Battery shell torque testing method and torque testing system

Abstract

The present invention discloses a battery shell torque testing method and torque testing system, including obtaining stress-strain relationship data and structural micro-strain monitoring data of the battery shell under the initial load state; torque loading the shell according to the obtained data, and obtaining crack evolution characteristic data including multi-band optical deformation data and micro-structure stress response data; determining and applying the torque loading gradient according to the characteristic data, and obtaining high-frequency strain data, real-time optical monitoring data and conductive network status data; performing segmented comparison on the preset area, obtaining time interval data and harmonic acoustic characteristic data, and generating repair efficiency evaluation data; generating a dynamic safety threshold curve and determining the torque limit range; and performing real-time judgment on the operating state of the shell to generate a use safety evaluation result. The technical solution of the present invention realizes the performance evaluation of the self-repairing shell in extreme environments, and provides a reliable basis for the safety judgment of the shell.

Description

Technical Field

[0001] The present invention relates to the technical field of battery casing torque testing, and in particular to a battery casing torque testing method and a torque testing system. Background Art

[0002] In modern high-performance battery systems, the structural integrity and reliability of the battery shell have a crucial impact on the overall performance. In order to improve the durability and safety of the battery shell in complex and extreme environments, researchers have developed and applied self-healing and self-monitoring materials. Such materials are usually composed of a composite structure with a microencapsulated repair agent or a nanoscale conductive fiber network embedded inside. When the battery shell develops microcracks or is subjected to stress concentration during use, the self-healing material can automatically release the repair agent to fill the cracks, thereby restoring the structural integrity of the shell. At the same time, the built-in conductive or optical sensing network can monitor the stress changes and crack development inside the shell in real time, providing instant health status feedback. These self-healing and self-monitoring technologies not only significantly enhance the fatigue resistance and corrosion resistance of the battery shell, but also enable it to maintain higher reliability and stability under harsh conditions such as temperature changes, frequent vibrations and chemical erosion, ensuring the efficient operation and long life of the battery system in a changing environment.

[0003] However, existing technologies face significant limitations when evaluating the mechanical properties of battery shells embedded with self-healing or self-monitoring materials. Under extreme environmental conditions, such as rapid temperature changes, high-frequency vibrations, and chemical corrosion, the formation and self-healing process of microcracks inside the battery shell becomes extremely complex and dynamic. Traditional torque testing methods mainly rely on macroscopic mechanical parameters, such as the relationship between overall torque and resistance, ignoring subtle changes in internal stress distribution and the dynamic process of release and diffusion of the repair agent. This method cannot capture and accurately evaluate the microscopic damage inside the material and its self-healing effect in real time, resulting in a deviation in the judgment of the actual safety margin of the battery shell. Specifically, the local stress concentration area and the real-time reaction of the repair agent have not been effectively monitored, so that in actual applications, the battery shell may fail prematurely due to failure to identify and repair microcracks in time, or introduce unnecessary redundancy in the design due to inaccurate evaluation. Therefore, there is an urgent need for a torque testing method that can combine the self-healing characteristics of the battery shell and monitor the evolution and repair process of microcracks in real time to accurately evaluate its performance in complex environments and ensure the efficient and safe operation of the battery system. Summary of the invention

[0004] The main purpose of the present invention is to solve the problem that the prior art ignores the dynamic process of internal stress distribution and repair agent release of the material when evaluating the mechanical properties of the battery casing embedded with self-healing or self-monitoring materials, resulting in the inability to accurately evaluate the actual safety margin of the battery casing under extreme environmental conditions.

[0005] A first aspect of the present invention provides a battery housing torque testing method, the battery housing torque testing method comprising:

[0006] Acquire stress-strain relationship data of the battery housing under an initial load state, and acquire structural microstrain monitoring data of the battery housing under the initial load state, and generate initial self-repair monitoring reference data of the battery housing according to the stress-strain relationship data and the structural microstrain monitoring data, wherein the initial load causes the battery housing to be in an elastic deformation stage;

[0007] Performing progressive torsion loading on the battery housing according to the initial self-repair monitoring benchmark data, acquiring multi-band optical deformation data and microstructure stress response data of the battery housing during the progressive torsion loading process, and generating crack evolution characteristic data of the battery housing according to the time correlation between the multi-band optical deformation data and the microstructure stress response data;

[0008] Determine a torque loading gradient according to the crack evolution characteristic data, apply the torque loading gradient to the battery housing, obtain high-frequency strain data, real-time optical monitoring data, and conductive network state data of the battery housing under different torque gradients, and generate repair process characterization data of the battery housing according to the high-frequency strain data, the real-time optical monitoring data, and the conductive network state data;

[0009] According to the repair process characterization data, a preset area of ​​the battery housing is segmented and compared under multiple torque gradients to obtain time interval data between a crack formation moment and a repair agent activation moment in the preset area, as well as harmonic acoustic characteristic data of the preset area, and according to the time interval data and the harmonic acoustic characteristic data, generate repair efficiency evaluation data of the preset area;

[0010] Analyzing and processing the repair efficiency evaluation data to generate a dynamic safety threshold curve of the battery housing, and determining a torque limit range of the battery housing according to the dynamic safety threshold curve;

[0011] According to the torque limit range and the dynamic safety threshold curve, the operating state of the battery housing is determined in real time to generate a safety assessment result of the battery housing.

[0012] Optionally, the step of acquiring stress-strain relationship data of the battery housing under an initial load state, and acquiring structural microstrain monitoring data of the battery housing under the initial load state, and generating initial self-repair monitoring reference data of the battery housing according to the stress-strain relationship data and the structural microstrain monitoring data, comprises:

[0013] Apply a torque loading sequence to the battery housing to obtain multi-level stress-strain relationship data of the battery housing under the torque loading sequence, wherein the torque loading sequence includes an initial loading segment and an incremental loading segment, the loading value of the initial loading segment is greater than the torque value of the elastic deformation starting point of the battery housing and less than the yield limit torque value of the battery housing, the starting loading value of the incremental loading segment is the loading value of the initial loading segment, the loading process of the incremental loading segment is divided into a plurality of loading sub-segments with equal time intervals, the torque value of each loading sub-segment increases a preset torque increment compared to the previous loading sub-segment, and the terminal loading value of the incremental loading segment does not exceed the yield limit torque value;

[0014] Acquire structural microstrain monitoring data of the battery housing through a preset collection point array on the surface of the battery housing, wherein the structural microstrain monitoring data includes initial structural microstrain distribution data under the loading value of the initial loading section and dynamic structural microstrain distribution data under the loading values ​​of each level of the incremental loading section;

[0015] During each loading value maintenance process of the initial loading section and the incremental loading section, regional scanning is performed on the corresponding structural microstrain distribution data to obtain the repair agent trigger signal data of each area of ​​the battery housing;

[0016] Performing time series correlation analysis on the multi-level stress-strain relationship data and the repair agent trigger signal data to generate stress fluctuation response characteristic data of the battery housing;

[0017] The self-repair sensitivity parameter of the battery casing is determined according to the stress fluctuation response characteristic data, and initial self-repair monitoring reference data of the battery casing is generated.

[0018] Optionally, performing a time series correlation analysis on the multi-level stress-strain relationship data and the repair agent trigger signal data to generate stress fluctuation response characteristic data of the battery housing includes:

[0019] The multi-level stress-strain relationship data are grouped according to the time series of applied loads, and the strain mutation points and strain recovery points in each group of data are extracted to generate key stress response time data of the battery housing;

[0020] Performing time domain analysis on the repair agent trigger signal data, recording the initial activation moment and complete diffusion moment of the repair agent in each torque loading sub-segment, pairing and marking the repair agent trigger signal data based on the key stress response moment data, and generating repair response synchronization data of the battery housing;

[0021] The stress recovery coefficient and the diffusion coefficient of the repair agent of the battery housing in different torque loading sub-segments are calculated according to the repair response synchronization data to generate stress fluctuation response characteristic data of the battery housing.

[0022] Optionally, the battery case is subjected to progressive torsion loading according to the initial self-repair monitoring benchmark data, multi-band optical deformation data and microstructure stress response data of the battery case during the progressive torsion loading process are obtained, and crack evolution characteristic data of the battery case is generated according to the time correlation between the multi-band optical deformation data and the microstructure stress response data, including:

[0023] Dividing the initial self-repair monitoring benchmark data into regions to generate a surface stress distribution map of the battery housing and an internal repair agent distribution map;

[0024] Determine a progressive torque loading scheme according to the surface stress distribution map and the internal repair agent distribution map, wherein the progressive torque loading scheme divides the progressive torque loading into a plurality of loading cycles, the holding time of each loading cycle is proportional to the maximum torque value in the corresponding loading cycle, the torque value in each loading cycle increases step by step according to a preset interval, and there is an overlapping section of the torque values ​​between adjacent loading cycles;

[0025] In each loading cycle, the surface deformation image of the battery shell is collected through ultraviolet, visible light and near infrared three frequency bands to obtain the strain field data of the battery shell in different frequency bands, and the strain field data of the battery shell in the three frequency bands of ultraviolet, visible light and near infrared are compared and synthesized to generate multi-band optical deformation data;

[0026] Acquiring conductive network signal data and temperature gradient distribution data inside the battery housing in the overlapping section of each loading cycle, and generating microstructure stress response data according to the conductive network signal data and the temperature gradient distribution data;

[0027] Pairing the multi-band optical deformation data and the microstructure stress response data according to timestamps to generate crack characterization data of the battery housing in each loading cycle, wherein the crack characterization data includes surface crack morphology data and internal crack extension data;

[0028] The crack characterization data in each loading cycle is subjected to time series superposition analysis to obtain crack evolution correlation data between different loading cycles, and to generate crack evolution characteristic data of the battery shell.

[0029] Optionally, determining a torque loading gradient according to the crack evolution characteristic data, applying the torque loading gradient to the battery housing, acquiring high-frequency strain data, real-time optical monitoring data, and conductive network state data of the battery housing under different torque gradients, and generating repair process characterization data of the battery housing according to the high-frequency strain data, the real-time optical monitoring data, and the conductive network state data, including:

[0030] Classifying the crack evolution characteristic data according to crack depth and propagation speed to generate crack classification data, and determining the classification interval and torque load sequence of the torque loading gradient according to the crack classification data, wherein the torque load sequence includes a constant torque load and a pulse vibration load;

[0031] The constant torque load and the pulse vibration load are sequentially applied to the battery housing, wherein the constant torque load is increased in a stepwise manner at a preset rate, and the pulse vibration load is superimposed with a vibration stress of a preset amplitude on the basis of the current constant torque load, and high-frequency strain data is acquired by a dual time scale sampling method of millisecond and sub-millisecond levels;

[0032] During the loading process of the constant torque load and the pulse vibration load, synchronously collecting data on the battery housing, acquiring surface strain transient distribution data and internal conductivity fluctuation data, and generating real-time optical monitoring data and conductive network status data;

[0033] Performing time synchronization pairing on the high-frequency strain data, the real-time optical monitoring data and the conductive network state data to generate repair behavior characteristic data of the battery housing under the torque loading gradient;

[0034] The strain recovery rate parameters, the flow characteristic parameters of the repair agent and the conductive path reconstruction parameters of the battery shell are extracted according to the repair behavior characteristic data to generate the repair process characterization data of the battery shell.

[0035] Optionally, the preset area of ​​the battery housing is segmentedly compared under multiple torque gradients according to the repair process characterization data, and the time interval data between the crack formation time and the repair agent activation time of the preset area and the harmonic acoustic characteristic data of the preset area are obtained; and the repair efficiency evaluation data of the preset area is generated according to the time interval data and the harmonic acoustic characteristic data, including:

[0036] Performing segmented analysis on the repair process characterization data according to the torque gradient, dividing the crack-prone areas of the battery housing, and generating segmented scanning data of a preset area;

[0037] According to the segmented scanning data of the preset area, the strain mutation point and the conductive path recovery point of the battery housing are recorded in each torque gradient interval, the time series evolution data of the preset area is generated, and the time interval data between the crack formation time and the repair agent activation time are extracted from the time series evolution data;

[0038] At the key monitoring position determined by the segmented scanning data of the preset area, the acoustic response data of the repair process is collected for the crack formation time and the repair agent activation time in the time series evolution data, and the harmonic acoustic characteristic data of the preset area is generated;

[0039] Cross-validating the time interval data and the harmonic acoustic feature data to generate repair quality evaluation data of the preset area at each torque gradient;

[0040] The repair depth parameter and the repair speed parameter of the preset area are determined according to the repair quality evaluation data, and the repair efficiency evaluation data of the preset area is generated.

[0041] Optionally, analyzing and processing the repair efficiency evaluation data to generate a dynamic safety threshold curve of the battery housing, and determining the torque limit range of the battery housing according to the dynamic safety threshold curve includes:

[0042] Performing torque gradient correlation analysis on the repair efficiency evaluation data, extracting repair depth data and repair speed data of the battery shell under different torque gradients, and generating repair performance grading data of the battery shell;

[0043] The repair saturation degree of the battery housing under each torque gradient is calibrated according to the repair performance classification data, the repair agent consumption rate data and the repair layer stability data are obtained, and the repair limit state data of the battery housing is generated;

[0044] The repair limit state data are grouped according to the safety margin requirements, and safety torque reference data of the battery housing in three repair efficiency intervals of high, medium and low are generated, and a dynamic safety threshold curve of the battery housing is drawn according to the safety torque reference data;

[0045] Acquiring critical torque data of the battery housing in different repair efficiency intervals based on the dynamic safety threshold curve, and performing correction coefficient compensation on the critical torque data to generate safety margin correction data of the battery housing;

[0046] The position of the warning point of the dynamic safety threshold curve is adjusted according to the safety margin correction data to determine the torque limit range of the battery housing.

[0047] Optionally, the operating state of the battery housing is determined in real time based on the torque limit range and the dynamic safety threshold curve to generate a use safety assessment result of the battery housing, including:

[0048] Dividing the torque limit range and the dynamic safety threshold curve into regions, dividing the operating state of the battery housing into a normal operating range, a warning monitoring range, and a risk control range, and generating operating state partition data of the battery housing;

[0049] Determining a monitoring parameter threshold of the battery housing in different operating intervals according to the operating status partition data, and performing torque gradient calibration on the monitoring parameter threshold to generate status monitoring criterion data of the battery housing;

[0050] Analyzing the strain response characteristics of the battery shell and the activity state of the repair agent according to the state monitoring criterion data, and generating safety risk level data of the battery shell, wherein the safety risk level data includes structural integrity data of the battery shell, remaining amount data of the repair agent, and re-repairability data;

[0051] Establishing a maintenance cycle index of the battery housing based on the safety risk level data, cross-validating the maintenance cycle index with the dynamic safety threshold curve, and generating safety margin evaluation data of the battery housing;

[0052] The key early warning nodes and emergency handling procedures of the battery case are set according to the safety margin evaluation data, and the use safety evaluation results of the battery case are generated.

[0053] Optionally, the strain response characteristics of the battery housing and the activity state of the repair agent are analyzed by comparing the state monitoring criterion data to generate safety risk level data of the battery housing, wherein the safety risk level data includes structural integrity data of the battery housing, remaining amount data of the repair agent, and re-repairability data, including:

[0054] Performing periodic analysis on the strain response characteristics in the state monitoring criterion data, extracting the strain accumulation rate and strain recovery ratio of the battery housing in a continuous working cycle, and generating structural dynamic stability data of the battery housing;

[0055] According to the structural dynamic stability data, a repair agent loss curve of the battery shell under different operating conditions is established, the residual activity level and diffusion penetration capacity of the repair agent are calculated, and the repair system availability data of the battery shell is generated;

[0056] The structural dynamic stability data and the repair system availability data are comprehensively evaluated to generate safety risk level data of the battery shell, and the safety risk level data reflects the structural integrity data, the remaining amount of the repair agent data and the re-repairability data of the battery shell.

[0057] A second aspect of the present invention provides a torque testing system, the torque testing system comprising:

[0058] An initial state monitoring module, used to obtain stress-strain relationship data of the battery housing under an initial load state, and obtain structural microstrain monitoring data of the battery housing under the initial load state, and generate initial self-repair monitoring reference data of the battery housing according to the stress-strain relationship data and the structural microstrain monitoring data, wherein the initial load causes the battery housing to be in an elastic deformation stage;

[0059] A progressive loading control module, used to perform progressive torsion loading on the battery housing according to the initial self-repair monitoring benchmark data, obtain multi-band optical deformation data and microstructure stress response data of the battery housing during the progressive torsion loading process, and generate crack evolution characteristic data of the battery housing according to the time correlation between the multi-band optical deformation data and the microstructure stress response data;

[0060] a gradient loading test module, for determining a torque loading gradient according to the crack evolution characteristic data, applying the torque loading gradient to the battery housing, acquiring high-frequency strain data, real-time optical monitoring data and conductive network state data of the battery housing under different torque gradients, and generating repair process characterization data of the battery housing according to the high-frequency strain data, the real-time optical monitoring data and the conductive network state data;

[0061] A repair efficiency evaluation module, used to perform segmented comparison of a preset area of ​​the battery housing under multiple torque gradients according to the repair process characterization data, obtain time interval data between the crack formation moment and the repair agent activation moment of the preset area, and harmonic acoustic characteristic data of the preset area, and generate repair efficiency evaluation data of the preset area according to the time interval data and the harmonic acoustic characteristic data;

[0062] A safety threshold analysis module, used to analyze and process the repair efficiency evaluation data, generate a dynamic safety threshold curve of the battery housing, and determine the torque limit range of the battery housing according to the dynamic safety threshold curve;

[0063] The operating state determination module is used to determine the operating state of the battery housing in real time according to the torque limit range and the dynamic safety threshold curve, and generate a safety evaluation result of the battery housing.

[0064] The technical solution provided by the embodiments of the present application has at least the following advantages:

[0065] First, the method generates initial self-repair monitoring benchmark data by acquiring the stress-strain relationship data of the battery shell under the initial load state and the structural microstrain monitoring data. The establishment of this benchmark data enables the accurate identification of the formation of microcracks and the release of self-repair agents during the subsequent torsion loading process, ensuring the accuracy of the test starting point and the comparability of the data. By applying an initial load to the shell, which is in the elastic deformation stage, the material is prevented from entering the plastic range too early, thereby ensuring the linear reliability of the stress-strain data.

[0066] Next, the implementation of progressive torque loading makes the torque application process more detailed and controllable. By gradually increasing the torque and synchronously acquiring multi-band optical deformation data and microstructural stress response data, the method can capture the evolution characteristics of the crack in real time. This combination of staged loading and monitoring can not only dynamically track the formation and expansion of cracks, but also monitor the release and effect of self-healing agents at the cracks in real time. Multi-band optical deformation data provides detailed deformation information on the shell surface, while microstructural stress response data reveals changes in internal stress distribution. This multi-dimensional data acquisition method enables the accurate generation of crack evolution characteristic data, providing a scientific basis for the determination of torque loading gradients in subsequent steps.

[0067] After determining the torque loading gradient, the corresponding torque loading is applied and high-frequency strain data, real-time optical monitoring data and conductive network state data are obtained to further generate repair process characterization data. This process can record in detail the key parameters in the self-repair process, such as strain recovery rate and repair agent flow characteristics, by combining high-frequency strain data and real-time optical monitoring data. The conductive network state data reflects the reconstruction of the conductive pathway inside the material and provides electrical verification of the repair effect. The generation of these repair process characterization data makes the evaluation of self-repair efficiency more comprehensive and accurate.

[0068] By comparing the repair process characterization data in sections, obtaining the time interval data between the crack formation moment and the repair agent activation moment and the harmonic acoustic characteristic data, the method can generate repair efficiency evaluation data. This evaluation process not only examines the filling effect of the self-healing agent at the crack, but also deeply analyzes the changes in the acoustic response inside the material through the harmonic acoustic characteristic data, further verifying the depth and quality of the repair. The generation of repair efficiency evaluation data ensures the accurate judgment of the effectiveness of the self-healing mechanism and provides solid data support for the formulation of the dynamic safety threshold curve.

[0069] Finally, by analyzing and processing the repair efficiency evaluation data, a dynamic safety threshold curve is generated, and the torque limit range is determined accordingly. This method realizes the dynamic safety evaluation of the battery shell under extreme environments. The dynamic safety threshold curve not only reflects the material's load-bearing capacity under different torque gradients, but also dynamically adjusts the torque limit according to real-time monitoring data to ensure that the battery shell is always within the safe working range. This dynamic judgment mechanism takes into account the performance changes of self-healing materials in extreme environments and ensures the efficient and safe operation of the battery system.

[0070] In summary, the battery shell torque test method combines multi-dimensional data synchronous acquisition, progressive torque loading and real-time monitoring to comprehensively solve the problem that traditional test methods cannot capture microcrack evolution and self-repair process in real time. By accurately generating initial benchmark data, dynamically capturing crack evolution characteristics, carefully evaluating repair efficiency and dynamically adjusting safety thresholds, the method achieves a comprehensive performance evaluation of self-repairing battery shells in complex environments, ensuring their reliability and safety under extreme conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0072] Figure 1 A schematic diagram of an embodiment of a battery housing torque testing method in an embodiment of the present invention;

[0073] Figure 2 FIG. 4 is a schematic diagram of an embodiment of a torque testing system in an embodiment of the present invention.

[0074] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0075] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0076] It should be noted that if the embodiments of the present invention involve directional indications (such as up, down, left, right, front, back...), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0077] In addition, the descriptions of "first", "second", etc. in the present invention are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, "and / or" in the full text includes three solutions. Taking A and / or B as an example, it includes technical solution A, technical solution B, and technical solution that satisfies both A and B. In addition, the technical solutions between the various embodiments can be combined with each other, which must be based on the ability of ordinary technicians in the field to implement. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0078] An embodiment of the present application provides a battery casing torque testing method. Figure 1 A flow chart of a battery housing torque testing method provided in one embodiment of the present application. In this embodiment, the method includes:

[0079] See also Figure 1 , obtaining stress-strain relationship data of the battery shell under an initial load state, and obtaining structural microstrain monitoring data of the battery shell under the initial load state, and generating initial self-repair monitoring benchmark data of the battery shell according to the stress-strain relationship data and the structural microstrain monitoring data, wherein the initial load causes the battery shell to be in an elastic deformation stage;

[0080] In one embodiment of the present invention, the stress-strain relationship data of the battery shell under the initial load state is obtained, and the structural microstrain monitoring data of the battery shell under the initial load state is obtained, and the initial self-repair monitoring benchmark data of the battery shell is generated according to the stress-strain relationship data and the structural microstrain monitoring data, including: applying a torque loading sequence to the battery shell, and obtaining multi-level stress-strain relationship data of the battery shell under the action of the torque loading sequence, wherein the torque loading sequence includes an initial loading segment and an incremental loading segment, the loading value of the initial loading segment is greater than the torque value of the elastic deformation starting point of the battery shell and less than the yield limit torque value of the battery shell, the starting loading value of the incremental loading segment is the loading value of the initial loading segment, and the loading process of the incremental loading segment is divided into a plurality of loading sub-segments with equal time intervals, and the torque value of each loading sub-segment increases by a preset torque value compared with the previous loading sub-segment. moment increment, the termination loading value of the incremental loading section does not exceed the yield limit torque value; the structural microstrain monitoring data of the battery shell is obtained through a preset acquisition point array on the surface of the battery shell, and the structural microstrain monitoring data includes the initial structural microstrain distribution data under the loading value of the initial loading section and the dynamic structural microstrain distribution data under the loading values ​​of each level of the incremental loading section; during the holding process of each loading value of the initial loading section and the incremental loading section, the corresponding structural microstrain distribution data is regionally scanned to obtain the repair agent trigger signal data of each area of ​​the battery shell; the multi-level stress-strain relationship data and the repair agent trigger signal data are time-series correlated and analyzed to generate stress fluctuation response characteristic data of the battery shell; the self-repair sensitivity parameter of the battery shell is judged according to the stress fluctuation response characteristic data, and the initial self-repair monitoring benchmark data of the battery shell is generated.

[0081] Specifically, first of all, the implementation of the initial loading section is to apply a load to the battery shell, and its loading value must be greater than the torque value at the starting point of elastic deformation of the battery shell and less than the yield limit torque value. The torque value at the starting point of elastic deformation corresponds to the lowest stress point at which the material just begins to undergo elastic deformation, and the yield limit torque value is the maximum load-bearing capacity of the material before entering plastic deformation. At this stage, the loading value is precisely set to the elastic interval to ensure the linear reliability of the data. In actual implementation, the torque is increased at a constant rate by the loading device, and the stress and strain data of the shell are recorded using a high-precision sensor. For example, taking the elastic deformation starting point of the battery shell as 10 Nm and the yield limit as 50 Nm as an example, the loading value of the initial loading section can be gradually increased from 12 Nm to 45 Nm. The key to this stage is to prevent the material from entering a plastic deformation state, thereby ensuring that the stress-strain relationship has linear characteristics and providing a scientific basis for subsequent data analysis.

[0082] In the incremental loading section, the loading value starts from the end value of the initial loading section, and increases according to the set torque increment to below the yield limit torque value, and the incremental process is divided into multiple loading sub-segments with equal time intervals. The torque increment of each sub-segment is predetermined to achieve precise control of the loading process. For example, when the initial loading section ends at 45 Nm, the incremental loading section can start from 45 Nm, and each sub-segment increases by 2 Nm until the yield limit of 50 Nm. This segmented loading method prevents excessive stress concentration caused by one-time loading while gradually approaching the material limit, ensuring the stability and safety of the test. The segmented design also enables each level of loading value to be maintained for a certain duration for accurate data collection.

[0083] During the loading process, the structural microstrain monitoring data is obtained through an array of preset collection points arranged on the surface of the battery shell. The array of collection points is usually evenly distributed on the shell surface, and each collection point records the strain changes caused by loading in real time through a highly sensitive displacement sensor or strain gauge. The strain data of the initial loading section is used to generate the initial structural microstrain distribution, while the strain data of the incremental loading section corresponds to the dynamic structural microstrain distribution. For example, in actual testing, 25 collection points can be arranged on the shell surface, and each collection point will record the corresponding strain value during the loading stage. These data are processed to form a two-dimensional or three-dimensional strain distribution diagram, which fully reflects the changes in the strain on the shell surface.

[0084] In order to further analyze the repair process, the structural microstrain distribution data of the shell is regionally scanned during the initial loading section and each level of the incremental loading section, and the repair agent trigger signal data is obtained at the same time. The repair agent trigger signal data is the change in optical signal, acoustic signal or electrical signal caused by crack filling or other chemical reactions after the repair agent is released. At each level of loading value, the location of the crack and the trigger time of the repair agent can be located by scanning the array of acquisition points. For example, under the loading condition of 45 Nm, if a crack is detected at a certain acquisition point and accompanied by an increase in the optical signal, it means that the repair agent has been released and has begun to fill the crack. These data are further associated with the loading value to provide comprehensive information on crack triggering and repair behavior.

[0085] Subsequently, the multi-level stress-strain relationship data and the repair agent trigger signal data are subjected to time series correlation analysis to generate stress fluctuation response characteristic data. This correlation analysis is based on data pairing on the time axis, matching the stress fluctuation caused by the loading value with the time series of the repair agent trigger signal, and extracting the time characteristics of crack formation, repair agent triggering and subsequent repair behavior. For example, under a certain loading value, the crack formation time is 2.8 seconds, while the repair agent trigger time is 3.2 seconds, with a time difference of 0.4 seconds. By counting the time differences of multiple loading stages, the stress fluctuation characteristics of the shell, as well as the response speed and effectiveness of the repair agent can be analyzed.

[0086] Finally, by analyzing the stress fluctuation response characteristic data, the self-repair sensitivity parameters of the battery shell are determined, and the initial self-repair monitoring benchmark data is generated. The self-repair sensitivity parameters integrate indicators such as crack formation time, repair agent trigger time, and crack filling efficiency, and are used to quantify the reaction speed and repair ability of the shell self-repair system. For example, if the time difference is small and the crack can be completely filled in a short time, the sensitivity parameter is high; if the time difference is large or the repair effect is incomplete, the sensitivity parameter is low. Based on the statistics and analysis of multiple sets of experimental data, the generated benchmark data fully reflects the self-repair performance of the shell under initial loading conditions.

[0087] In summary, by scientifically setting the torque loading sequence, accurately collecting microstrain data and repair agent trigger signals, and combining time series correlation analysis to generate characteristic data, this method achieves comprehensive quantification and benchmark data generation of the initial self-repairing ability of the battery shell. This process not only collects comprehensive data with high resolution, but also provides clear and verifiable data analysis, which can provide a reliable scientific basis for subsequent tests.

[0088] In one embodiment of the present invention, the multi-level stress-strain relationship data and the repair agent trigger signal data are subjected to time-series correlation analysis to generate stress fluctuation response characteristic data of the battery shell, including: grouping the multi-level stress-strain relationship data according to the time series of applied load, extracting strain mutation points and strain recovery points in each group of data, and generating key stress response moment data of the battery shell; performing time domain analysis on the repair agent trigger signal data, recording the initial activation moment and complete diffusion moment of the repair agent in each torque loading sub-segment, pairing and marking the repair agent trigger signal data based on the key stress response moment data, and generating repair response synchronization data of the battery shell; calculating the stress recovery coefficient and the repair agent diffusion coefficient of the battery shell in different torque loading sub-segments according to the repair response synchronization data, and generating stress fluctuation response characteristic data of the battery shell.

[0089] Specifically, first, the multi-level stress-strain relationship data is grouped by the time series of applied loads. This operation aims to logically classify the data according to the loading order on the time axis in order to capture the stress response characteristics of each loading sub-segment. Specifically, each loading sub-segment corresponds to a time interval, the starting point of which is the start time of loading and the end point is the end time of loading. The data grouping operation can be performed by synchronizing the loading record timestamp with the strain measurement data, and each group of data contains all stress and strain records within the time interval. For example, in a certain loading stage (such as a torque value of 40 Nm), the time interval from the start of loading to the completion of loading is set to 2 seconds, then all strain data within this time period will be classified into the same group. This grouping method ensures the temporal continuity of the data and lays the foundation for the subsequent extraction of strain mutation points and strain recovery points.

[0090] In each group of data after grouping, extracting strain mutation points and strain recovery points is a key step. The strain mutation point refers to the time node where a sudden and significant change occurs in the strain data, while the strain recovery point refers to the time node where the strain gradually returns to a stable state after a sudden change. In order to extract these characteristic points, an algorithm based on the rate of change can be used to analyze the strain data curve. The specific method includes calculating the slope of the data curve. When the slope exceeds the set mutation threshold, it is marked as a strain mutation point; when the slope gradually falls back to zero or approaches a stable threshold, it is marked as a strain recovery point. For example, when loaded to 45 Nm, it is observed that the strain suddenly increases from 0.2 millistrain to 0.8 millistrain, and the corresponding time point is the mutation point; after this, the strain value gradually falls back to 0.4 millistrain and stabilizes, and this moment is the recovery point. In this way, the key stress response moment data of the battery shell can be generated, which reflects the time characteristics of significant changes in stress distribution during loading.

[0091] Next, the repair agent trigger signal data is analyzed in the time domain to record the initial activation moment and complete diffusion moment of the repair agent. The purpose of this step is to capture the entire process from the release of the repair agent to the complete coverage of the crack. The trigger signal data of the repair agent may include changes in optical signals, acoustic signals, or conductivity signals. For example, when a crack appears, the release of the repair agent may cause an increase in the surface light reflectivity. This signal change can be monitored by optical detection equipment to mark the initial activation moment of the repair agent. At the same time, when the signal intensity reaches a stable state (that is, it no longer increases significantly), it can be marked as the moment of complete diffusion. Assuming that in a certain loading sub-segment, the crack occurrence time is 3 seconds, the signal enhancement point when the repair agent begins to release is 3.2 seconds, and the complete diffusion point is 4.5 seconds, then the initial activation time is recorded as 3.2 seconds, and the complete diffusion time is recorded as 4.5 seconds.

[0092] Based on the key stress response moment data, the repair agent trigger signal data is paired and marked. This step associates two independent data sets to analyze the time relationship between the stress mutation caused by loading and the release behavior of the repair agent. In actual operation, this can be achieved through a pairing algorithm on the time axis, comparing the key stress response moment (such as the strain mutation point) with the initial activation moment of the repair agent to find the closest time match. For example, in a certain loading stage, the strain mutation point is recorded as 3 seconds, and the initial activation moment of the repair agent is recorded as 3.2 seconds. The time difference between the two is 0.2 seconds. This time difference can reflect the response speed of the repair agent to the crack. Through this pairing mark, the generated repair response synchronization data contains the time series relationship between the stress response caused by each load and the release of the repair agent.

[0093] According to the repair response synchronization data, the stress recovery coefficient and the diffusion coefficient of the repair agent of the battery shell in different torque loading sub-segments are calculated. The stress recovery coefficient is a quantitative indicator of the stress recovery after loading, which can be calculated by the ratio of the strain difference and the time difference between the mutation point and the recovery point. For example, in a certain loading sub-segment, the strain recovers from 0.8 millistrain at the mutation point to 0.4 millistrain at the recovery point, and the time required is 2 seconds. The stress recovery coefficient is (0.8-0.4) / 2=0.2 millistrain / second. The diffusion coefficient of the repair agent reflects the efficiency of the repair agent in covering the crack, which can be calculated by the time difference between the time when the repair agent is completely diffused and the time when it is initially activated, and the ratio of the crack coverage area. For example, under the condition that the crack area is 2 square millimeters, if the time when the repair agent is completely diffused is 4.5 seconds and the initial activation time is 3.2 seconds, the diffusion coefficient of the repair agent is 2 square millimeters / (4.5-3.2 seconds)=1.54 square millimeters / second.

[0094] Finally, by integrating the stress recovery coefficient and the diffusion coefficient of the repair agent, the stress fluctuation response characteristic data of the battery shell is generated. This characteristic data reflects the stress recovery ability and repair efficiency of the battery shell under different loading conditions, and is a comprehensive quantification of the mechanical properties and self-healing characteristics of the shell. Through this process, the dynamic response behavior of the battery shell after loading-induced cracks can be fully captured, providing an important scientific basis for subsequent performance evaluation and optimization. Each step of the entire process has a clear operating method and analysis logic to ensure the specificity and operability of the technical implementation.

[0095] Please continue reading Figure 1 , performing progressive torsion loading on the battery housing according to the initial self-repair monitoring benchmark data, obtaining multi-band optical deformation data and microstructure stress response data of the battery housing during the progressive torsion loading process, and generating crack evolution characteristic data of the battery housing according to the time correlation between the multi-band optical deformation data and the microstructure stress response data;

[0096] In one embodiment of the present invention, the battery shell is progressively torque loaded according to the initial self-repair monitoring benchmark data, multi-band optical deformation data and microstructure stress response data of the battery shell during the progressive torque loading process are obtained, and crack evolution characteristic data of the battery shell is generated according to the time correlation between the multi-band optical deformation data and the microstructure stress response data, including: regional division of the initial self-repair monitoring benchmark data to generate a surface stress distribution map and an internal repair agent distribution map of the battery shell; a progressive torque loading scheme is determined according to the surface stress distribution map and the internal repair agent distribution map, wherein the progressive torque loading scheme divides the progressive torque loading into multiple loading cycles, the holding time of each loading cycle is proportional to the maximum torque value in the corresponding loading cycle, the torque value in each loading cycle increases step by step according to a preset interval, and there is an overlapping segment of the torque values ​​between adjacent loading cycles; in each loading cycle , the surface deformation image of the battery shell is collected by ultraviolet, visible light and near infrared three frequency bands, the strain field data of the battery shell in different frequency bands are obtained, and the strain field data of the battery shell in the three frequency bands of ultraviolet, visible light and near infrared are compared and synthesized to generate multi-band optical deformation data; in the overlapping section of each loading cycle, the conductive network signal data and the temperature gradient distribution data inside the battery shell are obtained, and the microstructure stress response data is generated according to the conductive network signal data and the temperature gradient distribution data; the multi-band optical deformation data and the microstructure stress response data are paired according to the timestamp to generate the crack characterization data of the battery shell in each loading cycle, and the crack characterization data includes surface crack morphology data and internal crack extension data; the crack characterization data in each loading cycle is subjected to time series superposition analysis to obtain the evolution correlation data of the crack between different loading cycles, and the crack evolution characteristic data of the battery shell is generated.

[0097] Specifically, first, the initial self-repair monitoring benchmark data is divided into regions. This step aims to clarify the stress distribution characteristics of the battery shell surface and the distribution law of the internal repair agent. The regional division is based on the geometric structure, material properties and acquisition point array data of the battery shell. Through the partition analysis of the monitoring benchmark data, the shell surface stress distribution map and the internal repair agent distribution map can be generated. The stress distribution map is a spatial mapping of the stress concentration in each area of ​​the surface under loading, which is usually generated by finite element analysis of the collected microstrain data and the mechanical model of the shell structure. For example, on a battery shell, it can be observed that torque loading causes stress concentration around the screw fixing position, which is manifested as a high stress area on the stress distribution map. The internal repair agent distribution map characterizes the pre-filled distribution of the repair agent in the shell, which can be detected by the optical signal (such as near-infrared transmission characteristics) or electrical signal changes of the material, thereby mapping the spatial distribution of the repair agent inside the shell.

[0098] Next, the progressive torque loading scheme is determined based on the generated surface stress distribution map and internal repair agent distribution map. The progressive torque loading scheme is designed as multiple loading cycles. The holding time of each loading cycle is proportional to the corresponding maximum torque value, and the loading value increases step by step at preset intervals. The setting of the holding time ensures that each level of loading stress has a full effect on the shell, and the increasing mode of the preset interval ensures the gradualness and controllability of the loading process. For example, assuming that the maximum torque value in a loading cycle is 50 Nm, the holding time can be set to 5 seconds, and the loading interval is 2 Nm. The loading values ​​of the loading cycle are 48 Nm and 50 Nm respectively. The overlapping section between adjacent loading cycles is set to 2 Nm to ensure the continuity of the loading process and avoid the impact of sudden stress interruption on the shell.

[0099] In each loading cycle, the surface deformation images of the battery shell are collected through three frequency bands: ultraviolet, visible light, and near-infrared, to obtain its strain field data in different frequency bands. The core of this step is to use the optical properties of different frequency bands to capture the dynamic changes of the shell surface cracks in multiple dimensions. The ultraviolet band is highly sensitive to the initial formation of microcracks on the shell surface, the visible light band has good analytical ability for the macroscopic changes of surface deformation, and the near-infrared band can penetrate materials to a certain depth and reflect the expansion of cracks under the shell surface. For example, in a certain loading stage, a slight increase in the crack width can be observed through the ultraviolet band, the direction of the crack extension can be determined through the visible light band, and the depth change of the crack can be detected through the near-infrared band. After comparing and synthesizing the data of these three bands, multi-band optical deformation data can be generated to form a comprehensive description of the crack deformation characteristics.

[0100] In the overlapping section of the loading cycle, the conductive network signal data and temperature gradient distribution data inside the battery shell are obtained to generate microstructure stress response data. The conductive network signal data can reflect the damage of the internal cracks of the shell to the conductive path, while the temperature gradient distribution data can reveal the changes in heat conduction around the cracks. For example, when loading to a certain overlapping section, the attenuation of the conductive signal can indicate that the internal cracks have affected the conductivity of the shell, and the abnormal changes in the temperature gradient can reflect the stress concentration and local deformation of the material around the crack. Comprehensive analysis of these data can generate stress response data reflecting the microstructure state inside the shell. Then, the multi-band optical deformation data and the microstructure stress response data are paired according to the timestamp to generate crack characterization data. The crack characterization data includes surface crack morphology data and internal crack extension data. Through synchronous pairing on the time axis, the behavior of surface cracks and internal cracks can be associated. For example, in a certain loading cycle, the expansion rate of the surface crack is 0.2 mm per second, while the corresponding internal crack depth increases to 0.1 mm per second, which indicates that the expansion of the crack has the synergistic characteristics of the surface and the interior. Through such paired analysis, the dynamic behavior of the crack in each loading cycle can be accurately described.

[0101] Finally, the crack characterization data within each loading cycle is subjected to time-series superposition analysis to obtain the evolution correlation data of the crack between different loading cycles. Time-series superposition analysis reveals the overall evolution law of the crack by accumulating the crack extension characteristics of each cycle. For example, the length of the surface crack gradually increases from 0.5 mm in the first loading cycle to 1.5 mm in the third loading cycle, while the depth of the internal crack increases from 0.2 mm to 0.8 mm. This time-series superposition can reveal the expansion trend and stage characteristics of the crack, and the generated crack evolution characteristic data summarizes the entire process from crack formation to expansion.

[0102] Please continue reading Figure 1 , determining a torque loading gradient according to the crack evolution characteristic data, and applying the torque loading gradient to the battery housing, acquiring high-frequency strain data, real-time optical monitoring data, and conductive network state data of the battery housing under different torque gradients, and generating repair process characterization data of the battery housing according to the high-frequency strain data, the real-time optical monitoring data, and the conductive network state data;

[0103] In one embodiment of the present invention, the torque loading gradient is determined according to the crack evolution characteristic data, and the torque loading gradient is applied to the battery shell, and the high-frequency strain data, real-time optical monitoring data and conductive network state data of the battery shell under different torque gradients are obtained, and the repair process characterization data of the battery shell is generated according to the high-frequency strain data, the real-time optical monitoring data and the conductive network state data, including: grading the crack evolution characteristic data according to the crack depth and propagation speed to generate crack grading data, determining the grading interval and torque load sequence of the torque loading gradient according to the crack grading data, and the torque load sequence includes a constant torque load and a pulse vibration load; applying the constant torque load and the pulse vibration load to the battery shell in sequence, wherein the constant torque load is applied at a preset rate. The pulse vibration load is increased in a step-by-step manner, and a vibration stress of a preset amplitude is superimposed on the current constant torque load, and high-frequency strain data is acquired by a dual time-scale sampling method of millisecond and sub-millisecond levels; during the loading process of the constant torque load and the pulse vibration load, synchronous data acquisition is performed on the battery shell to acquire surface strain transient distribution data and internal conductivity fluctuation data, and to generate real-time optical monitoring data and conductive network status data; the high-frequency strain data, the real-time optical monitoring data and the conductive network status data are time-synchronously paired to generate repair behavior characteristic data of the battery shell under the action of the torque loading gradient; according to the repair behavior characteristic data, the strain recovery rate parameters, the repair agent flow characteristic parameters and the conductive path reconstruction parameters of the battery shell are extracted to generate repair process characterization data of the battery shell.

[0104] Specifically, first, the crack evolution characteristic data is classified according to the crack depth and expansion rate. This step generates crack classification data by analyzing the changing characteristics of surface and internal data during crack expansion. The crack depth characterizes the extent to which the crack develops into the shell, while the crack expansion rate reflects the dynamic change trend of the crack under stress conditions. The classification of these data is completed through preset classification standards. For example, the crack depth can be divided into shallow layer (<0.5 mm), middle layer (0.5-1 mm) and deep layer (>1 mm), while the expansion rate can be divided into slow expansion (<0.1 mm / s), medium expansion (0.1-0.5 mm / s) and fast expansion (>0.5 mm / s). In specific implementation, these indicators are classified by the time series data of crack expansion. For example, when the depth of a crack expansion reaches 0.8 mm and the speed is 0.3 mm / s, it can be classified as middle layer and medium expansion level. This classification method structures the complex crack behavior data, making the subsequent loading design more accurate.

[0105] Based on the crack classification data, the classification interval of the torque loading gradient and the torque load sequence are determined. The classification interval of the torque loading gradient directly corresponds to the depth and extension speed level in the crack classification data. For example, the loading interval for shallow cracks may be set to 20-30 Nm, 30-40 Nm for medium cracks, and 40-50 Nm for deep cracks. In each interval, the torque load sequence consists of a constant torque load and a pulse vibration load, where the constant torque load increases stepwise at a preset rate to ensure the stability of the loading process; the pulse vibration load superimposes a vibration stress of a specific amplitude on the current constant torque to induce further crack expansion and stimulate the flow of the repair agent. For example, when loaded to 40 Nm, a pulse vibration load with an amplitude of ±2 Nm can be applied to simulate the effect of dynamic load on crack extension. The design of this loading sequence can not only capture the behavior of cracks under complex loading conditions, but also verify the response efficiency of the repair agent in a dynamic stress environment.

[0106] During the loading process, high-frequency strain data are obtained by dual time scale sampling of milliseconds and sub-milliseconds. Millisecond sampling is used to capture the overall trend of macroscopic strain changes during loading, while sub-millisecond sampling is used to identify transient strain mutations caused by crack extension in a short period of time. For example, in a certain loading stage, millisecond data can show that the strain gradually increases from 0.3 millistrain to 0.8 millistrain, while sub-millisecond data can reveal the instantaneous jump phenomenon of strain at a certain moment. Through the dual time scale sampling method, high-precision and comprehensive strain data can be obtained, providing rich information for subsequent dynamic behavior analysis.

[0107] At the same time, during the loading process, synchronous data collection is performed on the battery shell to obtain surface strain transient distribution data and internal conductivity fluctuation data. These data reflect the impact of crack extension on the shell surface deformation and internal conductivity, respectively. The surface strain transient distribution data records the deformation characteristics of the shell surface in real time through a high-precision displacement sensor. For example, during crack extension, the strain value of a certain area on the surface may increase rapidly from 0.5 millistrain to 1.2 millistrain, showing the characteristics of local stress concentration. The internal conductivity fluctuation data is collected through the conductive network detection module to evaluate the degree of damage to the conductive path inside the shell. For example, when the crack extends to a certain depth, the conductivity may drop by 10%. This data reflects the damage to the integrity of the internal structure. These data are further paired through time synchronization to generate real-time optical monitoring data and conductive network status data, providing multi-dimensional information for the comprehensive characterization of crack behavior.

[0108] Next, the high-frequency strain data, real-time optical monitoring data, and conductive network status data are synchronously paired according to timestamps to generate repair behavior feature data. Timestamp pairing ensures that multi-source data can achieve consistency in the time dimension by comparing the recording times of different data sources. For example, in a certain loading stage, the strain data shows that crack propagation occurs 5 seconds after loading, while the optical monitoring data and conductive network status data also show corresponding changes at the same time point. These data are marked as the same time event. Through time synchronization pairing, data from different detection modules can be integrated to generate comprehensive features that describe the crack propagation and repair agent release process.

[0109] Finally, the strain recovery rate parameters, flow characteristic parameters of the repair agent and the conductive path reconstruction parameters of the battery shell are extracted according to the repair behavior characteristic data to generate the repair process characterization data. The strain recovery rate parameter is calculated by comparing the strain difference and time interval before and after the crack extension. For example, when the crack extension triggers the strain to recover from 0.8 millistrain to 0.4 millistrain, the time required is 2 seconds, and the recovery rate is 0.2 millistrain / second. The flow characteristic parameters of the repair agent are calculated by the time interval from the triggering of the repair agent to the complete coverage of the crack and the flow path length. For example, the time from the release of the repair agent to the complete filling of the crack is 3 seconds, and the crack length is 1.5 mm, then the flow characteristic parameter is 0.5 mm / second. The conductive path reconstruction parameters are calculated by the time characteristics of the conductivity from the decrease to the recovery. For example, the conductivity recovers to 90% after the crack is completely repaired. This recovery time is combined with the repair behavior characteristic data to characterize the integrity of the conductive network. The repair process characterization data generated by the integration of these parameters fully reflects the dynamic repair performance of the battery shell under loading conditions, providing a scientific basis for subsequent performance optimization and safety assessment.

[0110] Please continue reading Figure 1 , performing segmented comparison on a preset area of ​​the battery housing under multiple torque gradients according to the repair process characterization data, obtaining time interval data between a crack formation moment and a repair agent activation moment in the preset area, and harmonic acoustic characteristic data of the preset area, and generating repair efficiency evaluation data of the preset area according to the time interval data and the harmonic acoustic characteristic data;

[0111] In one embodiment of the present invention, the preset area of ​​the battery shell is segmented and compared under multiple torque gradients according to the repair process characterization data, and the time interval data between the crack formation time and the repair agent activation time of the preset area and the harmonic acoustic characteristic data of the preset area are obtained. According to the time interval data and the harmonic acoustic characteristic data, the repair efficiency evaluation data of the preset area is generated, including: segmenting the repair process characterization data according to the torque gradient, dividing the high-incidence crack area of ​​the battery shell, and generating segmented scanning data of the preset area; according to the segmented scanning data of the preset area, recording the strain mutation point and the conductive path recovery time of the battery shell in each torque gradient interval. The method comprises the following steps: rechecking the time points, generating time series evolution data of a preset area, extracting the time interval data between the crack formation moment and the repair agent activation moment from the time series evolution data; collecting the acoustic response data of the repair process for the crack formation moment and the repair agent activation moment in the time series evolution data at the key monitoring positions determined by the segmented scanning data of the preset area, and generating the harmonic acoustic characteristic data of the preset area; cross-validating the time interval data and the harmonic acoustic characteristic data, and generating the repair quality evaluation data of the preset area under each torque gradient; determining the repair depth parameter and the repair speed parameter of the preset area according to the repair quality evaluation data, and generating the repair efficiency evaluation data of the preset area.

[0112] Specifically, when the repair process characterization data is segmented according to the torque gradient, the operation process and segmentation criteria need to be clarified. The torque gradient is divided into multiple intervals, such as 20-30 Nm, 30-40 Nm, 40-50 Nm, etc. The analysis object in each interval is the dynamic behavior characteristics of the cracks within the gradient range. The division of high-incidence areas of cracks is completed by combining finite element analysis and measured data. The surface of the battery shell is gridded, and the strain concentration of each grid unit is quantified as a strain gradient value. If the strain gradient value of a certain area exceeds a preset threshold (such as a 5% strain change rate), the area is marked as a high-incidence area of ​​cracks. For example, in the screw fixing hole area of ​​the battery shell, the stress concentration caused by the structural characteristics is usually manifested as a higher strain change rate, and the area will be preferentially marked as a high-incidence area of ​​cracks. Through this clear quantification standard and grid division method, segmented scanning data of preset areas containing high-incidence area information can be generated.

[0113] Next, in each torque gradient interval, the data is scanned in segments in the preset area to record the strain mutation point and the conductive path recovery point of the battery shell. The recording of the strain mutation point is completed by real-time monitoring of the rate of change of the strain data. When the rate of change exceeds the preset mutation threshold (such as 10 millistrain / second), the time point is automatically marked as the strain mutation point. For example, in the torque gradient interval of 30-40 Nm, the strain value of a crack area increases rapidly from 0.3 millistrain to 0.8 millistrain, and the corresponding time point is recorded as the mutation point. The conductive path recovery point is achieved through real-time conductivity monitoring. When the conductivity gradually recovers from a declining state to more than 90% and tends to stabilize, the moment is recorded as the conductive path recovery point. For example, when the conductivity gradually recovers from 70% to 90% during the crack repair process, and remains stable at this level for more than 1 second, this moment is the recovery point. These records lay a specific operational basis for generating time-series evolution data for preset areas.

[0114] When extracting the time interval data between the crack formation moment and the repair agent activation moment in the time series evolution data, the time axis comparison method is used to ensure the accuracy of the extraction. Specifically, by generating the timestamp curves of crack formation and repair agent activation in the time series data, the time difference between the two is automatically calculated using the timestamp comparison algorithm. For example, in crack area A, the timestamp of crack formation is 2.8 seconds, the timestamp of repair agent activation is 3.2 seconds, and the time interval data is 0.4 seconds. This timestamp-based algorithm ensures the efficiency and accuracy of the calculation, while avoiding the errors that may be caused by manual extraction.

[0115] At the key monitoring positions determined by the segmented scanning data in the preset area, the acoustic response data of the repair process is collected for the crack formation moment and the repair agent activation moment in the time series evolution data. The collection of acoustic response data relies on high-sensitivity acoustic sensors, which are arranged around the areas where cracks are prone to occur. The sensors monitor the acoustic wave signals generated during the repair process in real time, and remove environmental noise through filtering algorithms to retain characteristic signals related to the repair process. For example, when the repair agent is released from the storage position to the crack and begins to fill, an acoustic wave signal with a frequency of 20 kHz may be generated, and the intensity and frequency of these signals will change as the repair progresses. When the crack is completely filled and solidified, the amplitude of the acoustic wave signal may tend to be stable, indicating that the repair process has been completed. After processing, these acoustic data generate harmonic acoustic characteristic data of the preset area, reflecting the law of acoustic changes during the repair process.

[0116] The time interval data and harmonic acoustic feature data are cross-validated, and the accuracy and reliability of the data are ensured through data consistency analysis. The core of cross-validation is to compare whether the time correlation between the two types of data is consistent during the crack formation and repair process. For example, the time interval data shows that the repair agent begins to activate 0.4 seconds after the crack is formed, and the acoustic feature data also detects the first appearance of the acoustic wave signal at 0.4 seconds, and the two are highly consistent. This verification method can eliminate abnormal data points and provide a basis for generating high-quality repair evaluation data.

[0117] According to the repair quality evaluation data, the repair depth parameters and repair speed parameters of the preset area are calculated to generate the repair efficiency evaluation data. The repair depth parameter is calculated by the change in crack depth. For example, if the crack is reduced from an initial depth of 1 mm to 0.2 mm, the repair depth parameter is 0.8 mm. The repair speed parameter is calculated by the time interval from the formation of the crack to its complete filling. For example, if the crack is completely repaired from 1 mm in 8 seconds, the repair speed parameter is 0.125 mm / s. These data comprehensively reflect the efficiency and effect of the crack repair process. The repair efficiency evaluation data further integrates these parameters to quantify the repair capacity and behavior characteristics of the preset area.

[0118] Please continue reading Figure 1 , analyzing and processing the repair efficiency evaluation data, generating a dynamic safety threshold curve of the battery housing, and determining a torque limit range of the battery housing according to the dynamic safety threshold curve;

[0119] In one embodiment of the present invention, the analysis and processing of the repair efficiency evaluation data to generate a dynamic safety threshold curve of the battery shell, and determining the torque limit range of the battery shell according to the dynamic safety threshold curve, includes: performing torque gradient correlation analysis on the repair efficiency evaluation data, extracting repair depth data and repair speed data of the battery shell under different torque gradients, and generating repair performance grading data of the battery shell; calibrating the repair saturation degree of the battery shell under each torque gradient according to the repair performance grading data, obtaining repair agent consumption rate data and repair layer stability data, and generating repair limit state data of the battery shell; grouping the repair limit state data according to safety margin requirements, generating safety torque benchmark data of the battery shell in three repair efficiency intervals of high, medium and low, and drawing the dynamic safety threshold curve of the battery shell according to the safety torque benchmark data; obtaining critical torque data of the battery shell in different repair efficiency intervals based on the dynamic safety threshold curve, and performing correction coefficient compensation on the critical torque data to generate safety margin correction data of the battery shell; adjusting the warning point position of the dynamic safety threshold curve according to the safety margin correction data to determine the torque limit range of the battery shell.

[0120] Specifically, the core of the correlation analysis is to match the repair depth data and repair speed data in the repair efficiency evaluation data with the loading conditions corresponding to different torque gradient intervals, so as to extract the repair characteristics under each torque gradient. The repair depth data describes the change in the depth of the crack after loading. For example, under a torque gradient of 20-30 Nm, the crack depth is repaired from 1.0 mm to 0.2 mm, and the repair depth is 0.8 mm. The repair speed data quantifies the time required for the crack to be completely repaired from formation to complete repair. For example, under the same torque gradient, the crack repair time is 5 seconds, and the repair speed is 0.16 mm / s. These data are formed through correlation analysis to form the repair performance grading data of the battery shell, and the repair performance under different torque gradients is divided into three levels: high performance, medium performance and low performance. For example, when the repair depth is greater than 0.7 mm and the repair speed exceeds 0.1 mm / s, it can be defined as high performance; otherwise, it is classified as low performance.

[0121] Next, according to the repair performance grading data, the repair saturation degree of the battery shell at each torque gradient is calibrated, and the repair agent consumption rate data and the repair layer stability data are further obtained. The calibration of the repair saturation degree is completed by calculating the ratio of the actual consumption of the repair agent to the total volume of the crack when the crack is completely repaired. For example, under a torque gradient of 30-40 Nm, the repair agent consumption is 1.5 ml and the crack volume is 2.0 ml, then the repair saturation degree is 75%. The repair agent consumption rate is calculated by the ratio of the total consumption of the repair agent to the repair time. For example, if 1.5 ml of the repair agent is consumed within 5 seconds, the consumption rate is 0.3 ml / s. The repair layer stability data is generated by testing the mechanical properties of the crack area after the repair is completed, such as determining the deformation and fracture strength of the repair layer under the maximum bearing capacity through a loading test, thereby evaluating its stability.

[0122] Subsequently, the repair limit state data is generated based on the repair agent consumption rate data and the repair layer stability data. The repair limit state data reflects the maximum load state that the battery shell can withstand after the repair is completed and the durability of the repair agent. Specifically, when the deformation of the repair layer is less than 10% under a load of 40 Nm and the fracture strength reaches more than 90% of the initial material strength, the repair layer is considered stable; if the repair agent consumption rate exceeds the preset threshold (such as 0.5 ml / s), there may be a problem of insufficient supply of repair agent. These limit state data provide a scientific basis for subsequent safety assessments.

[0123] According to the repair limit state data, the repair performance is divided into three efficiency intervals of high, medium and low according to the safety margin requirements, and the safety torque benchmark data is generated. The safety margin refers to the safety range reserved before the load limit. Its definition depends on the specific loading conditions and repair performance. For example, the high efficiency interval is defined as the torque range where the repair depth exceeds 0.8 mm and the repair speed exceeds 0.12 mm / s; the medium efficiency interval is defined as the repair depth between 0.5-0.8 mm and the repair speed between 0.08-0.12 mm / s; the low efficiency interval is the rest. By analyzing the repair performance of each efficiency interval, the corresponding safety torque benchmark data can be generated. For example, in the high efficiency interval, the safety torque range is 20-30 Nm; the medium efficiency interval is 30-40 Nm; and the low efficiency interval is 40-50 Nm.

[0124] Next, the dynamic safety threshold curve of the battery case is drawn based on the safety torque benchmark data. This curve uses torque as the horizontal axis and repair efficiency as the vertical axis, showing the safety range of the battery case under different loading conditions. For example, the dynamic safety threshold curve may show that the repair efficiency remains stable in the range of 20-30 Nm, but the repair efficiency drops rapidly above 40 Nm. This dynamic curve can provide a visual safety reference for practical applications, allowing engineers to intuitively judge the safety of the battery case under various working conditions.

[0125] Based on the dynamic safety threshold curve, the critical torque data of the battery casing in different repair efficiency ranges are further obtained, and these critical torque data are compensated by correction coefficients. The critical torque data refers to the torque value at which the repair efficiency is converted between high, medium and low efficiency ranges. For example, the critical torque is 30 Nm when converting from high efficiency to medium efficiency, and 40 Nm when converting from medium efficiency to low efficiency. The correction coefficient compensation takes into account the impact of actual factors such as material aging and ambient temperature changes on the repair performance, and adjusts the critical torque value to reflect these factors. For example, in a high temperature environment, the correction coefficient is 0.9, the original critical torque is 30 Nm, and the compensated critical torque is 30×0.9=27 Nm. This process ensures the applicability and accuracy of critical data.

[0126] Finally, according to the safety margin correction data, the warning point position of the dynamic safety threshold curve is adjusted, and the torque limit range of the battery casing is determined. The warning point refers to the torque value when the repair efficiency drops to a specific threshold. For example, when the repair efficiency drops to 50%, the warning point can be set to 35 Nm. Adjusting the position of the warning point can optimize the limit range according to the actual working conditions, thereby ensuring the safety of the battery casing in various usage scenarios. The final torque limit range is 20-40 Nm, which not only reflects the repair performance of the battery casing, but also takes into account a variety of safety factors in actual applications.

[0127] Please continue reading Figure 1 According to the torque limit range and the dynamic safety threshold curve, the operating state of the battery housing is determined in real time to generate a safety assessment result of the battery housing.

[0128] In one embodiment of the present invention, the operating state of the battery housing is determined in real time based on the torque limit range and the dynamic safety threshold curve to generate a safety assessment result of the battery housing, including: dividing the torque limit range and the dynamic safety threshold curve into regions, dividing the operating state of the battery housing into a normal operating range, a warning monitoring range and a risk control range, and generating operating state partition data of the battery housing; determining the monitoring parameter thresholds of the battery housing in different operating ranges according to the operating state partition data, and performing torque gradient calibration on the monitoring parameter thresholds to generate a state monitoring judgment data of the battery housing. According to the status monitoring criterion data, the strain response characteristics of the battery shell and the activity state of the repair agent are analyzed, and the safety risk level data of the battery shell is generated, and the safety risk level data includes the structural integrity data of the battery shell, the remaining amount data of the repair agent and the re-repairability data; based on the safety risk level data, the maintenance cycle index of the battery shell is established, and the maintenance cycle index is cross-validated with the dynamic safety threshold curve to generate the safety margin evaluation data of the battery shell; according to the safety margin evaluation data, the key early warning nodes and emergency disposal process of the battery shell are set to generate the safety evaluation result of the battery shell.

[0129] Specifically, the dynamic safety threshold curve shows the working state of the shell under different torque gradients. By analyzing the distribution characteristics of the curve, the operating state can be divided into three intervals: normal operating interval, early warning monitoring interval and risk control interval. The normal operating interval corresponds to a state with low torque, high repair efficiency and stable crack extension, such as a torque range of 20-30 Nm; the early warning monitoring interval corresponds to a state with reduced repair efficiency and accelerated crack extension, such as 30-40 Nm; the risk control interval refers to a dangerous state where the crack may extend out of control or the repair agent is insufficient, such as greater than 40 Nm. Through the division of these areas, operating status partition data is generated, which provides a basis for subsequent monitoring and evaluation.

[0130] Next, based on the operating status partition data, the monitoring parameter thresholds of the battery casing in each operating interval are further determined, and the torque gradient calibration is completed. The monitoring parameters include key indicators such as strain change rate, conductivity, and repair agent response time. For example, in the normal operating range, the strain change rate is less than 5% and the conductivity is greater than 95%; in the early warning monitoring range, the strain change rate rises to 10% and the conductivity drops to 85%-90%; in the risk control range, the strain change rate exceeds 15% and the conductivity is less than 80%. These thresholds are obtained through experimental data fitting and mathematical analysis, and are calibrated based on the torque gradient, so that the state changes in each gradient interval have a clear monitoring benchmark. For example, in the torque range of 30-40 Nm, the monitoring parameters can be calibrated as a strain change rate of 7%-12% and a conductivity of 85%-90%.

[0131] After clarifying the condition monitoring criteria data, the strain response characteristics of the battery shell and the active state of the repair agent are analyzed to generate safety risk level data. The strain response characteristics are obtained by real-time monitoring of the strain distribution. For example, during the loading process, the strain in a certain area increases rapidly from 0.5 millistrain to 1.0 millistrain, reflecting the increased risk of crack propagation. The active state of the repair agent is analyzed by the delay time of the repair agent release and the diffusion efficiency. For example, when loaded to 30 Nm, the response delay of the repair agent is 0.4 seconds and the diffusion efficiency is 85%. Combining these data can evaluate the structural integrity of the shell, the remaining amount of repair agent, and the repair ability. If the crack depth exceeds 80% and the remaining amount of repair agent is less than 50%, it can be marked as a high risk level.

[0132] Based on the generated safety risk level data, maintenance cycle indicators for the battery shell are further established. The maintenance cycle is obtained by quantitatively analyzing the service life and repair ability of the shell. For example, when the remaining amount of repair agent is low but the structural integrity is high, a shorter maintenance cycle (such as 50 hours) can be set; if the repair ability and structural integrity are both in good condition, the maintenance cycle can be extended to 100 hours. These maintenance cycle indicators are then cross-validated with the dynamic safety threshold curve. For example, when the curve shows that the repair efficiency drops rapidly at 40 Nm, the maintenance cycle needs to be adjusted to avoid the shell being in a high-risk area for a long time. The cross-validated results generate safety margin evaluation data, which is used to quantify the safety margin of the shell under different operating conditions.

[0133] Finally, according to the safety margin evaluation data, key early warning nodes and emergency response processes are set to form the final safety assessment results. The early warning node defines the critical point where the operating state enters the risk control interval from the early warning monitoring interval. For example, when the torque exceeds 35 Nm and the strain change rate reaches 10%, and the repair agent delay time exceeds 0.5 seconds, the early warning is triggered. In the emergency response process, when entering the risk control interval, risk control can be carried out by reducing torque, increasing the flow of the repair agent, or stopping operation. At the same time, all key parameters are recorded to provide a basis for subsequent optimization design and safety assessment. This complete process ensures the safe operation of the battery casing under various complex conditions.

[0134] In one embodiment of the present invention, the strain response characteristics and the activity state of the repair agent of the battery shell are analyzed by comparing the state monitoring criterion data to generate safety risk level data of the battery shell, and the safety risk level data includes structural integrity data of the battery shell, remaining repair agent data and re-repairing ability data, including: periodically analyzing the strain response characteristics in the state monitoring criterion data, extracting the strain accumulation rate and strain recovery ratio of the battery shell in a continuous working cycle, and generating structural dynamic stability data of the battery shell; according to the structural dynamic stability data, establishing a repair agent loss curve of the battery shell under different operating conditions, calculating the remaining activity level and diffusion penetration capacity of the repair agent, and generating the repair system availability data of the battery shell; comprehensively evaluating the structural dynamic stability data and the repair system availability data to generate safety risk level data of the battery shell, and the safety risk level data reflects the structural integrity data, remaining repair agent data and re-repairing ability data of the battery shell.

[0135] Specifically, first, by periodically analyzing the strain response characteristics in the condition monitoring criterion data, the strain accumulation rate and strain recovery ratio of the battery shell in the continuous working cycle are extracted. This step aims to quantify the mechanical behavior characteristics of the battery shell in multiple loading cycles. The strain accumulation rate refers to the accumulation rate of permanent strain caused by crack propagation and deformation of the material in one working cycle. For example, in 10 loading cycles, the strain accumulates from the initial 0.5 millistrain to 1.5 millistrain, and the average increase per cycle is 0.1 millistrain. The accumulation rate is 0.1 millistrain / cycle. The strain recovery ratio reflects the elastic recovery ability of the material after unloading. For example, in a loading cycle, the strain recovers from 1 millistrain to 0.4 millistrain, and the recovery ratio is 60%. Through periodic analysis, the deformation law of the battery shell in repeated loading can be fully captured, and the structural dynamic stability data can be generated to evaluate the mechanical stability and durability of the shell.

[0136] Next, based on the structural dynamic stability data, the repair agent loss curve of the battery shell under different operating conditions is established. The repair agent loss curve describes the dynamic characteristics of the repair agent being consumed after the crack is induced by loading. The specific method is to associate the consumption of the repair agent with the crack extension and filling efficiency. For example, under the working condition of 20 Nm of torque, the crack extension rate is 0.5 mm / cycle and the repair agent consumption is 0.3 ml / cycle; under the condition of 40 Nm, the crack extension rate increases to 1 mm / cycle and the repair agent consumption increases to 0.8 ml / cycle. These data are fitted by multiple sets of experiments to obtain the repair agent loss curve, which reflects the response characteristics of the repair agent under different working conditions. In addition, the loss curve can also be combined with the diffusion and penetration ability of the repair agent to further quantify the residual activity level of the repair agent. The diffusion and penetration ability is calculated by the crack filling time and the coverage area. For example, when the crack length is 2 mm, the time required for the repair agent to completely fill is 5 seconds, and the diffusion rate is 0.4 mm / s. These calculation results can accurately reflect the real-time status and repair potential of the repair agent.

[0137] Based on the above data, the repair system availability data is generated, which quantifies the functional performance of the battery shell repair system under different operating conditions. The repair system availability data mainly includes the remaining amount of the repair agent, the release rate and the coverage efficiency. For example, after a certain working cycle, the remaining amount of the repair agent is 80%, the release rate is 0.5 ml / s, and the coverage efficiency is 90%. These indicators together describe whether the repair system can meet subsequent work requirements. If the remaining amount of the repair agent is less than 50% or the coverage efficiency drops below 70%, the availability of the repair system will drop significantly and additional measures will need to be taken.

[0138] Then, the structural dynamic stability data and the repair system availability data are comprehensively evaluated to generate the safety risk level data of the battery shell. The core of the comprehensive assessment is to jointly analyze the mechanical stability and repair ability. For example, when the structural dynamic stability data shows that the strain accumulation rate is 0.2 millistrain / cycle, the recovery ratio is 50%, and the repair system availability data shows that the remaining amount of the repair agent is 60% and the diffusion efficiency is 80%, it can be determined that the shell still has medium repair ability, but monitoring needs to be strengthened. If the strain accumulation rate rises to 0.5 millistrain / cycle, the recovery ratio drops to 30%, and the remaining amount of the repair agent drops to 40%, it is judged to be a high risk level and immediate measures to supplement the repair agent are required. The safety risk level data ultimately reflects the structural integrity of the battery shell (for example, whether the crack exceeds the warning depth), the remaining amount of the repair agent (for example, whether it is sufficient to support multiple repairs), and the re-repair capability (for example, whether it can respond quickly to new cracks).

[0139] The above describes the battery housing torque testing method in the embodiment of the present invention. The following describes the torque testing system in the embodiment of the present invention. Figure 2 , an embodiment of the torque testing system in the embodiment of the present invention includes:

[0140] The initial state monitoring module 101 is used to obtain stress-strain relationship data of the battery housing under the initial load state, and obtain structural microstrain monitoring data of the battery housing under the initial load state, and generate initial self-repair monitoring reference data of the battery housing according to the stress-strain relationship data and the structural microstrain monitoring data, wherein the initial load causes the battery housing to be in an elastic deformation stage;

[0141] A progressive loading control module 102, configured to perform progressive torsion loading on the battery housing according to the initial self-repair monitoring reference data, obtain multi-band optical deformation data and microstructure stress response data of the battery housing during the progressive torsion loading process, and generate crack evolution characteristic data of the battery housing according to the time correlation between the multi-band optical deformation data and the microstructure stress response data;

[0142] A gradient loading test module 103, for determining a torque loading gradient according to the crack evolution characteristic data, and applying the torque loading gradient to the battery housing, obtaining high-frequency strain data, real-time optical monitoring data, and conductive network state data of the battery housing under different torque gradients, and generating repair process characterization data of the battery housing according to the high-frequency strain data, the real-time optical monitoring data, and the conductive network state data;

[0143] A repair efficiency evaluation module 104 is used to perform segmented comparison of a preset area of ​​the battery housing under multiple torque gradients according to the repair process characterization data, obtain time interval data between the crack formation time and the repair agent activation time of the preset area, and harmonic acoustic characteristic data of the preset area, and generate repair efficiency evaluation data of the preset area according to the time interval data and the harmonic acoustic characteristic data;

[0144] A safety threshold analysis module 105, configured to analyze and process the repair efficiency evaluation data, generate a dynamic safety threshold curve of the battery housing, and determine a torque limit range of the battery housing according to the dynamic safety threshold curve;

[0145] The operating state determination module 106 is used to determine the operating state of the battery housing in real time according to the torque limit range and the dynamic safety threshold curve, and generate a safety evaluation result of the battery housing.

[0146] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. All equivalent structural changes made by using the contents of the present invention specification and drawings under the inventive concept of the present invention, or directly / indirectly applied in other related technical fields are included in the patent protection scope of the present invention.

Claims

1. A battery housing torque testing method, characterized in that: include: Acquire stress-strain relationship data of the battery housing under an initial load state, and acquire structural microstrain monitoring data of the battery housing under the initial load state, and generate initial self-repair monitoring reference data of the battery housing according to the stress-strain relationship data and the structural microstrain monitoring data, wherein the initial load causes the battery housing to be in an elastic deformation stage; Performing progressive torsion loading on the battery housing according to the initial self-repair monitoring benchmark data, acquiring multi-band optical deformation data and microstructure stress response data of the battery housing during the progressive torsion loading process, and generating crack evolution characteristic data of the battery housing according to the time correlation between the multi-band optical deformation data and the microstructure stress response data; Determine a torque loading gradient according to the crack evolution characteristic data, apply the torque loading gradient to the battery housing, obtain high-frequency strain data, real-time optical monitoring data, and conductive network state data of the battery housing under different torque gradients, and generate repair process characterization data of the battery housing according to the high-frequency strain data, the real-time optical monitoring data, and the conductive network state data; According to the repair process characterization data, a preset area of ​​the battery housing is segmented and compared under multiple torque gradients to obtain time interval data between the crack formation moment and the repair agent activation moment of the preset area, as well as harmonic acoustic characteristic data of the preset area, and according to the time interval data and the harmonic acoustic characteristic data, generate repair efficiency evaluation data of the preset area; Analyzing and processing the repair efficiency evaluation data to generate a dynamic safety threshold curve of the battery housing, and determining a torque limit range of the battery housing according to the dynamic safety threshold curve; According to the torque limit range and the dynamic safety threshold curve, the operating state of the battery housing is determined in real time to generate a safety assessment result of the battery housing.

2. The battery casing torque testing method according to claim 1, characterized in that: The step of obtaining stress-strain relationship data of the battery housing under the initial load state, and obtaining structural microstrain monitoring data of the battery housing under the initial load state, and generating initial self-repair monitoring reference data of the battery housing according to the stress-strain relationship data and the structural microstrain monitoring data, comprises: Apply a torque loading sequence to the battery housing to obtain multi-level stress-strain relationship data of the battery housing under the torque loading sequence, wherein the torque loading sequence includes an initial loading segment and an incremental loading segment, the loading value of the initial loading segment is greater than the torque value of the elastic deformation starting point of the battery housing and less than the yield limit torque value of the battery housing, the starting loading value of the incremental loading segment is the loading value of the initial loading segment, the loading process of the incremental loading segment is divided into a plurality of loading sub-segments with equal time intervals, the torque value of each loading sub-segment increases a preset torque increment compared to the previous loading sub-segment, and the terminal loading value of the incremental loading segment does not exceed the yield limit torque value; Acquiring structural microstrain monitoring data of the battery housing through a preset collection point array on the surface of the battery housing, wherein the structural microstrain monitoring data includes initial structural microstrain distribution data under the loading value of the initial loading section and dynamic structural microstrain distribution data under the loading values ​​of each level of the incremental loading section; During each loading value maintenance process of the initial loading section and the incremental loading section, the corresponding structural microstrain distribution data is regionally scanned to obtain the repair agent trigger signal data of each area of ​​the battery housing; Performing time series correlation analysis on the multi-level stress-strain relationship data and the repair agent trigger signal data to generate stress fluctuation response characteristic data of the battery housing; The self-repair sensitivity parameter of the battery casing is determined according to the stress fluctuation response characteristic data, and initial self-repair monitoring reference data of the battery casing is generated.

3. The battery casing torque testing method according to claim 2, characterized in that: The performing of time series correlation analysis on the multi-level stress-strain relationship data and the repair agent trigger signal data to generate stress fluctuation response characteristic data of the battery housing includes: The multi-level stress-strain relationship data are grouped according to the time series of applied loads, and the strain mutation points and strain recovery points in each group of data are extracted to generate key stress response time data of the battery housing; Performing time domain analysis on the repair agent trigger signal data, recording the initial activation moment and complete diffusion moment of the repair agent in each torque loading sub-segment, pairing and marking the repair agent trigger signal data based on the key stress response moment data, and generating repair response synchronization data of the battery housing; The stress recovery coefficient and the diffusion coefficient of the repair agent of the battery housing in different torque loading sub-segments are calculated according to the repair response synchronization data to generate stress fluctuation response characteristic data of the battery housing.

4. The battery casing torque testing method according to claim 1, characterized in that: The method of performing progressive torsion loading on the battery housing according to the initial self-repair monitoring benchmark data, obtaining multi-band optical deformation data and microstructure stress response data of the battery housing during the progressive torsion loading process, and generating crack evolution characteristic data of the battery housing according to the time correlation between the multi-band optical deformation data and the microstructure stress response data, comprises: Dividing the initial self-repair monitoring benchmark data into regions to generate a surface stress distribution map of the battery housing and an internal repair agent distribution map; Determine a progressive torque loading scheme according to the surface stress distribution map and the internal repair agent distribution map, wherein the progressive torque loading scheme divides the progressive torque loading into a plurality of loading cycles, the holding time of each loading cycle is proportional to the maximum torque value in the corresponding loading cycle, the torque value in each loading cycle increases step by step according to a preset interval, and there is an overlapping section of the torque values ​​between adjacent loading cycles; In each loading cycle, the surface deformation image of the battery shell is collected through ultraviolet, visible light and near infrared three frequency bands to obtain the strain field data of the battery shell in different frequency bands, and the strain field data of the battery shell in the three frequency bands of ultraviolet, visible light and near infrared are compared and synthesized to generate multi-band optical deformation data; Acquiring conductive network signal data and temperature gradient distribution data inside the battery housing in the overlapping section of each loading cycle, and generating microstructure stress response data according to the conductive network signal data and the temperature gradient distribution data; Pairing the multi-band optical deformation data and the microstructure stress response data according to timestamps to generate crack characterization data of the battery housing in each loading cycle, wherein the crack characterization data includes surface crack morphology data and internal crack extension data; The crack characterization data in each loading cycle is subjected to time series superposition analysis to obtain crack evolution correlation data between different loading cycles, and to generate crack evolution characteristic data of the battery shell.

5. The battery casing torque testing method according to claim 1, characterized in that: The step of determining a torque loading gradient according to the crack evolution characteristic data, applying the torque loading gradient to the battery housing, acquiring high-frequency strain data, real-time optical monitoring data, and conductive network state data of the battery housing under different torque gradients, and generating repair process characterization data of the battery housing according to the high-frequency strain data, the real-time optical monitoring data, and the conductive network state data, includes: Classifying the crack evolution characteristic data according to crack depth and propagation speed to generate crack classification data, and determining the classification interval and torque load sequence of the torque loading gradient according to the crack classification data, wherein the torque load sequence includes a constant torque load and a pulse vibration load; The constant torque load and the pulse vibration load are sequentially applied to the battery housing, wherein the constant torque load is increased in a stepwise manner at a preset rate, and the pulse vibration load is superimposed with a vibration stress of a preset amplitude on the basis of the current constant torque load, and high-frequency strain data is acquired by a dual time scale sampling method of millisecond and sub-millisecond levels; During the loading process of the constant torque load and the pulse vibration load, synchronously collecting data on the battery housing, acquiring surface strain transient distribution data and internal conductivity fluctuation data, and generating real-time optical monitoring data and conductive network status data; Performing time synchronization pairing on the high-frequency strain data, the real-time optical monitoring data and the conductive network state data to generate repair behavior characteristic data of the battery housing under the torque loading gradient; The strain recovery rate parameters, the flow characteristic parameters of the repair agent and the conductive path reconstruction parameters of the battery shell are extracted according to the repair behavior characteristic data to generate the repair process characterization data of the battery shell.

6. The battery casing torque testing method according to claim 1, characterized in that: The method comprises: performing segmented comparison on a preset area of ​​the battery housing under multiple torque gradients according to the repair process characterization data, obtaining time interval data between a crack formation moment and a repair agent activation moment in the preset area, and harmonic acoustic characteristic data of the preset area, and generating repair efficiency evaluation data of the preset area according to the time interval data and the harmonic acoustic characteristic data, including: Performing segmented analysis on the repair process characterization data according to the torque gradient, dividing the crack-prone areas of the battery housing, and generating segmented scanning data of a preset area; According to the segmented scanning data of the preset area, the strain mutation point and the conductive path recovery point of the battery housing are recorded in each torque gradient interval, the time series evolution data of the preset area is generated, and the time interval data between the crack formation time and the repair agent activation time are extracted from the time series evolution data; At the key monitoring position determined by the segmented scanning data of the preset area, the acoustic response data of the repair process is collected for the crack formation time and the repair agent activation time in the time series evolution data, and the harmonic acoustic characteristic data of the preset area is generated; Cross-validating the time interval data and the harmonic acoustic feature data to generate repair quality evaluation data of the preset area at each torque gradient; The repair depth parameter and the repair speed parameter of the preset area are determined according to the repair quality evaluation data, and the repair efficiency evaluation data of the preset area is generated.

7. The battery casing torque testing method according to claim 1, characterized in that: The analyzing and processing the repair efficiency evaluation data to generate a dynamic safety threshold curve of the battery housing, and determining the torque limit range of the battery housing according to the dynamic safety threshold curve, includes: Performing torque gradient correlation analysis on the repair efficiency evaluation data, extracting repair depth data and repair speed data of the battery shell under different torque gradients, and generating repair performance grading data of the battery shell; The repair saturation degree of the battery housing under each torque gradient is calibrated according to the repair performance classification data, the repair agent consumption rate data and the repair layer stability data are obtained, and the repair limit state data of the battery housing is generated; The repair limit state data are grouped according to the safety margin requirements, and safety torque reference data of the battery housing in three repair efficiency intervals of high, medium and low are generated, and a dynamic safety threshold curve of the battery housing is drawn according to the safety torque reference data; Acquiring critical torque data of the battery housing in different repair efficiency intervals based on the dynamic safety threshold curve, and performing correction coefficient compensation on the critical torque data to generate safety margin correction data of the battery housing; The position of the warning point of the dynamic safety threshold curve is adjusted according to the safety margin correction data to determine the torque limit range of the battery housing.

8. The battery casing torque testing method according to claim 1, characterized in that: The step of determining the operating state of the battery housing in real time based on the torque limit range and the dynamic safety threshold curve, and generating a safety evaluation result of the battery housing, includes: Dividing the torque limit range and the dynamic safety threshold curve into regions, dividing the operating state of the battery housing into a normal operating range, a warning monitoring range, and a risk control range, and generating operating state partition data of the battery housing; Determining a monitoring parameter threshold of the battery housing in different operating intervals according to the operating status partition data, and performing torque gradient calibration on the monitoring parameter threshold to generate status monitoring criterion data of the battery housing; Analyzing the strain response characteristics of the battery shell and the activity state of the repair agent according to the state monitoring criterion data, and generating safety risk level data of the battery shell, wherein the safety risk level data includes structural integrity data of the battery shell, remaining amount data of the repair agent, and re-repairability data; Establishing a maintenance cycle index of the battery housing based on the safety risk level data, cross-validating the maintenance cycle index with the dynamic safety threshold curve, and generating safety margin evaluation data of the battery housing; The key early warning nodes and emergency handling procedures of the battery case are set according to the safety margin evaluation data, and the use safety evaluation results of the battery case are generated.

9. The battery casing torque testing method according to claim 8, characterized in that: The method of analyzing the strain response characteristics of the battery shell and the activity state of the repair agent by comparing the state monitoring criterion data to generate safety risk level data of the battery shell, wherein the safety risk level data includes structural integrity data of the battery shell, remaining amount data of the repair agent, and re-repairability data, including: Performing periodic analysis on the strain response characteristics in the state monitoring criterion data, extracting the strain accumulation rate and strain recovery ratio of the battery housing in a continuous working cycle, and generating structural dynamic stability data of the battery housing; According to the structural dynamic stability data, a repair agent loss curve of the battery shell under different operating conditions is established, the residual activity level and diffusion penetration capacity of the repair agent are calculated, and the repair system availability data of the battery shell is generated; The structural dynamic stability data and the repair system availability data are comprehensively evaluated to generate safety risk level data of the battery shell, and the safety risk level data reflects the structural integrity data, the remaining amount of the repair agent data and the re-repairability data of the battery shell.

10. A torque testing system, characterized in that: The torque testing system adopts the battery housing torque testing method according to any one of claims 1 to 9, and the torque testing system comprises: An initial state monitoring module, used to obtain stress-strain relationship data of the battery housing under an initial load state, and obtain structural microstrain monitoring data of the battery housing under the initial load state, and generate initial self-repair monitoring reference data of the battery housing according to the stress-strain relationship data and the structural microstrain monitoring data, wherein the initial load causes the battery housing to be in an elastic deformation stage; A progressive loading control module, used to perform progressive torsion loading on the battery housing according to the initial self-repair monitoring benchmark data, obtain multi-band optical deformation data and microstructure stress response data of the battery housing during the progressive torsion loading process, and generate crack evolution characteristic data of the battery housing according to the time correlation between the multi-band optical deformation data and the microstructure stress response data; a gradient loading test module, for determining a torque loading gradient according to the crack evolution characteristic data, applying the torque loading gradient to the battery housing, acquiring high-frequency strain data, real-time optical monitoring data and conductive network state data of the battery housing under different torque gradients, and generating repair process characterization data of the battery housing according to the high-frequency strain data, the real-time optical monitoring data and the conductive network state data; A repair efficiency evaluation module, used to perform segmented comparison of a preset area of ​​the battery housing under multiple torque gradients according to the repair process characterization data, obtain time interval data between the crack formation moment and the repair agent activation moment of the preset area, and harmonic acoustic characteristic data of the preset area, and generate repair efficiency evaluation data of the preset area according to the time interval data and the harmonic acoustic characteristic data; A safety threshold analysis module, used to analyze and process the repair efficiency evaluation data, generate a dynamic safety threshold curve of the battery housing, and determine the torque limit range of the battery housing according to the dynamic safety threshold curve; The operating state determination module is used to determine the operating state of the battery housing in real time according to the torque limit range and the dynamic safety threshold curve, and generate a safety evaluation result of the battery housing.

Images