A fiber-guided battery bulge positioning method, system, electronic device and medium

Abstract

This invention provides a fiber-optic guided method, system, electronic device, and medium for locating battery bulges, belonging to the field of battery safety monitoring technology. The method includes: acquiring strain data from distributed optical fibers deployed on the battery surface to form a strain sequence distributed along spatial locations; acquiring battery operating state parameters time-aligned with the strain sequence to construct an operating condition background; grouping and modeling historical strain sequences to construct a reference model; performing consistency analysis between the current strain sequence and the reference model under the corresponding operating condition background to generate reliability evaluation parameters; performing reliability compensation processing on the current strain sequence to generate a reliability-compensated equivalent strain sequence; and performing bulge feature recognition and spatial positioning analysis to output the bulge location result. This invention achieves stable positioning of battery bulges under long-term operating conditions by evaluating and compensating for the reliability of distributed optical fiber strain data under operating condition constraints.

Description

Technical Field

[0001] This invention relates to the field of battery safety monitoring technology, specifically to a fiber optic-guided method, system, electronic device, and medium for locating battery bulges. Background Technology

[0002] As power batteries and energy storage battery systems develop towards higher energy density and longer lifespan, bulging issues caused by internal aging and fluctuations in operating conditions during long-term operation are gradually becoming a significant hidden danger affecting system safety. Once bulging occurs, it is often accompanied by structural stress concentration, deterioration of heat dissipation conditions, and even the risk of failure. Therefore, timely and accurate monitoring and location of battery bulges is of great engineering significance.

[0003] Existing methods for monitoring battery bulging often rely on visual inspection, point sensing, or manual inspection, which suffer from limited monitoring coverage, sensitivity to obstruction, and difficulty in achieving continuous online monitoring. Distributed fiber optic sensing technology, due to its advantages such as spatial continuity, resistance to electromagnetic interference, and ease of large-scale deployment, is increasingly being applied to battery surface strain monitoring and for the identification and localization of bulge features. However, in practical engineering applications, distributed fiber optics typically need to operate in battery systems for extended periods. The strain data collected by these fibers exhibits decreased stability and changes in statistical characteristics over time, making it difficult to guarantee the reliability of directly judging bulges based on raw strain data.

[0004] Currently, the processing of long-term operational reliability of distributed fiber optic strain data largely relies on periodic calibration, manual experience correction, or simple threshold judgment. These methods often require the introduction of additional operating conditions or human intervention, making them difficult to adapt to the actual needs of online battery system operation. Furthermore, existing technologies lack a solution that can evaluate the applicability of fiber optic strain data for bulge localization based solely on strain data and operational status information during operation, without introducing calibration conditions, and effectively feed this evaluation result back into bulge localization analysis.

[0005] Therefore, there is an urgent need for a bulge localization method for long-term battery operation scenarios. This method should evaluate the reliability of distributed fiber optic strain data without changing the existing hardware structure, and compensate and constrain the bulge localization results based on the reliability, thereby improving the reliability and practicality of battery bulge localization results in engineering applications. Summary of the Invention

[0006] The purpose of this invention is to provide a fiber-optic guided method, system, electronic device, and medium for locating battery bulges, so as to at least solve the problem that the reliability of the results is difficult to guarantee when locating bulges based on distributed fiber optic strain data under long-term battery operation conditions.

[0007] To achieve the above objectives, a first aspect of the present invention provides a fiber-optic guided method for locating battery bulges. The method includes: acquiring strain data from distributed optical fibers deployed on the battery surface to form a strain sequence distributed along spatial locations, and acquiring battery operating state parameters time-aligned with the strain sequence to construct an operating condition background; grouping and modeling historical strain sequences based on the operating condition background to construct a reference model, and performing consistency analysis between the current strain sequence and the reference model under the corresponding operating condition background to generate a reliability evaluation parameter; performing reliability compensation processing on the current strain sequence based on the reliability evaluation parameter to generate a reliability-compensated equivalent strain sequence; and performing bulge feature recognition and spatial positioning analysis based on the equivalent strain sequence to output a bulge location result.

[0008] Optionally, the process involves acquiring strain data from distributed optical fibers deployed on the battery surface to form a strain sequence distributed along spatial locations, and acquiring battery operating state parameters aligned with the time of the strain sequence to construct an operating condition background. This includes: determining the time intervals in which the operating state changes during battery operation based on the change process of the battery operating state parameters in the time dimension; extracting strain sequence segments from the distributed optical fiber strain data that completely overlap with the time interval within each time interval of the operating state change; performing spatial consistency calculations on each strain sequence segment to generate spatial statistical results characterizing the spatial distribution characteristics of the strain sequence segment; performing temporal stability calculations on each strain sequence segment to generate temporal statistical results characterizing the temporal evolution characteristics of the strain sequence segment; and generating operating condition characterization quantities describing the operating state characteristics of the corresponding time interval based on the spatial and temporal statistical results, and constructing an operating condition background based on the operating condition characterization quantities.

[0009] Optionally, a reference model is constructed by grouping historical strain sequences based on the aforementioned operating conditions background. This includes: screening historical strain sequences for consistency based on the operating conditions background generated during historical operation to form multiple strain data sets with consistent operating conditions; calculating statistical descriptive quantities of the strain sequences in terms of spatial distribution and temporal evolution characteristics for each strain data set; and constructing a reference model based on the statistical descriptive quantities to characterize the typical evolutionary behavior of strain sequences under the corresponding operating conditions background.

[0010] Optionally, a consistency analysis is performed between the current strain sequence and a reference model under the corresponding operating conditions to generate reliability evaluation parameters. This includes: selecting a reference model that matches the current operating conditions within the current operating period; calculating the deviation between the current strain sequence and the reference model in terms of spatial distribution characteristics to obtain a spatial consistency evaluation result; calculating the deviation between the current strain sequence and the reference model in terms of temporal evolution characteristics to obtain a temporal consistency evaluation result; and generating reliability evaluation parameters based on the spatial consistency evaluation result and the temporal consistency evaluation result to characterize the reliability of the current strain sequence under the current operating conditions.

[0011] Optionally, a credibility compensation process is performed on the current strain sequence based on the credibility evaluation parameters to generate a credibility-compensated equivalent strain sequence, including: determining the credibility weight corresponding to each spatial location in the current strain sequence based on the distribution of the credibility evaluation parameters in the spatial location dimension; applying the corresponding credibility weight to the strain value at each spatial location in the current strain sequence to generate a credibility-weighted strain sequence; performing spatial aggregation processing on the weighted strain sequence to weaken the strain fluctuations corresponding to low credibility spatial locations and retain the strain characteristics corresponding to high credibility spatial locations; and using the strain result after spatial aggregation processing as the credibility-compensated equivalent strain sequence.

[0012] Optionally, the rules for performing spatial aggregation processing on the weighted strain sequence are as follows: along the spatial location dimension of the distributed optical fiber, for each spatial location, select neighborhood strain data centered on that spatial location and covering a preset spatial length; according to the confidence weight of the neighborhood strain data, perform weighted aggregation calculation on the neighborhood strain data to increase the proportion of strain data with higher confidence weight in the aggregation result; use the result obtained from the weighted aggregation calculation as the aggregated strain value at the corresponding spatial location; use the aggregated strain values ​​corresponding to each spatial location to form a spatially continuous strain distribution result, which is used as the spatial aggregation processing result.

[0013] Optionally, based on the equivalent strain sequence, bulge feature recognition and spatial positioning analysis are performed to output bulge positioning results, including: identifying a set of spatial locations in the equivalent strain sequence where the strain amplitude exceeds a preset discrimination threshold, forming candidate bulge response intervals; calculating strain peak features and strain accumulation features for the equivalent strain sequence within each candidate bulge response interval to form bulge feature parameters used to characterize the bulge response intensity; filtering the candidate bulge response intervals based on the bulge feature parameters to determine target bulge intervals that meet preset bulge judgment conditions; and outputting the spatial position of the target bulge interval in the equivalent strain sequence as the spatial positioning result of the battery bulge.

[0014] A second aspect of the present invention provides an optical fiber-guided battery bulge localization system, the system comprising: a data acquisition unit, configured to acquire strain data from distributed optical fibers deployed on the battery surface to form a strain sequence distributed along a spatial location, and to acquire battery operating state parameters time-aligned with the strain sequence to construct an operating condition background; a processing unit, configured to group and model historical strain sequences based on the operating condition background to construct a reference model, and to perform consistency analysis between the current strain sequence and the reference model under the corresponding operating condition background to generate a reliability evaluation parameter; a compensation unit, configured to perform reliability compensation processing on the current strain sequence based on the reliability evaluation parameter to generate a reliability-compensated equivalent strain sequence; and a result output unit, configured to perform bulge feature recognition and spatial positioning analysis based on the equivalent strain sequence to output the bulge localization result.

[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 fiber-optic guided battery bulge positioning 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 aforementioned fiber-optic guided battery bulge positioning method.

[0017] Through the above technical solution, this invention simultaneously acquires distributed optical fiber strain data and battery operating state parameters during battery operation, constructing an operating condition background that matches the strain evolution process. This ensures that strain data analysis is based on clear operating condition constraints, thereby avoiding interference from mixed operating states in bulge detection. Based on this, by grouping historical strain sequences into operating condition models and performing consistency analysis between the current strain sequence and the reference model under corresponding operating conditions, a quantitative evaluation of the applicability of the current strain data to bulge analysis can be achieved without introducing calibration conditions, forming a reliability evaluation parameter. Furthermore, the reliability evaluation parameter is used to compensate for the current strain sequence, suppressing low-reliability strain responses and preserving high-reliability strain characteristics, thus obtaining a more reliable equivalent strain sequence. Finally, bulge feature recognition and spatial positioning analysis are performed based on the equivalent strain sequence, ensuring that the output bulge positioning results maintain high stability and engineering usability even under long-term battery operation scenarios, effectively improving the reliability of the distributed optical fiber bulge positioning method in practical applications.

[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 fiber-optic guided battery bulge positioning method according to one embodiment of the present invention. Figure 2 This is a system structure diagram of a fiber-optic guided battery bulge positioning 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, an embodiment of the present invention provides a fiber-optic guided method for locating battery bulges, the method comprising: Step S10: Obtain the strain data of the distributed optical fibers deployed on the battery surface to form a strain sequence distributed along the spatial location, and obtain the battery operating state parameters that are time-aligned with the strain sequence to construct the operating condition background.

[0022] Specifically, based on the changes in battery operating state parameters over time, the time intervals in which the operating state changes during battery operation are determined. Within each time interval of the change in operating state, strain sequence segments that completely overlap with that time interval are extracted from the distributed optical fiber strain data. Spatial consistency calculations are performed on each strain sequence segment to generate spatial statistical results characterizing the spatial distribution characteristics of that strain sequence segment. Temporal stability calculations are performed on each strain sequence segment to generate temporal statistical results characterizing the temporal evolution characteristics of that strain sequence segment. Based on the spatial and temporal statistical results within the corresponding time interval, operating condition characterization quantities describing the operating state characteristics of that time interval are generated, and an operating condition background is constructed based on these operating condition characterization quantities.

[0023] In this embodiment of the invention, strain data from distributed optical fibers deployed on the surface of a battery are acquired. The distributed optical fibers are deployed along a predetermined path on the battery surface to continuously collect strain response information of the battery during operation. Through a distributed optical fiber sensing unit, a strain sequence distributed along the spatial location of the optical fiber is obtained within a sampling period; this strain sequence is denoted as... ,in Indicates a spatial location index along a distributed optical fiber. This represents the sampling time index. The strain sequence is used to characterize the strain state of the battery at different spatial locations over time, and serves as the basic input for subsequent operation condition division and data segment extraction.

[0024] Simultaneously, battery operating status parameters synchronized with the strain sequence time are obtained from the battery management system or equivalent operation monitoring module. These battery operating status parameters are denoted as... This is used to reflect the changes in battery operating conditions during operation and to provide operating condition information over time. The battery operating state parameters are not limited to a specific set of physical quantities, but are required to reflect the characteristics of changes in operating state over time, so as to be used for determining the range of operating state changes and constructing the operating condition background.

[0025] Based on battery operating status parameters The changes over time are analyzed to understand the battery's operation and determine the set of time intervals in which its operating state changes. Specifically, this is achieved through... The changing trends, magnitudes of changes, state transition markers, or periods of stable development are identified to obtain a set of time intervals. Each of them These are all continuous time intervals on the time axis, used to characterize whether the battery is in a relatively consistent operating state or undergoes a specific change in operating state within that time interval. This set of time intervals serves to constrain the truncation boundaries of strain data, ensuring that subsequent processing is carried out on clearly defined segments of operating conditions.

[0026] In a defined time interval set Then, in each time interval Within, extracting data from distributed fiber optic strain data within a time interval Completely overlapping strain sequence segments yield a set of interval strain sequence segments. ,in Indicates the time interval The strain sequence segments extracted from the interval. These interval strain sequence segments serve as the direct processing objects for subsequent spatial consistency calculations and temporal stability calculations, thereby ensuring that the statistical results correspond one-to-one with specific operating conditions.

[0027] For each interval strain sequence segment Spatial consistency calculations are performed to generate spatial statistical results characterizing the spatial distribution features of strain sequence fragments within a time interval. The input to the spatial consistency calculation is spatially distributed data within the same time interval, and the output is a set of statistics or features that can be used to describe the spatial distribution characteristics. These spatial statistical results are used to characterize the spatial distribution within the time interval. The degree of consistency and spatial fluctuation pattern of the internal strain response across spatial dimensions. This spatial statistical result serves as a source of information on the spatial dimension in the subsequent construction of operational condition characterization quantities.

[0028] Meanwhile, for the same interval strain sequence segment The system performs time stability calculations, generating temporal statistics to characterize the temporal evolution of interval strain sequence segments. The input to the time stability calculation is a time interval. For time series data within a given time interval, the output of time stability calculation is a set of statistics or features that can be used to describe the characteristics of time evolution. These time statistics are used to characterize the time interval. The degree of fluctuation and stable state of the internal strain response over time. This time-related statistical result serves as a source of information on the time dimension in the subsequent construction of operating condition characterization quantities.

[0029] In the time interval After obtaining the corresponding spatial and temporal statistical results, an operational condition characterization quantity is generated based on these results. This operational condition characterization quantity is used to describe the time interval. The corresponding operating status characteristics, and the time interval Establish connections. Furthermore, construct operational condition backgrounds based on the operational condition characterization quantities corresponding to each time interval, enabling the operational condition backgrounds to segment the operational process in units of time intervals, and providing clear operational condition constraints and data organization basis for subsequent grouping modeling of historical strain sequences, consistency analysis of current strain sequences, and generation of credibility evaluation parameters.

[0030] In another possible implementation, the operating condition backgrounds for each time interval are constructed as described above, and the operating condition backgrounds of adjacent time intervals are correlated to form an operating condition background sequence. Subsequently, the equivalent strain sequence change trend of the corresponding spatial location in the continuous operating condition background sequence is analyzed to identify strain anomaly patterns that persist and have stable spatial locations across multiple operating condition backgrounds. For strain anomalies that appear only in a single operating condition background but cannot maintain consistent spatial characteristics in adjacent operating condition backgrounds, their weight in the bulge localization analysis is reduced or they are excluded from bulge determination.

[0031] Step S20: Based on the aforementioned operating conditions, group and model the historical strain sequences to construct a reference model, and perform consistency analysis between the current strain sequence and the reference model under the corresponding operating conditions to generate credibility evaluation parameters.

[0032] Specifically, the historical strain sequences are grouped and modeled based on the aforementioned operating conditions to construct a reference model, including: screening the historical strain sequences for consistency of operating conditions based on the operating conditions generated during historical operation to form multiple strain data sets with consistent operating conditions; calculating statistical descriptive quantities of the strain sequences in terms of spatial distribution and temporal evolution characteristics for each strain data set; and constructing a reference model based on the statistical descriptive quantities to characterize the typical evolutionary behavior of the strain sequences under the corresponding operating conditions.

[0033] Furthermore, a consistency analysis is performed between the current strain sequence and a reference model under the corresponding operating conditions to generate reliability evaluation parameters. This includes: selecting a reference model that matches the current operating conditions within the current operating period; calculating the deviation between the current strain sequence and the reference model in spatial distribution characteristics to obtain a spatial consistency evaluation result; calculating the deviation between the current strain sequence and the reference model in temporal evolution characteristics to obtain a temporal consistency evaluation result; and generating reliability evaluation parameters based on the spatial consistency evaluation result and the temporal consistency evaluation result to characterize the reliability of the current strain sequence under the current operating conditions.

[0034] In this embodiment of the invention, after the construction of the operating condition background is completed, the historical strain sequences are further grouped and modeled based on the operating condition background to construct a reference model for subsequent consistency analysis, and on this basis, consistency analysis is performed on the current strain sequence to generate credibility evaluation parameters.

[0035] During historical operation, the historical strain sequences are screened for consistency under various established operating conditions. Specifically, the historical strain sequences are divided according to the operating conditions to which the corresponding time period belongs, and only strain data whose similarity in the characteristic quantities of the operating conditions meets the preset consistency requirements are retained, thus forming multiple strain data sets with consistent operating conditions. Each strain data set corresponds to a specific operating condition background, used to describe the typical evolution behavior of battery surface strain under that operating condition.

[0036] For each strain data set operating under consistent conditions, spatial distribution morphology analysis and temporal evolution characteristic analysis are performed on the strain sequences. Spatially, the strain sequences are represented as... ,in This indicates the background category of the operating conditions. By statistically analyzing the strain distribution at different spatial locations within the same time section, feature vectors characterizing the spatial distribution pattern are extracted. In this embodiment, the spatial distribution morphology features include not only strain amplitude information, but also spatial gradient and spatial fluctuation characteristics, and its construction form can be expressed as follows:

[0037] in Indicates the background of operating conditions The mean strain characteristics at this spatial location, This indicates the corresponding spatial fluctuation characteristics.

[0038] In the time dimension, time evolution characteristic statistics are performed on strain sequences under the same operating conditions to form a time feature vector describing the law of strain change over time. The time evolution characteristics are used to reflect the stability and changing trend of the strain sequence under the operating conditions, and their construction form can be expressed as:

[0039] in This represents the average evolution level of strain over a time interval. Indicates time fluctuation characteristics, Indicates spatial location In time The strain value at any given moment.

[0040] In obtaining spatial distribution morphological characteristics and temporal evolution characteristics Then, the two are combined to construct a reference model for characterizing the typical evolutionary behavior of strain sequences under corresponding operating conditions. The reference model is not a single scalar, but a multi-dimensional model structure composed of spatial and temporal features, represented as follows: By using the above method, a one-to-one correspondence between the operating conditions and the reference model was established, providing a clear benchmark for subsequent consistency analysis.

[0041] Within the current runtime segment, a reference model matching the runtime environment corresponding to the current strain sequence is first selected. Subsequently, the current strain sequence is represented as... Consistency analysis was performed with the selected reference model in both spatial and temporal dimensions.

[0042] In the spatial dimension, the spatial consistency evaluation result is obtained by calculating the deviation between the spatial distribution characteristics of the current strain sequence and the spatial characteristics of the reference model. Its calculation form can be expressed as:

[0043] in Indicates the current strain sequence in space. Spatial characteristics of the location Indicates spatial location The spatial characteristics of the reference model at that location.

[0044] In the time dimension, the time consistency evaluation result is obtained by calculating the deviation between the time evolution characteristics of the current strain sequence and the time characteristics of the reference model. Its calculation form can be expressed as:

[0045] in This indicates the time evolution characteristics of the current strain sequence. This represents the temporal characteristics of the reference model.

[0046] The time consistency evaluation results are used to reflect the degree of deviation between the current strain sequence and historical typical behavior in terms of the time evolution pattern.

[0047] After obtaining the spatial consistency evaluation results and the temporal consistency evaluation results, the two are fused to generate a reliability evaluation parameter that characterizes the reliability of the current strain sequence under the current operating conditions. .

[0048] Through the above processing flow, under the constraints of the operating conditions, historical strain sequences are grouped and modeled, and a reference model is constructed. Furthermore, based on spatial consistency and temporal consistency analysis, credibility evaluation parameters are generated for the current strain sequence, providing quantifiable and traceable input basis for subsequent credibility compensation processing and bulge localization analysis.

[0049] Step S30: Perform confidence compensation processing on the current strain sequence based on the confidence evaluation parameters to generate a confidence-compensated equivalent strain sequence.

[0050] Specifically, based on the distribution of the credibility evaluation parameters in the spatial location dimension, the credibility weight corresponding to each spatial location in the current strain sequence is determined; the corresponding credibility weight is applied to the strain value of each spatial location in the current strain sequence to generate a credibility-weighted strain sequence; spatial aggregation processing is performed on the weighted strain sequence to weaken the strain fluctuations corresponding to low credibility spatial locations and retain the strain characteristics corresponding to high credibility spatial locations; the strain result after spatial aggregation processing is used as the equivalent strain sequence after credibility compensation.

[0051] Furthermore, the rules for performing spatial aggregation processing on the weighted strain sequence are as follows: along the spatial location dimension of the distributed optical fiber, for each spatial location, select neighborhood strain data centered on that spatial location and covering a preset spatial length; according to the confidence weight of the neighborhood strain data, perform weighted aggregation calculation on the neighborhood strain data to increase the proportion of strain data with higher confidence weight in the aggregation result; use the result obtained from the weighted aggregation calculation as the aggregated strain value at the corresponding spatial location; use the aggregated strain values ​​corresponding to each spatial location to form a spatially continuous strain distribution result, which is used as the spatial aggregation processing result.

[0052] In this embodiment of the invention, after obtaining the credibility evaluation parameters, credibility compensation processing is further performed on the current strain sequence based on the credibility evaluation parameters to generate a credibility-compensated equivalent strain sequence, thereby providing stable data input for subsequent bulge feature recognition and spatial positioning analysis.

[0053] The current strain sequence is represented as ,in Indicates a spatial location index along a distributed optical fiber. This represents the time index within the current analysis period. The credibility evaluation parameters are generated from the aforementioned consistency analysis process and can be expressed as follows: This is used to characterize the reliability of strain data analysis at different spatial locations under current operating conditions. The reliability evaluation parameters are distributed along the spatial location dimension to perform spatial differentiation processing on the current strain sequence.

[0054] Based on the distribution of credibility evaluation parameters along the spatial dimension, the credibility weight corresponding to each spatial location in the current strain sequence is first determined. Specifically, the credibility evaluation parameters are... Mapped to a confidence weight function for numerical computation ,in and Maintaining a monotonic correspondence is used to reflect the degree of participation of strain data from different spatial locations in subsequent compensation processing. In this embodiment, the confidence weighting function can be directly composed of confidence evaluation parameters, i.e. Alternatively, the credibility evaluation parameters can be obtained by normalizing or interval mapping according to engineering requirements, but this does not affect the execution logic of subsequent compensation processing.

[0055] After determining the confidence weight for each spatial location, the corresponding confidence weight is applied to the strain value at each spatial location in the current strain sequence, generating a confidence-weighted strain sequence. The weighted strain sequence can be represented as:

[0056] Through the above processing, the spatial locations with higher reliability retain their original response characteristics in the strain sequence, while the strain response corresponding to the spatial locations with lower reliability is numerically suppressed, thereby reducing the interference of low-reliability data on subsequent analysis at the data level.

[0057] Obtaining the weighted strain sequence Subsequently, spatial aggregation processing is performed on the weighted strain sequence. The purpose of spatial aggregation processing is not to change the physical meaning of the strain sequence, but to use statistical fusion within the spatial neighborhood to ensure that the strain results at local spatial locations are simultaneously constrained by the confidence distribution of adjacent spatial locations, thereby forming a spatially continuous and stable strain distribution result.

[0058] Specifically, along the spatial location dimension of the distributed optical fiber, for each spatial location Select a neighborhood spatial interval centered on the spatial location and covering a preset spatial length. The neighborhood space interval Determined by a preset spatial length, it is used to include several adjacent strain sampling points near the spatial location. The setting of the neighborhood spatial interval is used to ensure that the spatial aggregation processing has local characteristics and avoid unnecessary interference from strain responses at long distances to the current spatial location.

[0059] In determining the neighborhood space interval Then, the corresponding weighted strain data within the neighborhood spatial interval is extracted. And simultaneously obtain the credibility weights corresponding to each spatial location within the neighborhood spatial interval. Based on this, a weighted aggregation calculation is performed on the neighborhood strain data according to the confidence weight corresponding to the neighborhood strain data, so that the strain data with higher confidence weight occupies a higher proportion in the aggregation result.

[0060] In this embodiment, spatial aggregation calculation can be performed using a weighted average method, and the calculation process can be expressed as follows:

[0061] in, Indicates spatial location The aggregated strain value is obtained after spatial aggregation processing. Through this calculation method, strain responses with higher confidence in the neighborhood have a more significant impact on the aggregation result, while strain responses with lower confidence in the neighborhood are naturally weakened during the aggregation process.

[0062] After completing the above spatial aggregation calculations at all spatial locations along the distributed optical fiber, the aggregated strain values ​​corresponding to each spatial location are... The strains are combined in spatial order to form a spatially continuous strain distribution. This spatially continuous strain distribution is the output of the spatial aggregation process, reflecting the overall strain distribution pattern under confidence constraints. The strain result after spatial aggregation is used as the equivalent strain sequence after confidence compensation, denoted as... ， The equivalent strain sequence after credibility compensation maintains the spatial distribution structure of the original strain sequence while introducing constraints on the spatial location dimension of the credibility evaluation parameters, so that subsequent bulge feature recognition and spatial positioning analysis are performed based on the credibility-compensated data.

[0063] Through the aforementioned credibility compensation process, spatial location differentiation compensation and neighborhood aggregation constraints are achieved for the current strain sequence without introducing additional sensing information or changing the strain data acquisition method, providing a stable, continuous, and controllable data input foundation for subsequent bulge localization analysis.

[0064] Step S40: Perform bulge feature recognition and spatial localization analysis based on the equivalent variation sequence to output bulge localization results.

[0065] Specifically, the set of spatial locations in the equivalent strain sequence where the strain amplitude exceeds a preset threshold is identified to form candidate bulge response intervals; strain peak characteristics and strain accumulation characteristics are calculated for the equivalent strain sequence within each candidate bulge response interval to form bulge feature parameters for characterizing the intensity of the bulge response; the candidate bulge response intervals are effectively screened based on the bulge feature parameters to determine the target bulge interval that meets the preset bulge judgment conditions; and the spatial location of the target bulge interval in the equivalent strain sequence is output as the spatial positioning result of the battery bulge.

[0066] In this embodiment of the invention, after obtaining the equivalent strain sequence after confidence compensation, bulge feature recognition and spatial positioning analysis are further performed based on the equivalent strain sequence to output the bulge positioning result of the corresponding battery. The equivalent strain sequence is obtained by the aforementioned confidence compensation processing and is used to reflect the strain distribution state of the battery surface along the distributed optical fiber under confidence constraints.

[0067] Specifically, the equivalent variation sequence is represented as ,in Indicates a spatial location index along a distributed optical fiber. This represents the time index within the current analysis period. Based on the equivalent strain sequence, the strain amplitude corresponding to each spatial location is analyzed in a traversal manner to identify spatial locations where the strain amplitude exceeds a preset discrimination threshold. The preset discrimination threshold is used to distinguish between normal strain fluctuations and abnormal strain responses that may be related to bulging; its value can be set based on engineering experience or historical data.

[0068] After identifying spatial locations where the strain amplitude exceeds a preset threshold, adjacent or consecutive spatial locations are further merged based on spatial adjacency to form several candidate bulge response intervals. Each candidate bulge response interval corresponds to a continuous spatial interval in the equivalent strain sequence, representing the region where a bulge response may exist. This method avoids misjudgment of single-point anomalies, allowing subsequent analysis to be based on strain responses with spatial continuity.

[0069] For each candidate bulge response interval, feature extraction processing is performed on the equivalent strain sequence within that interval. Specifically, the strain peak feature within the candidate bulge response interval is calculated to characterize the maximum strain response level within that interval; simultaneously, the strain cumulative feature within the candidate bulge response interval is calculated to characterize the overall strain response intensity in the spatial dimension within that interval. The strain peak feature and the strain cumulative feature together constitute the bulge feature parameters used to characterize the bulge response intensity.

[0070] After obtaining the bulge characteristic parameters corresponding to each candidate bulge response interval, the candidate bulge response intervals are filtered for validity based on preset bulge determination criteria. The bulge determination criteria may include threshold constraints or combined discrimination rules for strain peak characteristics and strain accumulation characteristics, used to exclude candidate intervals with insufficient response intensity or spatial distribution characteristics that do not conform to bulge characteristics. Through the above filtering process, target bulge intervals that meet the preset bulge determination criteria are determined.

[0071] Finally, the spatial location range of the target bulge area in the equivalent strain sequence is output as the spatial positioning result of the battery bulge. The spatial positioning result is used to indicate the spatial location area of ​​the bulge on the battery surface and can serve as the basis for subsequent alarms, maintenance, or further analysis.

[0072] In another possible implementation, within the current analysis period, the equivalent strain sequence is divided into multiple local spatial segments according to a preset spatial length, and a normalized strain morphology description vector is extracted for each spatial segment to characterize the relative morphological features of the strain distribution within that segment.

[0073] Furthermore, the spatial morphological description vectors obtained in the current time period are compared with the set of spatial morphological description vectors formed under the same operating conditions in historical operations to identify morphological segments that repeatedly appear in multiple operations and have relatively stable spatial positions. For spatial segments that appear only once in the current time period but lack similar morphological support in historical operations, their participation weight in the bulge determination is reduced or they are not considered as valid bulge candidate intervals.

[0074] like Figure 2 As shown, this invention provides an optical fiber-guided battery bulge localization system. The system includes: a data acquisition unit 201, used to acquire strain data from distributed optical fibers deployed on the battery surface to form a strain sequence distributed along spatial locations, and to acquire battery operating state parameters aligned with the time of the strain sequence to construct an operating condition background; a processing unit 202, used to group and model historical strain sequences based on the operating condition background to construct a reference model, and to perform consistency analysis between the current strain sequence and the reference model under the corresponding operating condition background to generate a reliability evaluation parameter; a compensation unit 203, used to perform reliability compensation processing on the current strain sequence based on the reliability evaluation parameter to generate a reliability-compensated equivalent strain sequence; and a result output unit 204, used to perform bulge feature recognition and spatial positioning analysis based on the equivalent strain sequence to output the bulge localization result.

[0075] The present invention also provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the aforementioned fiber-optic guided battery bulge positioning method.

[0076] 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 fiber-optic guided battery bulge positioning method as described above.

[0077] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 3 As 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 fiber-optic guided battery bulge positioning method.

[0078] 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.

[0079] 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.

[0080] 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 fiber-optic guided method for locating battery bulges, characterized in that, The method includes: Distributed optical fiber strain data deployed on the battery surface are acquired to form a strain sequence distributed along the spatial location, and battery operating state parameters aligned with the time of the strain sequence are acquired to construct the operating condition background. Based on the aforementioned operating conditions, historical strain sequences are grouped and modeled to construct a reference model, and the consistency analysis between the current strain sequence and the reference model under the corresponding operating conditions is performed to generate credibility evaluation parameters. Based on the aforementioned reliability evaluation parameters, a reliability compensation process is performed on the current strain sequence to generate a reliability-compensated equivalent strain sequence. Based on the equivalent variation sequence, perform bulge feature recognition and spatial positioning analysis to output bulge positioning results.

2. The fiber-optic guided battery bulge positioning method according to claim 1, characterized in that, Strain data from distributed optical fibers deployed on the battery surface are acquired to form a strain sequence distributed along spatial locations. Battery operating state parameters, time-aligned with the strain sequence, are then acquired to construct an operating condition background, including: Based on the change process of battery operating state parameters over time, the time interval of battery operating state changes during operation is determined. Within each time interval where the operating state changes, extract the strain sequence segment from the distributed optical fiber strain data that completely overlaps with that time interval; For each strain sequence segment, spatial consistency calculation is performed to generate spatial statistical results that characterize the spatial distribution features of the strain sequence segment. Perform time stability calculations on each strain sequence segment to generate time statistics that characterize the time evolution of the strain sequence segment; Based on the spatial and temporal statistical results within the corresponding time interval, an operational condition characterization quantity is generated to describe the operational state characteristics of that time interval, and an operational condition background is constructed based on the operational condition characterization quantity.

3. The fiber-optic guided battery bulge positioning method according to claim 2, characterized in that, Based on the aforementioned operating conditions, historical strain sequences are grouped and modeled to construct a reference model, including: Based on the background of operating conditions generated during historical operation, the historical strain sequences are screened for consistency of operating conditions to form multiple strain data sets with consistent operating conditions. For each strain data set, the statistical descriptive quantities of the strain sequence in terms of spatial distribution and temporal evolution characteristics are calculated. Based on the statistical descriptors, a reference model is constructed to characterize the typical evolutionary behavior of strain sequences under corresponding operating conditions.

4. The fiber-optic guided battery bulge positioning method according to claim 3, characterized in that, Consistency analysis is performed between the current strain sequence and the reference model under corresponding operating conditions to generate reliability evaluation parameters, including: Within the current runtime segment, select a reference model that matches the current runtime conditions. Calculate the deviation between the current strain sequence and the reference model in terms of spatial distribution characteristics to obtain the spatial consistency evaluation result; The deviation between the current strain sequence and the reference model in terms of time evolution characteristics is calculated to obtain the time consistency evaluation result; Based on the spatial consistency evaluation results and the temporal consistency evaluation results, a reliability evaluation parameter is generated to characterize the reliability of the current strain sequence under the current operating conditions.

5. The fiber-optic guided battery bulge positioning method according to claim 1, characterized in that, Based on the aforementioned reliability evaluation parameters, reliability compensation processing is performed on the current strain sequence to generate a reliability-compensated equivalent strain sequence, including: Based on the distribution of the credibility evaluation parameters in the spatial location dimension, the credibility weights corresponding to each spatial location in the current strain sequence are determined. Apply corresponding confidence weights to the strain values ​​at each spatial location in the current strain sequence to generate a confidence-weighted strain sequence; Spatial aggregation processing is performed on the weighted strain sequence to reduce strain fluctuations corresponding to low-confidence spatial locations and retain strain characteristics corresponding to high-confidence spatial locations. The strain results after spatial aggregation are used as the equivalent strain sequence after confidence compensation.

6. The fiber-optic guided battery bulge positioning method according to claim 5, characterized in that, The rules for performing spatial aggregation processing on the weighted strain sequence are as follows: Along the spatial location dimension of the distributed optical fiber, for each spatial location, neighborhood strain data centered on that spatial location and covering a preset spatial length are selected; Based on the confidence weight of the neighborhood strain data, a weighted aggregation calculation is performed on the neighborhood strain data to increase the proportion of strain data with higher confidence weight in the aggregation result; The result obtained from the weighted aggregation calculation is used as the aggregation strain value at the corresponding spatial location; The spatially continuous strain distribution result is constructed using the aggregated strain values ​​corresponding to each spatial location, which serves as the spatial aggregated processing result.

7. The fiber-optic guided battery bulge positioning method according to claim 1, characterized in that, Based on the equivalent variation sequence, perform bulge feature recognition and spatial localization analysis to output bulge localization results, including: In the equivalent strain sequence, identify the set of spatial locations where the strain amplitude exceeds a preset discrimination threshold to form candidate bulge response intervals; For each candidate bulge response interval, the strain peak characteristics and strain cumulative characteristics are calculated to form bulge characteristic parameters used to characterize the intensity of the bulge response; Based on the bulge feature parameters, the candidate bulge response intervals are effectively screened to determine the target bulge intervals that meet the preset bulge judgment conditions. The spatial location of the target bulge region in the equivalent strain sequence is output as the spatial localization result of the battery bulge.

8. A fiber-optic guided battery bulge positioning system, characterized in that, The system includes: The acquisition unit is used to acquire distributed optical fiber strain data deployed on the battery surface to form a strain sequence distributed along the spatial location, and to acquire battery operating state parameters that are time-aligned with the strain sequence to construct the operating condition background. The processing unit is used to group and model historical strain sequences based on the operating conditions background to construct a reference model, and to perform consistency analysis between the current strain sequence and the reference model under the corresponding operating conditions background to generate credibility evaluation parameters. The compensation unit is used to perform confidence compensation processing on the current strain sequence based on the confidence evaluation parameters to generate a confidence-compensated equivalent strain sequence; The result output unit is used to perform bulge feature recognition and spatial positioning analysis based on the equivalent variation sequence to output the bulge positioning result.

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 fiber-optic guided battery bulge positioning 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 fiber-optic guided battery bulge positioning method as described in any one of claims 1-7.

Images