate controller LAN communication. When the BMS receives the risk level, it can perform the target operation corresponding to that level. The target operation may include one or more of the following: power outage, cooling, pressure relief, reporting to the cloud, etc.

[0276] In some implementations, the security level of the battery pack, such as S1 or S2, is sent to the cloud or terminal via Controller Area Network (CAN FD) communication.

[0277] In some implementations, the cloud platform and data management support event replay, health scoring, and remote upgrade functions. Event replay is used to reconstruct the vibration process of the battery pack via a timeline; health scoring is used to generate structural health indicators for the battery pack; and remote upgrade is used to remotely update thresholds and algorithms.

[0278] This application provides a method for assessing the health of a battery pack, such as... Figure 6 As shown, the process includes steps S601 to S606: Step S601: Acquire fatigue acceleration event data (i.e., the fourth vibration signal mentioned above); Step S602: Event type classification; In some implementations, the system can be pre-defined with acceleration intervals corresponding to different event types, and a baseline number (N) for each acceleration interval. f ).

[0279] For example, the acceleration range includes 1.5g-2.0g, 2.0g-2.5g, and 2.5g-3.0g. Among them, the reference number of accelerations is 10,000 when the acceleration range is 1.5g-2.0g, 5,000 when the acceleration range is 2.0g-2.5g, and 2,000 when the acceleration range is 2.5g-3.0g.

[0280] Step S603: Cumulative damage calculation; In some implementations, the cumulative damage value can be determined using Miner's linear cumulative damage theory. This cumulative damage value can be obtained using the following formula (1): (1); in, n This indicates the number of events (i.e., the number of acceleration intervals mentioned above). i Indicates the event type (i.e., the identifier of the acceleration range mentioned above). N i This indicates the process fatigue of the battery pack (i.e., the target quantity in the above acceleration range). N f (i) This indicates the allowable fatigue level of the battery pack.

[0281] Understandably, the damage value for each event type can be obtained using Formula 1 above. For example, with 3 events, the damage values ​​for each event type could include D1, D2, and D3. Then, the cumulative damage value is obtained by summing D1, D2, and D3.

[0282] Step S604: Determine whether critical damage has been reached; Here, the battery pack is determined to have reached critical damage when the cumulative damage value is greater than or equal to 1.

[0283] Step S605: Trigger S3 level response; In some implementations, after triggering an S3 level response, the high-voltage busbar is disconnected, the pressure relief valve is opened, and an emergency distress signal is sent.

[0284] Step S606: Continuous monitoring and early warning.

[0285] Here, by continuously monitoring the fatigue acceleration of the battery pack, early maintenance of the battery pack can be achieved.

[0286] The battery pack health assessment method provided in this application is an intelligent criterion method for long-term structural health monitoring of battery packs. This method statistically analyzes the number of acceleration events exceeding a specific threshold (e.g., >1.5g) experienced by the battery pack during operation. Combining this with a time window, frequency distribution, and a structural fatigue model (implemented using Miner's linear cumulative theory), it achieves quantitative assessment and graded response to progressive structural degradation (such as bolt loosening, increased gaps in connectors, and seal aging), resulting in the following beneficial effects: 1. Enables early identification of structural fatigue damage, filling the "blind spots" of traditional monitoring. Existing technologies only use single high-amplitude accelerations (e.g., impact acceleration > 5g) as triggering conditions, lacking the ability to identify low-amplitude, high-frequency fatigue vibrations (e.g., daily bumps, rapid acceleration). This method establishes a two-dimensional judgment mechanism through event counting and time windows. This mechanism can effectively capture fatigue excitations exceeding 50 times / 7 days cumulatively, achieving early warning of structural deterioration. Furthermore, this method considers the number of fatigue events (i.e., the aforementioned...) N i As a core indicator of structural health, it enables the monitoring of battery structure to leap from impact response to fatigue evolution.

[0287] 2. Possesses tiered response capabilities, enhancing system intelligence and reliability. The tiered mechanism can adopt a three-level triggering logic. For example, when an S1 level warning is triggered, a warning message is generated. When an S2 level warning is triggered, not only is a warning message generated, but the operating status of the battery pack is also adjusted. Then, when an S3 level warning is triggered, a warning message is generated and the high voltage of the battery pack is cut off. That is, as the risk level gradually increases, the risk response strategy for the battery pack is also adjusted accordingly, realizing refined control of risk response. This alleviates the structural misjudgment of the battery pack caused by a one-size-fits-all strategy, and realizes intelligent tiered control of "early warning for small problems and intervention for big problems", significantly improving the robustness of the system.

[0288] 3. Strong anti-interference capability, effectively reducing false alarm rate. Traditional systems are prone to misinterpreting single bumps or braking impacts as structural anomalies, leading to frequent alarms. However, this method only triggers an S1-level warning after monitoring multiple consecutive events (≥50 times / 7 days), which can significantly filter out the interference of random events on the battery pack structure. At the same time, it can combine a sliding window and an exponentially weighted moving average (EWMA) algorithm to enhance the sensitivity to trend changes in the battery pack structure and improve the accuracy of battery pack structure monitoring.

[0289] 4. Supports dynamic threshold adjustment to adapt to different operating conditions and vehicle statuses. The trigger threshold (i.e., the above-mentioned multiple acceleration ranges) can be dynamically optimized based on the following parameters (i.e., the above-mentioned state parameters): (1) Temperature: Material fatigue accelerates at high temperatures, and the threshold is automatically lowered by 10%; (2) Charge and discharge state: Vibration is enhanced during high-rate charge and discharge, and the system automatically identifies and adjusts it; (3) Driving mileage: As the mileage increases, the fatigue accumulation effect is enhanced, and the response threshold is gradually tightened. It is understandable that by dynamically adjusting the trigger threshold, the intelligent criterion of "environmental self-adaptation and state self-sensing" can be realized, thereby improving the system's generalization ability.

[0290] 5. Integration with structural fatigue models to achieve physical interpretability. This method can be combined with the Miner linear cumulative damage model to convert acceleration events into cumulative damage values. This model is based on experimental data of materials mechanics and has physical interpretability. It upgrades the method of judging structural damage from empirical judgment to physical model-driven, thereby enhancing the reliability of monitoring structural damage of battery packs.

[0291] 6. Supports remote upgrades and strategy optimization via Over-The-Air (OTA) technology. Supports OTA updates for: trigger thresholds; tiered response strategies; fatigue model parameters; and event statistics algorithms. It can be customized based on different vehicle models, road conditions, and user habits, enabling continuous system evolution.

[0292] This application provides a battery pack health assessment device, such as... Figure 7 As shown, the device 700 includes: a first acquisition module 710, used to acquire a first vibration signal of the battery pack; the first vibration signal is acquired by an acceleration sensor installed on the battery pack during the operation of the battery pack; Analysis module 720 is used to perform modal analysis on the battery pack based on the first vibration signal to obtain the first natural frequency of the battery pack; first evaluation module 730 is used to evaluate the structural health of the battery pack based on the first natural frequency to obtain a first evaluation result.

[0293] In some embodiments, the first evaluation module includes: a first acquisition unit, configured to acquire a reference natural frequency of the battery pack; the reference natural frequency is determined based on historical monitoring data of the battery pack, the historical monitoring data being obtained by monitoring the vibration signal of the battery pack at historical moments; a first determination unit, configured to determine a first frequency offset rate between the first natural frequency and the reference natural frequency; and a first evaluation unit, configured to evaluate the structural health of the battery pack based on the first frequency offset rate and a first offset threshold, and obtain a first evaluation result.

[0294] In some embodiments, the evaluation unit includes one of the following: a first determining subunit, configured to determine a first evaluation result as a first result if the absolute value of the first frequency offset rate is greater than or equal to a first offset threshold; the first result characterizes that the structural health of the battery pack is at a first risk level; and an evaluation subunit, configured to, for each of a plurality of time intervals, determine a set of frequency offset rates corresponding to the time interval based on a plurality of second vibration signals collected by an accelerometer within the time interval if the absolute value of the first frequency offset rate is less than the first offset threshold; the plurality of time intervals are later than the acquisition time of the first vibration signals; the set of frequency offset rates corresponding to the time intervals includes a second frequency offset rate between the second natural frequency and the reference natural frequency corresponding to the plurality of second vibration signals collected within the time intervals; and to evaluate the structural health of the battery pack based on the set of frequency offset rates corresponding to the plurality of time intervals to obtain the first evaluation result.

[0295] In some embodiments, the evaluation subunit includes: for each set of frequency offset rates, calculating the average of multiple second frequency offset rates in the set of frequency offset rates to obtain an average offset rate; if the absolute value of the average offset rate corresponding to each of the multiple sets of frequency offset rates is greater than or equal to a second offset threshold, determining the first evaluation result as the second result; the second result characterizes that the structural health of the battery pack is at a second risk level.

[0296] In some embodiments, the evaluation subunit includes: if the first evaluation result is the second result, determining a third natural frequency of the battery pack based on an acquired third vibration signal; the third vibration signal is acquired by an accelerometer after acquiring multiple second vibration signals; determining a third frequency offset rate between the third natural frequency and a reference natural frequency; if the absolute value of the third frequency offset rate is greater than or equal to a third offset threshold, determining that the structural health of the battery pack is at a third risk level; the third offset threshold is not less than a second offset threshold.

[0297] In some embodiments, the first acquisition module includes: a second acquisition unit for acquiring the initial calibration frequency of the battery pack; and a second determination unit for determining the reference inherent frequency based on the historical monitoring data of the battery pack and the initial calibration frequency; the historical monitoring data includes at least one of the following: historical frequency offset rate, historical reference inherent frequency, and historical evaluation results.

[0298] In some embodiments, the analysis module includes: an analysis unit for performing frequency domain analysis on the first vibration signal to obtain power spectral density distribution information of the battery pack; and an identification unit for performing modal parameter identification on the power spectral density distribution information to obtain a first natural frequency.

[0299] In some embodiments, the first determining module is used to determine and execute a target strategy corresponding to the first evaluation result; the target strategy is used to output alarm information corresponding to at least the risk level of the structure of the battery pack.

[0300] In some embodiments, the first determining module includes: a first output unit, configured to output a first alarm message when the first evaluation result is a first result; the first result indicates that the structural health of the battery pack is at a first risk level; a first control unit, configured to output a second alarm message and control the operating parameters of the battery pack to not exceed target parameter thresholds when the first evaluation result is a second result; the operating parameters include at least one of the following: temperature, charge / discharge rate; the second result indicates that the structural health of the battery pack is at a second risk level; a second control unit, configured to cut off the high voltage of the battery pack and control the opening of the pressure relief valve of the battery pack and output a third alarm message when the first evaluation result is a third result; the third result indicates that the structural health of the battery pack is at a third risk level.

[0301] In some embodiments, the device further includes: a second acquisition module, configured to acquire multiple fourth vibration signals collected by an accelerometer within a second preset time period; a second determination module, configured to, for each fourth vibration signal, determine the acceleration generated by the battery pack after being subjected to stress based on the fourth vibration signal; a third determination module, configured to determine the target quantity corresponding to each of the multiple acceleration intervals; the target quantity corresponding to the acceleration interval is the number of accelerations within the acceleration interval among the accelerations corresponding to each of the fourth vibration signals; and a third evaluation module, configured to evaluate the structural fatigue of the battery pack based on the target quantity corresponding to each of the multiple acceleration intervals, and obtain a second evaluation result.

[0302] In some embodiments, the third evaluation module includes: a third determining unit, configured to determine the structural damage value corresponding to each acceleration interval based on the number of targets corresponding to the acceleration interval; a summing unit, configured to sum the structural damage values ​​corresponding to multiple acceleration intervals respectively to obtain a cumulative damage value; and a second evaluation module, configured to evaluate the structural fatigue of the battery pack based on the cumulative damage value to obtain a second evaluation result.

[0303] In some embodiments, the second evaluation module includes: a fourth determining unit, configured to determine the second evaluation result as a fourth result when the cumulative damage value is greater than or equal to the damage threshold; the fourth result is used to characterize the structural fatigue of the battery pack as being at a third risk level.

[0304] In some embodiments, the third determining module further includes: a third acquiring unit, configured to acquire at least one state parameter of the battery pack; the state parameter includes one or more of the following: temperature, charging parameters, and discharging parameters; and an updating unit, configured to update multiple acceleration intervals based on at least one state parameter to obtain updated multiple acceleration intervals.

[0305] In some embodiments, the first evaluation result includes the target fault type of the battery pack structure; the first evaluation module includes: an identification unit, used to identify the structural fault of the battery pack based on a first natural frequency using a target model to obtain the target fault type; the target model is trained based on at least one historical natural frequency of the battery pack and the fault type labels corresponding to each historical natural frequency.

[0306] It should be noted that, in the embodiments of this application, if the above-mentioned battery pack health assessment method is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0307] Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to related technologies, can be embodied in the form of a software product. This software product is stored in a storage medium 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 methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, external hard drive, read-only memory (ROM), magnetic disk, or optical disk. Therefore, the embodiments of this application are not limited to any specific hardware and software combination.

[0308] This application provides a battery management system for implementing some or all of the steps in the above method.

[0309] This application provides a battery pack, including at least one battery cell and the aforementioned battery management system.

[0310] This application provides an electrical device including the battery pack described above.

[0311] This application provides a computer device, which includes: a memory for storing computer-executable instructions or computer programs; and a processor for executing some or all of the steps in the above method when executing the computer-executable instructions or computer programs stored in the memory.

[0312] This application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements some or all of the steps in the above-described method. The computer-readable storage medium can be transient or non-transient.

[0313] This application provides a computer program product, including a computer program or instructions, which, when executed by a processor, implement some or all of the steps in the above-described method.

[0314] It should be noted that the descriptions of the above-described storage media, computer program products, and device embodiments are similar to the descriptions of the above-described method embodiments, and have similar beneficial effects. For technical details not disclosed in the embodiments of the storage media, computer program products, and devices of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0315] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in the various embodiments of this application, the sequence numbers of the above-described processes do not imply a sequential order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application. The sequence numbers of the above-described embodiments are merely descriptive and do not represent the superiority or inferiority of the embodiments.

[0316] 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 apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0317] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0318] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0319] In addition, each functional unit in the various embodiments of this application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0320] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, read-only memory (ROM), magnetic disks, or optical disks.

[0321] Alternatively, if the integrated units described above are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence or the part that contributes to related technologies, can be embodied in the form of a software product. This computer software product is stored in a storage medium 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 methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0322] The above are merely embodiments of this application and are not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, and improvements made within the spirit and scope of this application are included within the scope of protection of this application.

Claims

1. A method for health assessment of a battery pack, characterized in that, include: Acquire the first vibration signal of the battery pack; The first vibration signal was collected by an accelerometer mounted on the battery pack during the operation of the battery pack; Based on the first vibration signal, modal analysis is performed on the battery pack to obtain the first natural frequency of the battery pack; Based on the first inherent frequency, the structural health of the battery pack is evaluated to obtain a first evaluation result; The step of evaluating the structural health of the battery pack based on the first inherent frequency to obtain a first evaluation result includes: The reference natural frequency of the battery pack is obtained; the reference natural frequency is determined based on the historical monitoring data of the battery pack, and the historical monitoring data is obtained by monitoring the vibration signal of the battery pack at historical moments; Determine a first frequency offset rate between the first natural frequency and the reference natural frequency; Based on the first frequency offset rate and the first offset threshold, the structural health of the battery pack is evaluated to obtain the first evaluation result.

2. The health assessment method for a battery pack according to claim 1, characterized in that, The structural health of the battery pack is evaluated based on the first frequency offset rate and the first offset threshold to obtain the first evaluation result, which includes one of the following: If the absolute value of the first frequency offset rate is greater than or equal to the first offset threshold, the first evaluation result is determined as the first result; the first result indicates that the structural health of the battery pack is at a first risk level. If the absolute value of the first frequency offset rate is less than the first offset threshold, for each of the multiple time intervals, based on the multiple second vibration signals collected by the accelerometer within the time interval, a set of frequency offset rates corresponding to the time interval is determined; the multiple time intervals are later than the acquisition time of the first vibration signal; the set of frequency offset rates corresponding to the time interval includes the second frequency offset rate between the second natural frequency corresponding to the multiple second vibration signals collected within the time interval and the reference natural frequency; based on the set of frequency offset rates corresponding to the multiple time intervals, the structural health of the battery pack is evaluated to obtain the first evaluation result.

3. The health assessment method for the battery pack according to claim 2, characterized in that, The assessment of the structural health of the battery pack based on the frequency offset rate sets corresponding to the multiple time intervals, to obtain the first assessment result, includes: For each set of frequency offset rates, the average offset rate is obtained by averaging multiple second frequency offset rates in the set of frequency offset rates. If the absolute value of the average offset rate corresponding to each of the plurality of frequency offset rate sets is greater than or equal to the second offset threshold, the first evaluation result is determined to be the second result; the second result indicates that the structural health of the battery pack is at the second risk level.

4. The health assessment method for the battery pack according to claim 3, characterized in that, The method further includes: If the first evaluation result is the second result, the third natural frequency of the battery pack is determined based on the acquired third vibration signal; the third vibration signal is acquired by the accelerometer after acquiring the plurality of second vibration signals; Determine the third frequency offset rate between the third natural frequency and the reference natural frequency; If the absolute value of the third frequency offset rate is greater than or equal to the third offset threshold, the structural health of the battery pack is determined to be at the third risk level; the third offset threshold is not less than the second offset threshold.

5. The health assessment method for a battery pack according to claim 1, characterized in that, The step of obtaining the reference inherent frequency of the battery pack includes: Obtain the initial calibration frequency of the battery pack; Based on the historical monitoring data of the battery pack and the initial calibration frequency, the reference natural frequency is determined; the historical monitoring data includes at least one of the following: historical frequency offset rate, historical reference natural frequency, and historical evaluation results.

6. The health assessment method for a battery pack according to any one of claims 1 to 5, characterized in that, The step of performing modal analysis on the battery pack based on the first vibration signal to obtain the first natural frequency of the battery pack includes: Frequency domain analysis is performed on the first vibration signal to obtain the power spectral density distribution information of the battery pack; Modal parameter identification is performed on the power spectral density distribution information to obtain the first natural frequency.

7. The health assessment method for a battery pack according to any one of claims 1 to 5, characterized in that, The method further includes: Determine and execute the target strategy corresponding to the first evaluation result; the target strategy is used to output alarm information corresponding to at least the risk level of the battery pack structure.

8. The health assessment method for a battery pack according to claim 7, characterized in that, The step of determining and executing the target strategy corresponding to the first evaluation result includes: If the first assessment result is the first result, a first alarm message is output; the first result indicates that the structural health of the battery pack is at the first risk level. If the first evaluation result is the second result, a second alarm message is output, and the operating parameters of the battery pack are controlled to not exceed the target parameter threshold; the operating parameters include at least one of the following: temperature, charge / discharge rate; the second result indicates that the structural health of the battery pack is at the second risk level; If the first assessment result is the third result, the high voltage of the battery pack is cut off, the pressure relief valve of the battery pack is opened, and a third alarm message is output; the third result indicates that the structural health of the battery pack is at the third risk level.

9. The health assessment method for a battery pack according to any one of claims 1 to 5, characterized in that, The method further includes: Acquire multiple fourth vibration signals collected by the accelerometer within a second preset time period; For each fourth vibration signal, the acceleration generated by the vibration of the battery pack after being subjected to stress is determined based on the fourth vibration signal; Determine the number of targets corresponding to each of the multiple acceleration intervals; the number of targets corresponding to each acceleration interval is the number of accelerations that fall within the acceleration interval among the accelerations corresponding to each of the fourth vibration signals. Based on the target quantity corresponding to the multiple acceleration intervals, the structural fatigue of the battery pack is evaluated to obtain a second evaluation result.

10. The health assessment method for a battery pack according to claim 9, characterized in that, The structural fatigue of the battery pack is evaluated based on the target quantities corresponding to the multiple acceleration intervals, resulting in a second evaluation result, including: For each acceleration interval, the structural damage value corresponding to the acceleration interval is determined based on the number of targets corresponding to the acceleration interval; The cumulative damage value is obtained by summing the structural damage values ​​corresponding to the multiple acceleration intervals respectively. Based on the cumulative damage value, the structural fatigue of the battery pack is evaluated to obtain a second evaluation result.

11. The health assessment method for a battery pack according to claim 10, characterized in that, The assessment of the structural fatigue of the battery pack based on the cumulative damage value yields a second assessment result, including: If the cumulative damage value is greater than or equal to the damage threshold, the second evaluation result is determined as the fourth result; the fourth result is used to characterize the structural fatigue of the battery pack as being at the third risk level.

12. The health assessment method for a battery pack according to claim 9, characterized in that, The method further includes: Obtain at least one state parameter of the battery pack; the state parameter includes one or more of the following: temperature, charging parameters, and discharging parameters; Based on the at least one state parameter, the plurality of acceleration intervals are updated to obtain the updated plurality of acceleration intervals.

13. The health assessment method for a battery pack according to any one of claims 1 to 5, characterized in that, The first evaluation result includes the target failure type of the battery pack structure; Based on the first inherent frequency, the structural health of the battery pack is evaluated to obtain a first evaluation result, including: Using the target model, based on the first inherent frequency, the structural faults of the battery pack are identified to obtain the target fault type; The target model is trained based on at least one historical inherent frequency of the battery pack and the fault type labels corresponding to each historical inherent frequency.

14. A battery management system, characterized in that, The battery management system is used to implement the method according to any one of claims 1 to 13.

15. A battery pack, characterized in that, The battery pack includes at least one battery cell and the battery management system as described in claim 14.

16. An electrical appliance, characterized in that, The electrical equipment includes the battery pack as described in claim 15.

17. A health assessment device for a battery pack, characterized in that, The device includes: The first acquisition module is used to acquire a first vibration signal of the battery pack; the first vibration signal is collected by an acceleration sensor installed on the battery pack during the operation of the battery pack; An analysis module is used to perform modal analysis on the battery pack based on the first vibration signal to obtain the first natural frequency of the battery pack; The first evaluation module is used to evaluate the structural health of the battery pack based on the first inherent frequency and obtain a first evaluation result. The first evaluation module includes: The first acquisition unit is used to acquire the reference natural frequency of the battery pack; the reference natural frequency is determined based on the historical monitoring data of the battery pack, and the historical monitoring data is obtained by monitoring the vibration signal of the battery pack at historical moments; The first determining unit is configured to determine a first frequency offset rate between the first intrinsic frequency and the reference intrinsic frequency; The first evaluation unit is used to evaluate the structural health of the battery pack based on the first frequency offset rate and the first offset threshold, and obtain the first evaluation result.

18. A computer device, characterized in that, The computer device includes: Memory is used to store executable instructions or computer programs. A processor, when executing computer-executable instructions or computer programs stored in the memory, implements the method according to any one of claims 1 to 13.

19. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the steps of the method as described in any one of claims 1 to 13.

20. A computer program product, characterized in that, It includes a computer program or instructions that, when executed by a processor, implement the steps of the method according to any one of claims 1 to 13.

Images