Battery bulge multi-modal detection method and system, electronic device, and storage medium

Abstract

This invention provides a method, system, electronic device, and storage medium for multimodal detection of battery bulging, belonging to the field of battery state detection technology. The method includes: acquiring surface strain data and corresponding temperature data collected along an optical fiber distributed along the battery surface; performing temperature compensation processing on the surface strain data based on the temperature data to obtain temperature-compensated strain data; calculating the strain response consistency characteristics at each spatial location to determine the bonding strain transfer parameters characterizing the strain transfer state of the optical fiber; identifying and suppressing spurious strain components to generate corrected strain data; based on the corrected strain data and the bonding strain transfer parameters, inverting the true bulging strain corresponding to the battery volume expansion, and outputting the battery bulging determination result according to the spatial distribution of the true bulging strain. This invention achieves the identification and location of the true battery bulging strain by performing temperature compensation, transfer correction, and spatial consistency inversion on the strain.

Description

Technical Field

[0001] This invention relates to the field of battery state detection technology, and more specifically to a method, system, electronic device, and storage medium for multimodal detection of battery bulging. Background Technology

[0002] With the widespread application of lithium-ion batteries in energy storage systems, electric vehicles, and portable devices, battery safety during operation has become an increasingly important concern. Among these concerns, bulging, a phenomenon that may occur during charging, discharging, or aging, is often related to internal side reactions, gas generation, or structural instability. If this bulging develops uncontrollably, it can easily lead to capacity decay, thermal runaway, or even safety accidents. Therefore, early identification and accurate assessment of battery bulging has become a critical technical requirement in battery operation monitoring.

[0003] Existing methods for detecting battery bulges mainly include visual inspection, point-based displacement or strain sensing, and indirect judgment methods based on electrochemical parameters such as voltage and current. Each of these methods has limitations in engineering applications: visual inspection is easily affected by obstructions and ambient light, making continuous online monitoring difficult; point-based sensing has limited coverage and cannot reflect the spatial distribution characteristics of the bulge on the battery surface; and judgment methods relying solely on electrochemical parameters often only show obvious abnormalities after the bulge has developed to a certain stage, failing to meet the needs of early identification.

[0004] In recent years, distributed fiber optic sensing technology has been gradually introduced into battery bulge detection scenarios due to its ability to continuously monitor strain along the battery surface. This technology can acquire surface strain information with high spatial resolution, providing a new technical means for bulge location identification. However, in practical engineering applications, the surface strain acquired by the fiber optic cable is not entirely equivalent to the actual bulge strain caused by battery volume expansion. Temperature changes generated during battery operation introduce significant thermal strain components. Simultaneously, the adhesive layer between the fiber optic cable and the battery surface may degrade under long-term service or environmental conditions, leading to strain transmission distortion and introducing spurious strain components into the fiber optic strain signal.

[0005] In existing technologies, most solutions only provide simple compensation for temperature factors or assume that the strain transfer relationship between the optical fiber and the battery is stable over a long period, lacking effective means to identify and address the impact of adhesive layer degradation. In this situation, strain anomalies caused by temperature drift or adhesive degradation are easily misjudged as battery bulging, or the actual degree of bulging in the adhesive failure area is underestimated, affecting the reliability and engineering applicability of the detection results. Therefore, there is an urgent need for a detection method that can distinguish between the actual battery volume expansion strain and the pseudo-strain introduced by temperature and adhesive degradation, based on distributed optical fiber strain monitoring, to achieve reliable determination and accurate location of battery bulging. Summary of the Invention

[0006] The purpose of this invention is to provide a method, system, electronic device, and storage medium for multimodal detection of battery bulging, so as to at least solve the problem of difficulty in distinguishing between true bulging strain and pseudo-strain due to temperature drift and adhesive layer degradation in distributed optical fiber strain monitoring.

[0007] To achieve the above objectives, a first aspect of the present invention provides a multimodal detection method for battery bulging. The method includes: acquiring surface strain data and corresponding temperature data collected by optical fibers distributed along the battery surface; performing temperature compensation processing on the surface strain data based on the temperature data to obtain temperature-compensated strain data; calculating strain response consistency characteristics at each spatial location based on the temperature-compensated strain data; determining bonding strain transfer parameters characterizing the strain transfer state of the optical fiber based on the strain response consistency characteristics; identifying and suppressing spurious strain components and generating corrected strain data based on the temperature-compensated strain data, the bonding strain transfer parameters, and the spatial strain continuity relationship; inverting the true bulging strain corresponding to the battery volume expansion based on the corrected strain data and the bonding strain transfer parameters; and outputting a battery bulging determination result based on the spatial distribution of the true bulging strain.

[0008] Optionally, temperature compensation processing is performed on the surface strain data based on the temperature data to obtain temperature-compensated strain data, including: acquiring surface strain data and corresponding temperature data collected by distributed optical fibers under different temperature conditions under a preset baseline state in which no battery volume expansion occurs; constructing thermal strain compensation parameters corresponding to each spatial location based on the response relationship between the surface strain data and the temperature data; calculating the thermal strain component introduced by temperature change at each spatial location based on the thermal strain compensation parameters and the real-time acquired temperature data during the detection process; and removing the thermal strain component from the surface strain data at the corresponding spatial location to obtain temperature-compensated strain data.

[0009] Optionally, calculating the strain response consistency characteristics of each spatial location based on the temperature-compensated strain data includes: under preset strain excitation conditions, acquiring a strain response sequence formed by the change of temperature-compensated strain data over time at each spatial location; based on the strain response sequence, extracting amplitude features to characterize the stability of the strain response and / or phase features to characterize the temporal relationship of the strain response; comparing the amplitude features and / or phase features with reference features of the corresponding spatial location under a preset reference state, and calculating strain response consistency characteristics to characterize the consistency of the strain response at the spatial location.

[0010] Optionally, the preset strain excitation conditions include at least one of the following changes: current change, state of charge change, and temperature change, causing observable strain response changes on the battery surface; based on the strain response sequence, amplitude features for characterizing strain response stability and / or phase features for characterizing the temporal relationship of the strain response are extracted, including: performing time-domain or frequency-domain analysis on the strain response sequence at each spatial location to extract amplitude statistical parameters characterizing the strain response intensity change, wherein the amplitude statistical parameters include any one or more of the mean, fluctuation range, and standard deviation of the strain response amplitude; performing temporal consistency analysis on the strain response sequence to extract phase features characterizing the relative temporal relationship of the strain response, wherein the phase features include the phase difference and / or phase stability of the strain response relative to a preset reference sequence; and using the amplitude statistical parameters and the phase features as the strain response features at the corresponding spatial location.

[0011] Optionally, determining the bonding strain transfer parameters characterizing the strain transfer state of the optical fiber based on the strain response consistency characteristics includes: comparing the strain response consistency characteristics corresponding to each spatial location with the reference consistency characteristics under a preset benchmark state, and calculating a consistency deviation parameter used to characterize the degree of strain response deviation; determining the bonding strain transfer parameters for the corresponding spatial location based on the consistency deviation parameters, wherein the bonding strain transfer parameters are used to characterize the effectiveness of the strain acquired by the optical fiber in transferring to the battery surface.

[0012] Optionally, based on the temperature-compensated strain data, the bond strain transfer parameters, and the spatial strain continuity relationship, spurious strain components are identified and suppressed to generate corrected strain data. This includes: constructing a strain continuity model between adjacent spatial locations based on the temperature-compensated strain data; calculating the continuity deviation parameter of the strain data at each spatial location relative to its neighborhood strain distribution; combining the continuity deviation parameter with the bond strain transfer parameters at the corresponding spatial location to determine spurious strain components introduced by bond layer degradation in the temperature-compensated strain data; performing suppression processing on the strain data determined to be spurious strain components, the suppression processing including replacing with neighborhood continuous strain values ​​or reducing the weight of the spurious strain components; and generating corrected strain data based on the suppressed strain data.

[0013] Optionally, based on the corrected strain data and the bonding strain transfer parameters, the actual bulging strain corresponding to the battery volume expansion is obtained by inversion, and the battery bulging determination result is output according to the spatial distribution of the actual bulging strain. This includes: performing transfer compensation calculation on the corrected strain data and the corresponding bonding strain transfer parameters at each spatial location; obtaining the actual bulging strain data at the corresponding spatial location based on the ratio of the corrected strain data to the bonding strain transfer parameters; performing connectivity analysis on the actual bulging strain data at adjacent spatial locations based on the spatial distribution of the actual bulging strain data to form at least one strain aggregation region; calculating the maximum value and spatial integral value of the actual bulging strain within each strain aggregation region, using the maximum value as an intensity feature characterizing the degree of local volume expansion of the battery, and using the spatial integral value as an expansion feature characterizing the degree of spatial expansion of the bulging region; determining whether the corresponding strain aggregation region continues to exist within a preset time range and shows a cumulative growth trend based on the evolution of the intensity feature and the expansion feature in the time dimension, and determining the strain aggregation region as a battery bulging region when the determination condition is met, and outputting the corresponding battery bulging location result.

[0014] A second aspect of the present invention provides a multimodal detection system for battery bulging. The system includes: a data acquisition unit, configured to acquire surface strain data and corresponding temperature data collected along an optical fiber distributed on the battery surface, and perform temperature compensation processing on the surface strain data based on the temperature data to obtain temperature-compensated strain data; a processing unit, configured to calculate strain response consistency characteristics at each spatial location based on the temperature-compensated strain data, and determine bonding strain transfer parameters characterizing the strain transfer state of the optical fiber according to the strain response consistency characteristics; a correction unit, configured to identify and suppress spurious strain components and generate corrected strain data based on the temperature-compensated strain data, the bonding strain transfer parameters, and the spatial strain continuity relationship; and a result output unit, configured to invert the true bulging strain corresponding to the battery volume expansion based on the corrected strain data and the bonding strain transfer parameters, and output a battery bulging determination result according to the spatial distribution of the true bulging strain.

[0015] A third aspect of the present invention provides an electronic device, comprising: one or more processors; and a storage device having stored one or more programs thereon, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the battery bulging multimodal detection method as described above.

[0016] On the other hand, the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the above-described battery bulging multimodal detection method.

[0017] Through the above technical solution, this invention achieves effective differentiation of different source components in fiber optic strain by performing temperature compensation, strain response consistency analysis, and bonding strain transfer modeling on the battery surface strain acquired by distributed optical fibers. This method can suppress the spurious strain effects introduced by temperature drift and bonding layer degradation without relying on changes to the battery structure, extract the true bulge strain that is physically consistent with the battery volume expansion, and determine the bulge location and state by combining its spatial distribution characteristics, thereby improving the reliability and engineering applicability of battery bulge detection results.

[0018] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0019] The accompanying drawings are provided to further illustrate embodiments of the present invention and form part of the specification. They are used together with the following detailed description to explain the embodiments of the present invention, but do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of the steps of a multimodal detection method for battery bulging provided in one embodiment of the present invention; Figure 2 This is a system structure diagram of a battery bulging multimodal detection system provided in one embodiment of the present invention; Figure 3 This is an internal structural diagram of a computer device provided in one embodiment of the present invention. Detailed Implementation

[0020] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0021] like Figure 1 As shown, this invention provides a multimodal detection method for battery bulging, the method comprising: Step S10: Acquire surface strain data and corresponding temperature data collected by optical fibers distributed along the battery surface, and perform temperature compensation processing on the surface strain data based on the temperature data to obtain temperature-compensated strain data.

[0022] Specifically, under a preset baseline state where no battery volume expansion occurs, surface strain data and corresponding temperature data collected by distributed optical fibers under different temperature conditions are acquired. Based on the response relationship between the surface strain data and the temperature data, thermal strain compensation parameters corresponding to each spatial location are constructed. During the detection process, based on the thermal strain compensation parameters and the real-time acquired temperature data, the thermal strain component introduced by temperature change at each spatial location is calculated. The thermal strain component is removed from the surface strain data at the corresponding spatial location to obtain temperature-compensated strain data.

[0023] In this embodiment of the invention, surface strain data collected at multiple spatial locations by optical fibers distributed along the battery surface, and temperature data corresponding to the surface strain data in time and space, are acquired. The surface strain data reflects the strain response of the optical fibers on the battery surface over time, and the temperature data reflects the temperature change on the battery surface or its adjacent area at the corresponding location. Since the optical fiber material itself is sensitive to temperature changes, and the battery inevitably undergoes temperature rise or fall during operation, directly using the raw surface strain data for analysis can easily lead to misinterpreting the thermal strain caused by temperature changes as mechanical strain resulting from battery volume expansion, thus affecting the accuracy of subsequent judgments.

[0024] Therefore, in this embodiment, before performing actual testing, the response relationship between fiber strain and temperature is calibrated under a preset reference state in which no battery volume expansion occurs.

[0025] Specifically, the battery is placed in a state where it will not bulge or expand in volume, such as in a static state or a low-load operating state. In this state, by changing the ambient temperature or controlling the battery's operating conditions, the surface temperature of the battery is varied between multiple different temperature points, and surface strain data and corresponding temperature data of the distributed optical fiber are collected simultaneously at each spatial location. By statistically analyzing the strain data under different temperature conditions at the same spatial location, the response relationship of surface strain with temperature change at that spatial location is established, thereby obtaining the thermal strain compensation parameters for the corresponding location. The thermal strain compensation parameters are used to characterize the change in optical fiber strain under a unit temperature change condition, and can be obtained according to linear fitting, piecewise fitting, or other numerical fitting methods.

[0026] In the actual testing process, based on the real-time acquired temperature data and the aforementioned thermal strain compensation parameters, the thermal strain component introduced by temperature change at each spatial location is calculated. Specifically, for any spatial location, the difference between the current temperature value at that location and the temperature value under the reference state is first determined. Then, this temperature difference is multiplied by the corresponding thermal strain compensation parameter to obtain the thermal strain component caused by temperature change at that spatial location. This thermal strain component is used to quantitatively characterize the strain contribution introduced by temperature factors in the fiber optic strain signal under the current temperature conditions.

[0027] Furthermore, the thermal strain component is removed from the surface strain data at the corresponding spatial location to obtain temperature-compensated strain data. Specifically, a point-by-point subtraction method can be used to subtract the calculated thermal strain component from the original surface strain data, yielding the strain result under the current temperature conditions after eliminating the temperature effect. Through this processing, the temperature-compensated strain data primarily reflects the strain components in the optical fiber strain related to battery structure deformation, providing a stable and comparable input data foundation for subsequent analysis based on strain response consistency, bonding strain transfer state, and real bulge strain inversion.

[0028] In another possible implementation, during battery operation, a short-time series of strain changes over time is constructed from continuously acquired surface strain data at the same spatial location before temperature compensation. This short-time series is then subjected to trend decomposition processing, decomposing the strain change into slowly changing components and rapidly changing components. The slowly changing components characterize the low-frequency strain trend caused by temperature and environmental condition changes, while the rapidly changing components characterize the high-frequency strain characteristics caused by internal volume changes or local structural responses within the battery.

[0029] In this embodiment, the slowly changing component is used as an estimate of the equivalent thermal strain component and removed from the original surface strain data at the corresponding spatial location, thereby obtaining temperature-compensated strain data that does not require explicit temperature calibration. In this way, even in scenarios with limited temperature sensing accuracy or uneven local temperature distribution, temperature influence suppression can be achieved based on the temporal characteristics of the strain signal itself, providing a stable data foundation for subsequent bond state analysis and real bulge strain inversion.

[0030] Step S20: Calculate the strain response consistency characteristics at each spatial location based on the temperature-compensated strain data, and determine the bonding strain transfer parameters characterizing the strain transfer state of the optical fiber based on the strain response consistency characteristics.

[0031] Specifically, calculating the strain response consistency characteristics of each spatial location based on the temperature-compensated strain data includes: acquiring a strain response sequence formed by the change of temperature-compensated strain data over time at each spatial location under preset strain excitation conditions; extracting amplitude features to characterize the stability of the strain response and / or phase features to characterize the temporal relationship of the strain response based on the strain response sequence; comparing the amplitude features and / or phase features with reference features of the corresponding spatial location under a preset reference state to calculate strain response consistency characteristics to characterize the consistency of the strain response at the spatial location.

[0032] Furthermore, the preset strain excitation conditions include at least one of the following changes: current change, state of charge change, and temperature change, causing observable strain response changes on the battery surface; based on the strain response sequence, amplitude features for characterizing strain response stability and / or phase features for characterizing the temporal relationship of the strain response are extracted, including: performing time-domain or frequency-domain analysis on the strain response sequence at each spatial location to extract amplitude statistical parameters characterizing the strain response intensity change, wherein the amplitude statistical parameters include any one or more of the mean, fluctuation range, and standard deviation of the strain response amplitude; performing temporal consistency analysis on the strain response sequence to extract phase features characterizing the relative temporal relationship of the strain response, wherein the phase features include the phase difference and / or phase stability of the strain response relative to a preset reference sequence; and using the amplitude statistical parameters and the phase features as the strain response features at the corresponding spatial locations.

[0033] Furthermore, determining the bonding strain transfer parameters characterizing the strain transfer state of the optical fiber based on the strain response consistency characteristics includes: comparing the strain response consistency characteristics corresponding to each spatial location with the reference consistency characteristics under a preset benchmark state, and calculating the consistency deviation parameters used to characterize the degree of strain response deviation; determining the bonding strain transfer parameters for the corresponding spatial location based on the consistency deviation parameters, wherein the bonding strain transfer parameters are used to characterize the effectiveness of the strain acquired by the optical fiber in transferring to the battery surface.

[0034] In this embodiment of the invention, after completing the temperature compensation process and obtaining the temperature-compensated strain data, the strain response consistency characteristics at each spatial location are further calculated based on the temperature-compensated strain data, and the bonding strain transfer parameters characterizing the state of strain transfer from the optical fiber to the battery surface are determined accordingly. This process is used to characterize the response stability and temporal consistency of the strain acquired by the optical fiber at different spatial locations, thereby reflecting the changes in the strain transfer state between the optical fiber and the battery surface.

[0035] Specifically, a strain response sequence is constructed under preset strain excitation conditions. These preset strain excitation conditions are not externally imposed forced mechanical loading, but rather utilize objectively existing changes in the battery's operating conditions during normal operation as the strain excitation source, such as changes in charging and discharging current, changes in state of charge, and temperature changes caused by operation. Under these conditions, observable strain response changes will occur on the battery surface, thus providing a data basis for strain consistency analysis. For each spatial location distributed along the battery surface... In a continuous time interval Internally acquired temperature-compensated strain data were used to form a strain response sequence. ,in, Indicates spatial location Location, Time Temperature compensation strain value at any given time.

[0036] After obtaining the strain response sequence, amplitude and phase features are extracted from the sequence. First, the stability of the strain response sequence is quantitatively analyzed from the perspective of amplitude. In this embodiment, amplitude statistical parameters can be calculated for the strain response sequence at each spatial location to reflect the strain response intensity and its fluctuations. The amplitude statistical parameters can be expressed as follows: ,in, Representing the strain response sequence The mean over the time interval; The standard deviation of the strain response sequence is used to reflect the degree of strain fluctuation. This represents the amplitude variation range of the strain response sequence, used to characterize the difference between the maximum and minimum strain. Any one or more of the above amplitude statistical parameters can be used as amplitude features characterizing the stability of the strain response.

[0037] Based on the amplitude characteristics, the temporal relationship of the strain response sequence is further analyzed to extract phase characteristics. In this embodiment, a preset reference strain sequence can be selected. This reference strain sequence can be obtained from the strain response at the same spatial location under the reference state, or by combining the strain responses at adjacent spatial locations. Through the analysis of... and Perform correlation analysis or frequency domain analysis to obtain the corresponding phase characteristics. in, This represents the phase difference between the strain response sequence and the reference strain sequence, used to reflect the relative lag or advance of the strain response in time. Phase stability is a measure of the degree of fluctuation in the phase difference over a time interval. Higher phase stability indicates a more stable strain response time sequence at that spatial location.

[0038] In obtaining amplitude characteristics and phase characteristics Then, the two are combined to form the strain response characteristic vector corresponding to the spatial location. Subsequently, the strain response feature vector is compared with the reference feature vector under a preset reference state. By comparison, strain response consistency characteristics are calculated to characterize the consistency of the strain response. Specifically, a consistency metric function can be defined to quantify the difference between the current state and the baseline state:

[0039] in, Represents vector norm operations. The range of values ​​is used to characterize spatial location. The degree of consistency between the strain response at that location and the reference state. The larger the consistency eigenvalue, the closer the strain response at that location is to the reference state.

[0040] After obtaining the strain response consistency characteristics, bonding strain transfer parameters characterizing the fiber strain transfer state are further determined based on these characteristics. Specifically, the strain response consistency characteristics are compared with reference consistency characteristics under a baseline state to calculate the consistency deviation parameter.

[0041] in, It is used to characterize the degree of deviation of the strain response in the current state from the reference state.

[0042] Subsequently, the bond strain transfer parameters are constructed based on the consistency deviation parameters. This is used to quantitatively describe the effectiveness of strain acquisition via optical fiber in transferring to the battery surface. In this embodiment, the consistency deviation parameter is mapped to the bond strain transfer parameter, for example:

[0043] in, This is a proportional coefficient used to adjust the degree of influence of consistency deviation on the transfer parameter. Using the above method, when the strain response consistency is high, the bond strain transfer parameter takes a larger value, indicating good strain transfer between the optical fiber and the cell surface; when the consistency is significantly reduced, the bond strain transfer parameter decreases accordingly, reflecting the degradation of the bonding layer or strain transfer failure.

[0044] Through the above steps, a quantitative analysis of strain response stability and temporal consistency was achieved based on temperature-compensated strain data. Furthermore, the analysis results were transformed into adhesive strain transfer parameters that can be used for subsequent pseudo-strain identification and real bulge strain inversion.

[0045] Step S30: Based on the temperature-compensated strain data, the bonding strain transfer parameters, and the spatial strain continuity relationship, identify and suppress spurious strain components, and generate corrected strain data.

[0046] Specifically, based on the temperature-compensated strain data, a strain continuity model between adjacent spatial locations is constructed, and the continuity deviation parameter of the strain data at each spatial location relative to its neighborhood strain distribution is calculated. Combining the continuity deviation parameter with the bonding strain transfer parameter at the corresponding spatial location, pseudo-strain components introduced by bonding layer degradation in the temperature-compensated strain data are identified. Suppression processing is performed on the strain data identified as pseudo-strain components, including replacing them with neighborhood continuous strain values ​​or reducing the weight of the pseudo-strain components. Based on the strain data after suppression processing, corrected strain data is generated.

[0047] In this embodiment of the invention, temperature-compensated strain data is obtained. and the corresponding bond strain transfer parameters at spatial locations Subsequently, spurious strain components introduced by bond layer degradation in the temperature-compensated strain data were identified and suppressed to generate corrected strain data. This process is based on the continuous characteristics of strain spatial distribution and, combined with the bond strain transmission state, quantitatively identifies local anomalous strains.

[0048] A spatial strain continuity model is constructed based on the temperature-compensated strain data. Optical fibers distributed along the battery surface form a one-dimensional or quasi-two-dimensional sampling path in space. For any spatial location... Define its neighborhood spatial location set as The neighborhood set is composed of those along the fiber optic path and It consists of several adjacent sampling locations. At the same time. The neighborhood strain distribution is constructed using temperature-compensated strain data within the neighborhood, which is used to characterize the local spatial strain variation trend.

[0049] Based on this, a spatial continuity deviation parameter is calculated to characterize the degree of deviation of the strain at the current spatial location from the strain distribution in its neighborhood. Specifically, the continuity deviation parameter can be defined as:

[0050] in, Indicates spatial location In time Temperature compensation strain value at any given time; express The set of neighborhood locations; This represents the number of spatial locations within the neighborhood. The continuity deviation parameter... It is used to reflect the deviation of the strain value at this location from the average strain level of the neighborhood.

[0051] Furthermore, the continuity deviation parameter is compared with the bond strain transfer parameter at the corresponding spatial location. The bonding strain transfer parameter is used to determine the spurious strain component. It characterizes the effectiveness of strain transfer from the fiber optic acquisition to the battery surface; a smaller value indicates a worse strain transfer state at that location. In this embodiment, when a spatial location simultaneously satisfies the conditions of a large absolute value of continuity deviation and a low bonding strain transfer parameter, it can be considered that the temperature-compensated strain data at that location contains a spurious strain component introduced by bonding layer degradation. The above determination process can be formally expressed as:

[0052] in, This represents a pseudo-strain discrimination index, used to comprehensively reflect the degree of spatial discontinuity and the degree of bond transfer failure. When this index increases significantly, it indicates that the strain anomaly at that spatial location is more likely to originate from bond layer degradation rather than battery volume expansion.

[0053] For spatial locations identified as containing spurious strain components, this embodiment further performs suppression processing. This suppression processing corrects the original strain data, restoring its spatial continuity while reducing the impact of spurious strain on subsequent analysis. One implementation method is to use a neighborhood-based weighted correction method to adjust the original temperature-compensated strain data, obtaining corrected strain data:

[0054] in, This represents the corrected strain data. The above correction method introduces the bond strain transfer parameter as a weighting factor, so that in regions with poor bonding, the strain value at the current spatial location tends to converge more towards the continuous strain distribution in the neighborhood, while in regions with good bonding, the main information of the original temperature-compensated strain data is retained.

[0055] Through the aforementioned spatial continuity modeling, pseudo-strain discrimination, and weighted suppression processing, the generated corrected strain data exhibits better spatial continuity and effectively reduces localized anomalous strain caused by adhesive layer degradation. This corrected strain data serves as input for subsequent real bulge strain inversion, making the inversion process more consistent with the actual physical behavior of battery volume expansion, thus providing a stable and reliable data foundation for the identification and determination of battery bulge regions.

[0056] In another possible implementation, for the same spatial location Calculate the corresponding continuity deviation parameter sequence within multiple consecutive time windows. The temporal evolution trajectory of the continuity deviation was constructed based on this sequence. Trend analysis was performed on the evolution trajectory to distinguish between the slow, unidirectional drift characteristic of the continuity deviation over time and the instantaneous abrupt change characteristic. Specifically, a continuous increase in the continuity deviation over a longer timescale, asynchronous with the deviation changes at adjacent spatial locations, was considered a characteristic of gradual degradation of the adhesive layer; while a continuity deviation that appears and disappears rapidly within a short period was considered sporadic noise or instantaneous disturbance.

[0057] In this embodiment, the temporal evolution characteristics of the continuity deviation are used in conjunction with the adhesive strain transfer parameters at the corresponding spatial location. When the continuity deviation shows a continuous evolution trend and the adhesive strain transfer parameters decrease synchronously, it is determined that the abnormal strain at that spatial location mainly originates from the adhesive layer degradation process. For this type of pseudo-strain component, instead of directly using a neighborhood replacement method for suppression, a progressive reduction process is performed on it in the time dimension, ensuring a smooth transition of the corrected strain data over time, thereby avoiding the introduction of new discontinuities due to one-time suppression. Through this method, this embodiment can identify and process pseudo-strains with progressive characteristics, providing more stable data support for battery bulging monitoring under long-term operating conditions.

[0058] Step S40: Based on the corrected strain data and the bonding strain transfer parameters, the actual bulging strain corresponding to the battery volume expansion is obtained by inversion, and the battery bulging determination result is output according to the spatial distribution of the actual bulging strain.

[0059] Specifically, a transfer compensation calculation is performed on the corrected strain data and the corresponding bonding strain transfer parameters at each spatial location. Based on the ratio of the corrected strain data to the bonding strain transfer parameters, the actual bulge strain data at the corresponding spatial location is obtained. Based on the distribution of the actual bulge strain data on the battery surface, connectivity analysis is performed on the actual bulge strain data at adjacent spatial locations to form at least one strain aggregation region. For each strain aggregation region, the maximum value and spatial integral value of the actual bulge strain within the region are calculated. The maximum value is used as an intensity feature characterizing the degree of local volume expansion of the battery, and the spatial integral value is used as an expansion feature characterizing the degree of spatial expansion of the bulge region. Based on the evolution of the intensity feature and the expansion feature over time, it is determined whether the corresponding strain aggregation region continues to exist within a preset time range and shows a cumulative growth trend. When the condition is met, the strain aggregation region is identified as a battery bulge region, and the corresponding battery bulge location result is output.

[0060] In this embodiment of the invention, after obtaining the corrected strain data and the bonding strain transfer parameters at the corresponding spatial location, strain inversion processing is further performed on the corrected strain data to obtain the true bulging strain that reflects the battery volume expansion behavior. Based on the distribution characteristics of the true bulging strain on the battery surface space, the identification and determination of the battery bulging area are completed.

[0061] Strain transfer compensation calculations are performed on the corrected strain data and corresponding bond strain transfer parameters at each spatial location. The corrected strain data is the strain result obtained after suppressing spurious strain components, and its value is still affected by the strain transfer efficiency between the optical fiber and the battery surface. To eliminate the influence of bond transfer attenuation on the true volume expansion strain, in this embodiment, the corrected strain data is compared with the bond strain transfer parameters at the corresponding spatial location to obtain the true bulge strain data. This process can be expressed as:

[0062] in, Indicates spatial location In time Real-time corresponding bump strain; This indicates the correction of strain data; This represents the bond strain transfer parameter at that spatial location. Through the above compensation calculations, it is possible to recover the strain amplitude corresponding to the battery volume expansion even in regions where bond transfer efficiency is reduced.

[0063] After obtaining the actual bulge strain data at each spatial location, its distribution on the battery surface is further analyzed. Since battery volume expansion typically manifests as a continuously expanding structural deformation within a localized region, the actual bulge strain associated with the bulge exhibits spatial connectivity. In this embodiment, connectivity analysis is performed on the actual bulge strain data at adjacent spatial locations, grouping spatially adjacent locations with continuously distributed actual bulge strains into the same strain aggregation region. This connectivity analysis can be based on the adjacent sampling position relationship along the fiber optic path or on the two-dimensional neighborhood relationship mapped from the battery surface.

[0064] Through the above connectivity analysis, at least one strain aggregation region is formed. Each strain aggregation region consists of multiple spatial locations and is used to characterize candidate regions where volume expansion may occur on the battery surface. For each strain aggregation region, characteristic parameters reflecting the degree of bulging are further calculated. In this embodiment, the maximum value and spatial integral value of the actual bulging strain within the strain aggregation region are calculated respectively. The maximum value of the actual bulging strain is used to characterize the local volume expansion intensity of the battery within this region, and its calculation method is as follows:

[0065] in, Indicates the first A strain accumulation zone.

[0066] Simultaneously, spatial integration is performed on the actual bulge strain within the strain aggregation region to characterize the spatial expansion of the bulge region, which can be specifically expressed as:

[0067] in, Indicates spatial location The corresponding equivalent area or equivalent length weights. The spatial integral value comprehensively reflects the cumulative strain effect of the bulge region and is used to characterize the expansion characteristics of the bulge in spatial range.

[0068] In obtaining the intensity characteristics of the strain accumulation region and extended features Subsequently, the evolution of the aforementioned features over time is further analyzed. In this embodiment, for the same strain accumulation region, the changing trends of its intensity and expansion characteristics are tracked over multiple consecutive time periods. When the strain accumulation region persists within a preset time range, and its corresponding intensity and expansion characteristics show a cumulative growth trend over time, it can be determined that the strain accumulation region is not caused by instantaneous disturbances or noise, but is consistent with the battery volume expansion process.

[0069] Specifically, through the and Trend analysis is performed on the changes over time to determine whether they exhibit a continuous increase. When the above criteria are met, the corresponding strain accumulation area is identified as the battery bulge area, and the spatial location of this bulge area on the battery surface is output as the battery bulge location result.

[0070] In another possible implementation, after obtaining the corrected strain data, a second-order spatial difference operation is performed on the corrected strain data along the spatial direction of the battery surface to calculate the strain curvature parameters at each spatial location, which are used to characterize the degree of bending of the strain in space. Since real battery bulges usually exhibit a convex structure in a local area, the corresponding strain distribution has obvious curvature concentration characteristics in space; while strain changes caused by overall thermal expansion or slow structural deformation often exhibit low curvature and gentle distribution characteristics.

[0071] In this embodiment, the strain curvature parameter is used in conjunction with the bond strain transfer parameter at the corresponding spatial location. Volumetric expansion inversion processing is performed on the corrected strain data within a spatial region only when a high strain curvature level is simultaneously observed within the effective transfer range of the bond strain transfer parameter. Subsequently, the boundary of the bulge region is determined based on the continuous spatial distribution of the strain curvature.

[0072] like Figure 2 As shown, this invention provides a multimodal detection system for battery bulging. The system includes: a data acquisition unit 201, used to acquire surface strain data and corresponding temperature data collected along an optical fiber distributed on the battery surface, and to perform temperature compensation processing on the surface strain data based on the temperature data to obtain temperature-compensated strain data; a processing unit 202, used to calculate the strain response consistency characteristics at each spatial location based on the temperature-compensated strain data, and to determine the bonding strain transfer parameters characterizing the strain transfer state of the optical fiber according to the strain response consistency characteristics; a correction unit 203, used to identify and suppress spurious strain components and generate corrected strain data based on the temperature-compensated strain data, the bonding strain transfer parameters, and the spatial strain continuity relationship; and a result output unit 204, used to invert the true bulging strain corresponding to the battery volume expansion based on the corrected strain data and the bonding strain transfer parameters, and to output the battery bulging determination result according to the spatial distribution of the true bulging strain.

[0073] The present invention also provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the above-described battery bulge multimodal detection method.

[0074] This invention also provides an electronic device, including: one or more processors; and a storage device storing one or more programs thereon, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the battery bulging multimodal detection method as described above.

[0075] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 3As shown, the computer device includes a processor A01, a network interface A02, a display screen A04, an input device A05, a memory (not shown), and a database (not shown) connected via a system bus. The processor A01 provides computing and control capabilities. The memory includes internal memory A03 and a non-volatile storage medium A06. The non-volatile storage medium A06 stores an operating system B01, a computer program B02, and a database (not shown). The internal memory A03 provides an environment for the operation of the operating system B01 and the computer program B02 stored in the non-volatile storage medium A06. The network interface A02 is used for communication with external terminals via a network connection. When the computer program B02 is executed by the processor A01, it implements a multimodal detection method for battery bulging.

[0076] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. This program is stored in a storage medium and includes several instructions to cause a microcontroller, chip, or processor to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0077] The optional embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the embodiments of the present invention are not limited to the specific details described above. Within the scope of the technical concept of the embodiments of the present invention, various simple modifications can be made to the technical solutions of the embodiments of the present invention, and these simple modifications all fall within the protection scope of the embodiments of the present invention. It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, the embodiments of the present invention will not further describe the various possible combinations.

[0078] Furthermore, various different embodiments of the present invention can be combined in any way, as long as they do not violate the spirit of the embodiments of the present invention, they should also be regarded as the content disclosed by the embodiments of the present invention.

Claims

1. A multimodal detection method for battery bulging, characterized in that, The method includes: Surface strain data and corresponding temperature data are acquired by optical fibers distributed along the battery surface, and temperature compensation processing is performed on the surface strain data based on the temperature data to obtain temperature-compensated strain data. Based on the temperature-compensated strain data, the strain response consistency characteristics at each spatial location are calculated, and the bonding strain transfer parameters characterizing the strain transfer state of the optical fiber are determined according to the strain response consistency characteristics. Based on the temperature-compensated strain data, the bonding strain transfer parameters, and the spatial strain continuity relationship, spurious strain components are identified and suppressed, and corrected strain data is generated. Based on the corrected strain data and the bonding strain transfer parameters, the actual bulging strain corresponding to the battery volume expansion is obtained by inversion, and the battery bulging judgment result is output according to the spatial distribution of the actual bulging strain.

2. The multimodal detection method for battery bulge according to claim 1, characterized in that, Temperature compensation processing is performed on the surface strain data based on the temperature data to obtain temperature-compensated strain data, including: Under a preset baseline state where no battery volume expansion occurs, surface strain data and corresponding temperature data collected by distributed optical fibers under different temperature conditions are obtained. Based on the response relationship between the surface strain data and the temperature data, thermal strain compensation parameters corresponding to each spatial location are constructed. During the detection process, based on the thermal strain compensation parameters and the real-time temperature data, the thermal strain components introduced by temperature changes at each spatial location are calculated. The thermal strain component is removed from the surface strain data at the corresponding spatial location to obtain temperature-compensated strain data.

3. The multimodal detection method for battery bulging according to claim 1, characterized in that, Based on the temperature-compensated strain data, the consistency characteristics of the strain response at each spatial location are calculated, including: Under preset strain excitation conditions, the strain response sequence formed by the change of temperature-compensated strain data over time at each spatial location is obtained. Based on the strain response sequence, strain response features are extracted, including amplitude features for characterizing strain response stability and / or phase features for characterizing strain response temporal relationship. The strain response characteristics are compared with the reference characteristics of the corresponding spatial location under a preset reference state to calculate the strain response consistency characteristics used to characterize the consistency of the strain response at the spatial location.

4. The multimodal detection method for battery bulge according to claim 3, characterized in that, The preset strain excitation conditions include at least one of the following changing states: current change, state of charge change, and temperature change, which causes observable strain response changes on the battery surface. Based on the strain response sequence, strain response features are extracted, including: Perform time-domain or frequency-domain analysis on the strain response sequence at each spatial location to extract amplitude statistical parameters characterizing the change in strain response intensity. The amplitude statistical parameters include any one or more of the mean, fluctuation range, and standard deviation of the strain response amplitude, and the amplitude characteristics are any one or more of the amplitude statistical parameters. A time-series consistency analysis is performed on the strain response sequence to extract phase features characterizing the relative temporal relationship of the strain response. The phase features include the phase difference and / or phase stability of the strain response relative to a preset reference sequence. The amplitude statistical parameters and the phase characteristics are used as the strain response characteristics of the corresponding spatial location.

5. The multimodal detection method for battery bulge according to claim 1, characterized in that, Based on the strain response consistency characteristics, the bonding strain transfer parameters characterizing the strain transfer state of the optical fiber are determined, including: The consistency characteristics of strain response at each spatial location are compared with the reference consistency characteristics under the preset benchmark state, and the consistency deviation parameter used to characterize the degree of deviation of strain response is calculated. The bonding strain transfer parameters for the corresponding spatial location are determined based on the consistency deviation parameters, wherein the bonding strain transfer parameters are used to characterize the effectiveness of the strain acquired by the optical fiber in transferring to the battery surface.

6. The multimodal detection method for battery bulging according to claim 1, characterized in that, Based on the temperature-compensated strain data, the bond strain transfer parameters, and the spatial strain continuity relationship, spurious strain components are identified and suppressed, and corrected strain data is generated, including: Based on the temperature-compensated strain data, a strain continuity model between adjacent spatial locations is constructed, and the continuity deviation parameter of the strain data at each spatial location relative to the strain distribution in its neighborhood is calculated. By combining the continuity deviation parameter with the adhesive strain transfer parameter at the corresponding spatial location, the pseudo-strain component introduced by adhesive layer degradation in the temperature-compensated strain data is determined. Suppression processing is performed on strain data identified as spurious strain components. The suppression processing includes replacing the spurious strain components with neighboring continuous strain values ​​or reducing the weight of the spurious strain components. Corrected strain data are generated based on the strain data after suppression processing.

7. The multimodal detection method for battery bulge according to claim 6, characterized in that, Based on the corrected strain data and the bonding strain transfer parameters, the actual bulging strain corresponding to the battery volume expansion is obtained by inversion, and the battery bulging determination result is output according to the spatial distribution of the actual bulging strain, including: A transfer compensation calculation is performed on the corrected strain data and the corresponding bond strain transfer parameters at each spatial location. Based on the ratio of the corrected strain data to the bond strain transfer parameters, the actual bulge strain data at the corresponding spatial location is obtained. Based on the distribution of the actual bulging strain data on the battery surface, connectivity analysis is performed on the actual bulging strain data at adjacent spatial locations to form at least one strain aggregation region. For each strain accumulation region, the maximum value and spatial integral value of the actual bulging strain within the strain accumulation region are calculated. The maximum value is used as the intensity feature characterizing the degree of local volume expansion of the battery, and the spatial integral value is used as the expansion feature characterizing the degree of spatial expansion of the bulging region. Based on the evolution of the intensity features and the expansion features over time, it is determined whether the corresponding strain accumulation region continues to exist within a preset time range and shows a cumulative growth trend. When the condition is met, the strain accumulation region is identified as a battery bulge region, and the corresponding battery bulge location result is output.

8. A multimodal detection system for battery bulging, characterized in that, The system includes: The acquisition unit is used to acquire surface strain data and corresponding temperature data collected by optical fibers distributed along the battery surface, and to perform temperature compensation processing on the surface strain data based on the temperature data to obtain temperature-compensated strain data. The processing unit is used to calculate the strain response consistency characteristics at each spatial location based on the temperature-compensated strain data, and to determine the bonding strain transfer parameters characterizing the strain transfer state of the optical fiber based on the strain response consistency characteristics. The correction unit is used to identify and suppress spurious strain components and generate corrected strain data based on the temperature-compensated strain data, the bonding strain transfer parameters, and the spatial strain continuity relationship. The result output unit is used to invert the actual bulging strain corresponding to the battery volume expansion based on the corrected strain data and the bonding strain transfer parameters, and output the battery bulging judgment result according to the spatial distribution of the actual bulging strain.

9. An electronic device, characterized in that, include: One or more processors; A storage device having stored one or more programs thereon, which, when executed by the one or more processors, cause the one or more processors to implement the battery bulge multimodal detection method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the battery bulge multimodal detection method according to any one of claims 1-7.

Images