A battery energy storage state monitoring and evaluation system and method

Abstract

The present application relates to the technical field of battery health state evaluation, and particularly relates to a battery energy storage state monitoring and evaluation system and method. The method comprises the following steps: collecting structural vibration response of the battery to obtain a set of frequency response characteristics of measuring points and a vibration modal distribution map; calculating mechanical impedance spectrum of the vibration modal distribution map according to the set of frequency response characteristics of measuring points to obtain a battery vibration response spectrum; identifying a resonance characteristic table from the battery vibration response spectrum; calculating a modal consistency index according to the battery vibration response spectrum based on the resonance characteristic table to obtain the modal consistency index; constructing a structural health degree vector according to the modal consistency index; and optimizing dielectric detection parameters according to the structural health degree vector to obtain a regional detection scheme. The present application realizes fine spatial mapping and accurate echelon utilization rating of the internal state of the battery by combining structural vibration response analysis and regional dielectric characteristic detection.

Description

Technical Field

[0001] This invention relates to the field of battery health status assessment technology, and in particular to a battery energy storage status monitoring and assessment system and method. Background Technology

[0002] Traditional battery energy storage state monitoring and assessment methods can only obtain overall battery performance parameters, such as total capacity and average internal resistance, and cannot distinguish the state differences between different internal regions. This results in low sensitivity to detect localized degradation or anomalies, especially when the overall battery performance has not yet significantly degraded but local defects already exist. These methods often fail to detect potential problems in a timely manner. Existing technologies lack the ability to accurately locate anomalies within the battery. Even if an anomaly is detected, it is difficult to determine its specific location and extent within the battery. This lack of spatial resolution makes it impossible to specifically assess the impact of anomalies on battery performance and safety, and also fails to provide a basis for accurate tiered utilization and grading. The problems of insufficient detection accuracy, inability to effectively identify internal structural anomalies, and lack of spatial resolution in existing technologies urgently need to be addressed. Summary of the Invention

[0003] Therefore, it is necessary to provide a battery energy storage state monitoring and evaluation system and method to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, a battery energy storage state monitoring and evaluation method includes the following steps: Step S1: Collect the structural vibration response of the battery to obtain the frequency response characteristic set and vibration mode distribution map of the measurement points; calculate the mechanical impedance spectrum of the vibration mode distribution map based on the frequency response characteristic set of the measurement points to obtain the battery vibration response spectrum; Step S2: Identify the resonance feature table from the battery vibration response spectrum; calculate the modal consistency index of the battery vibration response spectrum based on the resonance feature table to obtain the modal consistency index; construct the structural health vector based on the modal consistency index; Step S3: Optimize dielectric detection parameters based on structural health vector to obtain a regional detection scheme; collect multi-point dielectric response data based on the regional detection scheme to obtain a dielectric response dataset; perform dielectric parameter inversion calculation on the dielectric response dataset to obtain a regional dielectric parameter set; perform dielectric property spatial mapping and anomaly identification on the regional dielectric parameter set to obtain a local dielectric property map. Step S4: Perform structure-material correlation analysis on the structural health vector and local dielectric property spectrum to obtain failure mode identification results; perform battery energy storage health state mapping based on failure mode identification results to obtain local health index distribution; perform tiered utilization rating comprehensive based on local health index distribution to obtain tiered utilization evaluation index.

[0005] Preferably, the present invention also provides a battery energy storage state monitoring and evaluation system for performing the battery energy storage state monitoring and evaluation method described above, the battery energy storage state monitoring and evaluation system comprising: The structural vibration response module is used to acquire the structural vibration response of the battery, obtain the frequency response characteristic set of the measurement points and the vibration mode distribution map; based on the frequency response characteristic set of the measurement points and the vibration mode distribution map, the mechanical impedance spectrum is calculated to obtain the battery vibration response spectrum; The modal health assessment module is used to identify resonance feature tables from the battery vibration response spectrum; calculate modal consistency index based on the resonance feature tables to obtain the modal consistency index; and construct a structural health vector based on the modal consistency index. The dielectric property detection module is used to optimize dielectric detection parameters based on the structural health vector to obtain a regional detection scheme; to collect multi-point dielectric response data based on the regional detection scheme to obtain a dielectric response dataset; to perform dielectric parameter inversion calculation on the dielectric response dataset to obtain a regional dielectric parameter set; and to perform dielectric property spatial mapping and anomaly identification on the regional dielectric parameter set to obtain a local dielectric property map. The failure diagnosis and rating module is used to perform structure-material correlation analysis on the structural health vector and local dielectric property spectrum to obtain failure mode identification results; based on the failure mode identification results, the battery energy storage health status is mapped to obtain the local health index distribution; based on the local health index distribution, the tiered utilization rating is comprehensively performed to obtain the tiered utilization evaluation index.

[0006] This invention achieves non-destructive testing of the internal structure of batteries by applying precisely controlled weak broadband vibration excitation (amplitude controlled below 0.1g) and acquiring vibration response data with high spatial resolution (e.g., using a scanning laser Doppler vibrometer to measure densely packed points on the battery surface). Compared to traditional overall performance testing, this method can excite and capture minute vibration modes closely related to the internal structure of the battery, such as electrodes, separators, and encapsulation layers. Based on this detailed health status information, it can more accurately predict key performance parameters of batteries in secondary use (such as capacity retention, internal resistance, and safety risk level). By integrating local health index distribution, key area influences, and performance prediction results, it can accurately classify used batteries for secondary use, providing instructive evaluation indices and usage recommendations. This effectively avoids using batteries with safety hazards in high-demand scenarios or prematurely scrapping batteries with high remaining value, achieving accurate evaluation, safe utilization, and resource maximization of used batteries. Attached Figure Description

[0007] Figure 1 This is a flowchart illustrating the steps of a battery energy storage state monitoring and evaluation method.

[0008] Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S2. Figure 3 This is a schematic diagram of the internal structure of an energy storage battery; The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0009] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0010] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0011] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0012] To achieve the above objectives, please refer to Figures 1 to 3 The battery energy storage state monitoring and evaluation method includes the following steps: Step S1: Collect the structural vibration response of the battery to obtain the frequency response characteristic set and vibration mode distribution map of the measurement points; calculate the mechanical impedance spectrum of the vibration mode distribution map based on the frequency response characteristic set of the measurement points to obtain the battery vibration response spectrum; Step S2: Identify the resonance feature table from the battery vibration response spectrum; calculate the modal consistency index of the battery vibration response spectrum based on the resonance feature table to obtain the modal consistency index; construct the structural health vector based on the modal consistency index; Step S3: Optimize dielectric detection parameters based on structural health vector to obtain a regional detection scheme; collect multi-point dielectric response data based on the regional detection scheme to obtain a dielectric response dataset; perform dielectric parameter inversion calculation on the dielectric response dataset to obtain a regional dielectric parameter set; perform dielectric property spatial mapping and anomaly identification on the regional dielectric parameter set to obtain a local dielectric property map. Step S4: Perform structure-material correlation analysis on the structural health vector and local dielectric property spectrum to obtain failure mode identification results; perform battery energy storage health state mapping based on failure mode identification results to obtain local health index distribution; perform tiered utilization rating comprehensive based on local health index distribution to obtain tiered utilization evaluation index.

[0013] In one embodiment, structural vibration response acquisition of the battery under test is performed. The battery is fixed on a test fixture, and a controlled broadband small-amplitude vibration excitation is applied to the battery through a vibration excitation device. Multiple vibration measurement points are arranged at the battery casing and key structural locations, and the vibration response signals of each measurement point are acquired simultaneously. The acquired vibration response signals are converted from the time domain to the frequency domain, and the frequency response characteristic parameters corresponding to each measurement point are extracted to form a set of frequency response characteristics for each measurement point. Then, based on the distribution relationship of the frequency response characteristics of each measurement point in the spatial location of the battery, a vibration mode distribution map of the entire battery is constructed. On this basis, mechanical impedance spectrum calculation is performed on the vibration mode distribution map based on the set of frequency response characteristics for each measurement point, converting the vibration response into structural impedance characteristics, and obtaining the battery vibration response spectrum reflecting the overall structural state of the battery.

[0014] Structural health analysis is performed on the battery vibration response spectrum to identify the position, amplitude, and bandwidth parameters of each resonance peak, forming a resonance characteristic table. Based on this table, the response consistency between different modes is calculated to obtain a modal consistency index, which characterizes the overall coordination of the battery structure under different modes. Furthermore, a structural health vector is constructed based on the changes in the modal consistency index to describe the health level distribution of the battery under its current structural state in a multidimensional form.

[0015] The dielectric detection process is adaptively optimized based on the structural health vector. According to the degree of anomaly reflected in the structural health vector, regions with anomalies within the battery are located and prioritized, generating corresponding regional detection schemes. Following these regional detection schemes, multiple dielectric probes are deployed in the corresponding regions of the battery, and synchronous dielectric response data is acquired using matched sweep frequency parameters to obtain a dielectric response dataset. Subsequently, dielectric parameter inversion calculations are performed on the dielectric response dataset. By establishing a dielectric propagation model and iteratively solving the parameters, the dielectric parameter set corresponding to each detection region is obtained. Then, the regional dielectric parameter sets are spatially mapped, and anomaly identification is performed using parameter offset features to generate local dielectric property maps.

[0016] By jointly analyzing structural and material information, and performing unified feature analysis on the structural health vector and local dielectric property spectrum, the correlation between structural response characteristics and dielectric parameter changes is established, and the corresponding failure mode identification results are obtained. Then, based on the failure mode identification results, the battery energy storage health status is spatially mapped, and the local health index distribution corresponding to each region is calculated. On this basis, the overall suitability of the battery for secondary use is comprehensively evaluated by combining the local health index distribution, and finally, a secondary use evaluation index is obtained to characterize the secondary use value of the battery.

[0017] Please refer to [link / reference needed] for further information. Figure 3 The battery in the figure consists of multiple layers of electrode materials and a separator stacked together. The external metal frame is fastened with bolts to maintain structural stability. The grid-like surface corresponds to the array arrangement of the internal electrodes. Bolts and interfaces are used to connect vibration excitation and dielectric detection sensors. The mechanical impedance change is collected and analyzed by vibration response. At the same time, multi-point dielectric detection is achieved with the help of external interfaces to obtain the spatial distribution of dielectric parameters of the internal materials, thereby supporting structure-material correlation analysis and battery health status assessment.

[0018] Preferably, the structural vibration response acquisition in step S1 includes: Design and generate a wideband vibration excitation signal with a frequency range of 10Hz to 5000Hz and an amplitude controlled below 0.1g to obtain the vibration excitation signal; The vibration excitation signal is controlled by the excitation signal to obtain the calibrated vibration excitation; Spatial grid response acquisition of the battery yields a time-domain dataset of vibration response. The vibration response time-domain dataset is preprocessed to obtain vibration response data; Frequency domain transformation and feature extraction are performed on the vibration response data to obtain the frequency response characteristic set of the measurement points; By performing response spectrum spatial mapping based on the frequency response characteristic set of the measurement points, the vibration mode distribution map is obtained.

[0019] In one embodiment, a broadband vibration excitation signal for battery structure detection is designed and generated. The frequency range of the vibration excitation signal covers 10Hz to 5000Hz to ensure that the battery structure can be excited in both low-frequency overall modes and high-frequency local modes. Simultaneously, the amplitude of the vibration excitation is controlled below 0.1g to avoid interference or damage to the normal structure and internal material state of the battery. The generated signal serves as the initial vibration excitation signal. The vibration excitation signal is controlled by an excitation control module, which calibrates the output amplitude, frequency stability, and phase consistency of the excitation signal to ensure consistency and repeatability throughout the detection process. This calibrated vibration excitation is then used to apply controlled vibration input to the battery. Simultaneously with the application of the calibrated vibration excitation, spatial grid response acquisition is performed on the battery. Multiple vibration measurement points are arranged on the battery casing surface and key structural locations according to a preset spatial grid. The response signals of each measurement point under vibration excitation are simultaneously acquired, resulting in a vibration response time-domain dataset reflecting the dynamic characteristics of the overall battery structure.

[0020] The vibration response time-domain dataset undergoes response signal preprocessing. This preprocessing includes denoising, DC drift removal, and amplitude normalization of the original vibration response signal to eliminate environmental interference and acquisition errors, yielding vibration response data for subsequent analysis. Frequency domain transformation and feature extraction are then performed on the vibration response data. By performing frequency domain analysis on the vibration response data, the amplitude-frequency and phase characteristic parameters corresponding to each measuring point are extracted, forming a set of measuring point frequency response characteristics. This set characterizes the differences in structural response at different spatial locations and frequencies. Spatial mapping of the response spectrum is then performed based on the measuring point frequency response characteristic set. The frequency response characteristic parameters of each measuring point are combined and mapped according to their spatial location on the battery to construct a vibration mode distribution map reflecting the overall structural vibration characteristics of the battery, providing fundamental data for subsequent mechanical impedance spectrum calculation and structural health analysis.

[0021] Preferably, the mechanical impedance spectrum calculation in step S1 includes: The frequency response characteristic set of the measuring points is converted from acceleration to displacement to obtain the frequency domain displacement response; Calculate the point impedance spectrum based on the frequency domain displacement response; Extract the impedance characteristic parameter set of the potential impedance spectrum; Based on the set of impedance characteristic parameters, the spatial impedance distribution of the vibration mode distribution map is calculated to obtain the spatial impedance distribution map. Calculate the impedance anomaly distribution map of the impedance spatial distribution map; The battery vibration response spectrum is generated based on the impedance anomaly distribution diagram and the frequency domain displacement response.

[0022] In one embodiment, the response data of each measuring point in the aforementioned frequency response characteristic set are converted from acceleration to displacement. Based on the correspondence between frequency domain signals, the acceleration frequency response data of each measuring point is converted into the corresponding frequency domain displacement response, so that the vibration response data can directly reflect the deformation characteristics of the structure at different frequencies. The point impedance spectrum corresponding to each measuring point is calculated based on the frequency domain displacement response. Combined with the vibration excitation parameters applied to the battery structure, the frequency domain displacement response and excitation input are calculated to obtain the mechanical impedance changes of each measuring point at different frequencies, forming a point impedance spectrum. Impedance characteristic parameters that can characterize structural state changes are extracted from the point impedance spectrum. These impedance characteristic parameters include impedance amplitude changes, impedance abrupt changes in the resonant frequency band, and impedance gradient changes, etc., and are used to construct a point impedance characteristic parameter set.

[0023] Based on the aforementioned impedance characteristic parameter set, spatial impedance distribution calculations are performed on the aforementioned vibration mode distribution map. The impedance characteristic parameters of each measuring point are mapped and weighted according to their spatial position within the battery structure to generate an impedance spatial distribution map reflecting the overall impedance distribution characteristics of the battery structure. Impedance anomaly analysis is then performed on the impedance spatial distribution map. By comparing the relative changes in impedance characteristic parameters in different regions, abnormal regions with significantly deviated impedance from the surrounding areas are identified, forming an impedance anomaly distribution map used to locate locations where structural stiffness or connection status changes. The impedance anomaly distribution map is then comprehensively processed with the corresponding frequency domain displacement response, correlating and integrating structural deformation characteristics with impedance anomaly features to generate a battery vibration response spectrum characterizing the overall dynamic response characteristics of the battery structure. This provides input for subsequent resonance feature identification and structural health assessment.

[0024] Preferably, step S2 includes: Step S21: Identify the peak characteristics of the battery vibration response spectrum to obtain a resonance characteristic table; Step S22: Compare the resonance feature table with the pre-established healthy battery reference feature library to obtain the resonance offset index set; Step S23: Extract the damping parameters of the battery based on the resonance characteristic table and the battery vibration response spectrum to obtain the structural damping characteristic set; Step S24: Calculate the modal consistency index of the battery vibration response spectrum based on the resonance characteristic table to obtain the modal consistency index; Step S25: Perform local anomaly detection on the battery based on the modal consistency index and resonance shift index set to obtain a local anomaly map; Step S26: Calculate the spectral integrity index of the local anomaly map; Step S27: Construct a structural health vector from the resonance offset index set and the spectral integrity index.

[0025] In one embodiment, the amplitude variation of the vibration response spectrum within a preset frequency range is analyzed to identify the position, peak amplitude, and peak width parameters of each major resonance peak. The identified resonance peak features are then organized in frequency order to form a resonance feature table characterizing the dynamic properties of the battery structure. This resonance feature table is compared with a pre-established healthy battery reference feature library. The reference feature library is constructed from vibration test data of multiple batteries in normal structural states. By comparing the resonance features of the current battery with the reference features item by item, the resonance frequency offset and amplitude variation are calculated to form a resonance offset index set reflecting the degree of structural change. Damping parameters are extracted from the battery based on the resonance feature table and the battery vibration response spectrum. Using the attenuation characteristics and bandwidth variations of the resonance peaks, the damping characteristics corresponding to different modes are estimated, resulting in a structural damping characteristic set describing the energy dissipation characteristics of the battery structure.

[0026] Modal consistency indices are calculated for the battery vibration response spectrum based on the resonance characteristic table. By comprehensively analyzing the response morphology, frequency correspondence, and amplitude variation trends under different modes, a modal consistency index reflecting the degree of coordination between modes is calculated to determine whether the overall battery structure maintains stable modal distribution characteristics. Local anomalies are detected in the battery based on the modal consistency index and resonance shift index set. The modal consistency index and resonance shift index are correlated spatially to identify regions where both modal anomalies and significant resonance shifts exist simultaneously, generating local anomaly maps to describe the distribution of structural anomalies. Spectral integrity analysis is performed on these local anomaly maps, and corresponding spectral integrity indices are calculated. By evaluating the continuity and integrity of the spectral energy distribution within the anomaly region, the persistence and structural relevance of the anomaly are determined. A structural health vector is constructed by comprehensively integrating the resonance shift index set and the spectral integrity index. This structural health vector describes the battery's health status under different structural regions and modal conditions in a multi-dimensional parameter form, providing input for subsequent dielectric detection parameter optimization and material state analysis.

[0027] Most importantly, the modal consistency index is calculated as follows: Vibration phase space mapping was performed on the resonance characteristic table and the battery vibration response spectrum to obtain the vibration space distribution map; Remodeling state shape diagram based on vibration spatial distribution map; Calculate the modal complexity parameter set based on the vibration spatial distribution map; Symmetry and nodal line analysis were performed on the modal shape diagram and complexity parameter set to obtain structural characteristic indices; A comprehensive consistency evaluation of the complexity parameter set and structural feature indices is performed to obtain the modal consistency index.

[0028] In one embodiment, based on the aforementioned resonance characteristic table and battery vibration response spectrum, the vibration response of the battery structure is spatially mapped using vibration phase. Combining the phase information and amplitude distribution of each measuring point at different resonance frequencies, the vibration response is mapped to the corresponding spatial location of the battery structure, generating a vibration spatial distribution map reflecting the structural vibration state at different frequencies. Based on this vibration spatial distribution map, the modal shape of the battery is reconstructed. By combining the relative relationships of the vibration responses at each measuring point under the same mode, the overall vibration morphology of the corresponding mode is reconstructed, forming a modal shape map describing the vibration characteristics of the battery structure. A modal complexity parameter set is calculated based on the vibration spatial distribution map. These modal complexity parameters characterize the degree of spatial variation of modal vibration, including differences in vibration amplitude in different regions, continuity of phase changes, and the degree of local vibration concentration, reflecting the complexity characteristics of the structural vibration morphology.

[0029] Based on this, symmetry and nodal line analysis are performed on the modal shape diagram and modal complexity parameter set. The consistency of the modal shape distribution along the symmetry axis of the battery structure, as well as the spatial stability and continuity of the vibration nodal lines, are analyzed, and structural feature indices reflecting structural integrity and connection status are extracted. A comprehensive consistency evaluation is performed on the modal complexity parameter set and structural feature indices. By comprehensively analyzing the complexity variation trends and structural feature stability under different modes, the consistency level between the modal responses is determined, resulting in a modal consistency index used to characterize the overall coordination of the battery structure.

[0030] Preferably, step S3, the optimization of dielectric detection parameters, includes: Based on the structural health vector, abnormal areas of the battery are located and prioritized to obtain a priority table of detection areas. Structural anomaly classification analysis is performed on the structural health vector to obtain an anomaly type distribution map; Based on the anomaly type distribution map, the material property correspondence is established to obtain the material-frequency sensitivity table; Generate a regional frequency scheme based on the material-frequency sensitivity table; Based on the regional frequency scheme, the shell penetration parameters are calculated to obtain the power configuration table; Determine the spatial resolution configuration based on the anomaly type distribution map; A regional detection scheme is generated based on the regional frequency scheme, power configuration table, and spatial resolution configuration.

[0031] In one embodiment, the battery is used to locate and prioritize abnormal regions based on the aforementioned structural health vector. The health values ​​corresponding to each region in the structural health vector are compared with preset thresholds. Regions with significantly reduced health are identified as potential abnormal regions. These regions are then ranked according to the magnitude of the health decrease and their importance within the overall battery structure, forming a priority table for subsequent detection. The structural health vector is then analyzed for structural anomaly classification. By analyzing the differences in modal consistency, resonance shift, and spectral integrity among different abnormal regions, structural anomalies are classified into different types, such as loose connections, local stiffness degradation, or structural integrity failure. The distribution of each anomaly type within the battery space is then summarized to obtain an anomaly type distribution map.

[0032] Based on the anomaly type distribution map, a correspondence between material properties is established. Combining this with the characteristics of different structural parts and material composition within the battery, the sensitivity of various structural anomalies to changes in dielectric parameters is analyzed. A correspondence between anomaly types and changes in material dielectric properties is established, forming a material-frequency sensitivity table to reflect the response characteristics of different materials and anomaly types to dielectric detection frequencies. Based on this material-frequency sensitivity table, regional frequency schemes are generated. Dielectric detection frequency bands matching the corresponding material properties are selected for different anomaly regions, enhancing the response capability to changes in the target material's state, thus forming regional frequency schemes for different regions.

[0033] Penetration parameters of the casing are calculated based on the regional frequency scheme. By combining the battery casing thickness, material characteristics, and selected detection frequency, the attenuation of the dielectric signal in the casing is estimated to determine the transmit power and receive sensitivity parameters that meet the penetration requirements, forming a corresponding power configuration table. The spatial resolution configuration for dielectric detection is determined based on the anomaly type distribution map. For areas with high anomaly severity or strong structural sensitivity, a higher spatial resolution is set; for areas with low anomaly severity, a relatively lower spatial resolution is used to balance detection accuracy and efficiency. A regional detection scheme is generated by comprehensively configuring the regional frequency scheme, power configuration table, and spatial resolution configuration.

[0034] Preferably, the multi-point dielectric response data acquisition in step S3 includes: Based on the regionalized detection scheme, the dielectric probe array is configured to obtain an optimized probe configuration; An adaptive frequency sweep signal set is generated based on optimized probe configuration and regionalized detection scheme; Synchronous multi-point dielectric response acquisition is performed based on the adaptive frequency sweep signal set to obtain the dielectric response dataset.

[0035] In one embodiment, a dielectric probe array is configured according to the aforementioned regionalized detection scheme. Based on the abnormal region location, spatial resolution requirements, and detection priority determined in the regionalized detection scheme, multiple dielectric probes are deployed at corresponding positions on the outer surface of the battery casing. The spacing, arrangement, and coverage of the probes are adjusted to ensure that the probe array spatially corresponds to the region to be detected, thereby obtaining an optimized probe configuration adapted to the target region. An adaptive sweep frequency signal set is generated based on the optimized probe configuration and the regionalized detection scheme. For the detection regions corresponding to different probes, matching sweep frequency signals are generated by combining the regional frequency scheme and power configuration parameters. The start and end frequencies, sweep step, and excitation timing of the sweep frequency are configured so that the sweep frequency signal used by each probe can adapt to the material characteristics and penetration requirements of its corresponding region, forming an adaptive sweep frequency signal set. Synchronous multi-point dielectric response acquisition is performed based on the adaptive sweep frequency signal set. Under unified time control, each dielectric probe is driven to perform dielectric excitation on the battery according to the corresponding sweep frequency signal, and the dielectric response signal received by each probe is acquired synchronously to ensure the consistency of data between different detection points in the time and frequency dimensions, and finally form a dielectric response dataset containing dielectric response information of multiple detection areas.

[0036] Preferably, step S3, the dielectric parameter inversion calculation, includes: The scattering parameters of the dielectric response dataset are preprocessed to obtain the calibrated scattering parameters; An electromagnetic propagation model was constructed using the calibration scattering parameters to obtain a layered propagation model. Initial parameter estimation is performed on the layered propagation model to obtain initial parameter estimates. The theoretical scattering parameters are obtained by performing forward propagation calculations on the initial parameter estimates and the layered propagation model. The theoretical scattering parameters are optimized through iterative parameter optimization to obtain the optimized dielectric parameters; Frequency response characteristics are obtained by fitting the optimized dielectric parameters with frequency characteristics. Based on the frequency response characteristics and optimized dielectric parameters, the layer thickness parameters of the layered propagation model are corrected to obtain the corrected layer thickness parameters. Material state analysis was performed on the modified layer thickness parameters and optimized dielectric parameters to obtain the regional dielectric parameter set.

[0037] In one embodiment, the original dielectric response signal is converted into a corresponding scattering parameter representation. The scattering parameters are then subjected to amplitude calibration, phase alignment, and noise suppression to eliminate the influence of probe differences, signal transmission loss, and environmental interference on the measurement results, yielding calibrated scattering parameters for subsequent modeling and analysis. An electromagnetic propagation model is constructed based on these calibrated scattering parameters. According to the battery's structural hierarchy and material distribution, the internal structure of the battery is abstracted as a propagation structure formed by multiple equivalent dielectric layers. Combining the propagation and reflection characteristics of dielectric waves in different dielectric layers, a corresponding layered propagation model is established to describe the propagation process of the dielectric signal within the battery. Initial parameter estimation is performed on the layered propagation model. Based on the battery's design information, structural size range, and the overall trend of the calibrated scattering parameters, the initial dielectric parameters and layer thickness ranges of each dielectric layer are reasonably set, forming initial parameter estimates for subsequent inversion calculations.

[0038] Forward propagation calculations are performed based on initial parameter estimates and a layered propagation model. Under given initial parameters, the propagation and reflection responses of the dielectric signal in each dielectric layer are calculated using this model to obtain corresponding theoretical scattering parameters, which are then compared with the calibrated scattering parameters obtained from actual measurements. The theoretical scattering parameters are then iteratively optimized. By continuously adjusting the dielectric parameters, the theoretical scattering parameters gradually approach the calibrated scattering parameters until the difference between them meets a preset convergence condition, thus obtaining optimized dielectric parameters that truly reflect the dielectric properties of the material. Frequency characteristic fitting is then performed on the optimized dielectric parameters. By analyzing the variation trend of the optimized dielectric parameters at different frequencies, a frequency response characteristic of the dielectric parameters as a function of frequency is constructed to reflect the stability and variation law of the material's dielectric behavior under broadband conditions. Based on the frequency response characteristic and the optimized dielectric parameters, the layer thickness parameters of the layered propagation model are corrected. Combining the propagation characteristics of the dielectric signal at different frequencies, the original layer thickness parameters are adjusted so that the layered propagation model can more accurately reflect the actual thickness distribution of each layer structure inside the battery, resulting in corrected layer thickness parameters. Finally, material state analysis is performed on the corrected layer thickness parameters and the optimized dielectric parameters. By comprehensively analyzing the changes in dielectric parameters and layer thickness in different regions, it is determined whether there are any abnormalities or degradations in the dielectric properties of the material. The analysis results are then summarized according to spatial location to obtain a set of regional dielectric parameters used to characterize the material state of each detection region.

[0039] Preferably, the layer thickness parameter correction includes: Extract the phase frequency slope from frequency response characteristics and optimized dielectric parameters; Calculate the propagation velocity within the layer based on optimized dielectric parameters; The initial estimated layer thickness is obtained by estimating the base thickness based on the propagation velocity and phase frequency slope within the layer. Based on the initial estimated layer thickness, the initial estimated layer thickness is verified by multiple frequencies to obtain the multi-frequency thickness set and thickness dispersion. The interface transition layer parameters are obtained by optimizing the dielectric parameters, multi-frequency thickness set, and thickness dispersion. The abnormal region is refined based on the transition layer parameters to obtain the refined thickness distribution; The corrected layer thickness parameters are generated based on the refined thickness distribution, transition layer parameters, and intralayer propagation velocity.

[0040] In one embodiment, the slope information of phase change with frequency is extracted from the obtained frequency response characteristics and optimized dielectric parameters. By analyzing the phase change trend of the dielectric response at different frequency points, the rate of phase change relative to frequency is calculated to characterize the phase accumulation characteristics generated during the propagation of the dielectric signal within the material. The propagation speed of the dielectric signal within the corresponding material layer is calculated based on the optimized dielectric parameters. Combined with the dielectric properties reflected by the optimized dielectric parameters, the propagation speed of the dielectric signal within the material layer is determined to establish the correspondence between propagation time and spatial distance. The basic thickness is estimated based on the propagation speed within the layer and the phase frequency slope. By mapping the propagation characteristics of the dielectric signal within the material layer to the phase change, the equivalent distance corresponding to the propagation of the dielectric signal within the material layer is estimated, resulting in the initial estimated layer thickness value for each material layer.

[0041] Based on this, the initial estimated layer thickness is verified using multi-frequency thickness verification. Multiple frequency points are selected to estimate the material layer thickness, and the thickness results obtained at different frequencies are compared to form a multi-frequency thickness set. Simultaneously, the thickness dispersion is calculated based on the degree of difference between the multi-frequency thickness results to evaluate the stability and reliability of the initial estimated layer thickness. Interface transition layer processing is performed based on optimized dielectric parameters, the multi-frequency thickness set, and the thickness dispersion. When significant dispersion exists in the thickness estimation results at different frequencies, it is determined that there are transition regions or interface inhomogeneities between the material layers. Transition layer parameters are introduced to correct the interface influence between material layers, resulting in transition layer parameters describing interface continuity. Abnormal regions are refined based on these transition layer parameters. Combining the spatial distribution of the transition layer parameters, local refinement analysis is performed on regions with abnormal changes in the initial estimated layer thickness, adjusting the thickness distribution corresponding to the abnormal regions to obtain a more refined thickness distribution result. A corrected layer thickness parameter is generated through comprehensive calculation based on the refined thickness distribution, transition layer parameters, and intralayer propagation velocity. The corrected layer thickness parameter can more accurately reflect the actual thickness of each material layer inside the battery and serves as an important input parameter for subsequent material state analysis and dielectric property evaluation.

[0042] Preferably, step S4 includes the following steps: Step S41: Perform a unified feature space transformation on the structural health vector and the local dielectric property map to obtain a unified feature dataset; Step S42: Perform structure-material correlation analysis on the unified feature dataset to obtain failure mode identification results; Step S43: Optimize the health feature weights based on the failure mode identification results and the unified feature dataset to obtain the optimized feature weight set; Step S44: Calculate the local health index distribution based on the optimized feature weight set and the unified feature dataset; Step S45: Conduct a key area impact assessment based on the local health index distribution to obtain the key area impact score; Step S46: Based on the impact score of key areas and the distribution of local health index, predict performance parameters to obtain performance prediction results; Step S47: Combine the distribution of local health index, the impact score of key areas and the performance prediction results to obtain the tiered utilization evaluation index.

[0043] In this embodiment, the structural health vector is composed of structural response data such as structural strain indices, vibration characteristic parameters, and equivalent stiffness changes obtained in the preceding detection steps. The local dielectric property map is formed by mapping the regional dielectric parameter set obtained in step S3 according to spatial location, with each map unit corresponding to a detection area in the structure. A unified feature space transformation is performed on the structural health vector and the local dielectric property map. The structural parameters in the structural health vector are spatially rearranged according to the detection area and aligned one-to-one with the dielectric parameters of the corresponding area. Simultaneously, normalization is performed on structural and dielectric parameters of different dimensions to map them to the same feature scale, ultimately forming a unified feature dataset containing structural features and material dielectric features. Based on the unified feature dataset, structure-material correlation analysis is performed. By statistically analyzing the correlation between changes in structural features and changes in dielectric properties within each detection area, the coupling mode between structural anomalies and dielectric parameter anomalies is identified, and different types of failure modes are distinguished accordingly, such as material degradation-dominated, structural damage-dominated, or structure-material coupled failure, thus obtaining the failure mode identification result.

[0044] Based on the identified failure modes, the weights of various health features in the unified feature dataset are optimized. For different failure modes, the weight coefficients of structural features or dielectric features sensitive to the mode are increased, while the weights of features with weaker correlation are decreased, forming an optimized feature weight set that matches the failure mode. The unified feature dataset is then weighted based on the optimized feature weight set to obtain the local health index distribution of each detection area. This local health index characterizes the overall health level of the structural and material states within the corresponding area, and its value reflects the degree of degradation in that area. The impact of the local health index distribution on key areas in the structure is assessed. By analyzing the spatial distribution, concentration, and interrelationships of low-health-index areas in the overall structure, key areas with a significant impact on overall performance are identified, and a corresponding impact score is generated for each key area. Based on the key area impact scores and the local health index distribution, the performance parameters of the structure are predicted. Using the health status of key areas as the primary input, combined with the health index of non-key areas, the overall load-bearing capacity, reliability indicators, or service life-related performance parameters of the structure are predicted, yielding performance prediction results.

[0045] The distribution of local health index, the impact score of key areas and the performance prediction results are comprehensively processed for tiered utilization rating. Based on the health level of each area and its impact on performance, the structure or equipment is graded and evaluated to form a tiered utilization assessment index that can be used for maintenance decision-making or tiered utilization judgment.

[0046] Of particular importance is the structure-material correlation analysis, which specifically includes: Anomaly region spatial mapping is performed on the unified feature dataset to obtain a dual anomaly distribution map; The anomaly overlap degree is calculated on the double anomaly distribution map to obtain the anomaly spatial correlation index; Based on the anomaly spatial correlation index, a parameter correlation matrix is ​​generated from the unified feature dataset to obtain the parameter correlation matrix; The parameter correlation matrix is ​​matched with a pre-established battery failure mode feature library to obtain a list of mode similarities. A multidimensional discriminant analysis was performed on the pattern similarity list to obtain the pattern discrimination score; The regional failure type map is determined based on the pattern discrimination score and the dual anomaly distribution map; Failure propagation risk analysis was performed on the regional failure type map to obtain failure mode identification results.

[0047] In this embodiment, the unified feature dataset includes structural health features divided by detection areas and dielectric property parameters of the corresponding areas, where each detection area has a clear spatial location identifier. Anomaly region spatial mapping processing is performed on the unified feature dataset. Based on the deviation of the structural health features and the change in dielectric parameters relative to the baseline state, it is determined whether each detection area belongs to a structurally abnormal region or a materially abnormal region. Regions that simultaneously meet the criteria for both structural and material abnormality are marked as dual-abnormal regions, and a dual-abnormality distribution map is generated according to their spatial location. Anomaly overlap is calculated on the dual-abnormality distribution map. By statistically analyzing the proportion of dual-abnormal regions in the overall detection area, the degree of spatial clustering, and the continuity characteristics between adjacent regions, the spatial overlap level between structurally abnormal regions and materially abnormal regions is calculated, thereby obtaining an anomaly spatial correlation index to characterize the spatial coupling strength between structural and material abnormalities.

[0048] A parameter correlation matrix is ​​generated from the unified feature dataset based on the anomaly spatial correlation index. Using the detection area as the sample unit, the correlation between structural feature parameters and dielectric parameters is calculated item by item. The correlation results are then weighted using the anomaly spatial correlation index to form a parameter correlation matrix reflecting the relationship between structural and material parameters. This parameter correlation matrix is ​​then matched with a pre-established battery failure mode feature library. This library stores templates of structural feature variation patterns, dielectric property variation characteristics, and their correlation relationships corresponding to various known failure modes. By calculating the similarity between the parameter correlation matrix and each failure mode template, a pattern similarity list containing multiple failure modes is obtained. Multidimensional discriminant analysis is performed on the pattern similarity list, comprehensively considering the similarity distribution of each failure mode across different parameter dimensions. A corresponding pattern discrimination score is generated for each failure mode to reflect the degree of matching between the current detection state and each failure mode. Based on the pattern discrimination score and the dual-anomaly distribution map, the failure type of each detection area is determined. Failure modes with high discrimination scores are prioritized, and combined with the spatial distribution of the dual-anomaly areas, each detection area is assigned a corresponding regional failure type, forming a regional failure type map. Finally, a failure propagation risk analysis is performed on the regional failure type map. Based on the spatial relationship between adjacent regions and the distribution of their failure types, the potential propagation paths and expansion trends of failures between regions are analyzed, thus obtaining the final failure mode identification results.

[0049] Preferably, the present invention also provides a battery energy storage state monitoring and evaluation system for performing the battery energy storage state monitoring and evaluation method described above, the battery energy storage state monitoring and evaluation system comprising: The structural vibration response module is used to acquire the structural vibration response of the battery, obtain the frequency response characteristic set of the measurement points and the vibration mode distribution map; based on the frequency response characteristic set of the measurement points and the vibration mode distribution map, the mechanical impedance spectrum is calculated to obtain the battery vibration response spectrum; The modal health assessment module is used to identify resonance feature tables from the battery vibration response spectrum; calculate modal consistency index based on the resonance feature tables to obtain the modal consistency index; and construct a structural health vector based on the modal consistency index. The dielectric property detection module is used to optimize dielectric detection parameters based on the structural health vector to obtain a regional detection scheme; to collect multi-point dielectric response data based on the regional detection scheme to obtain a dielectric response dataset; to perform dielectric parameter inversion calculation on the dielectric response dataset to obtain a regional dielectric parameter set; and to perform dielectric property spatial mapping and anomaly identification on the regional dielectric parameter set to obtain a local dielectric property map. The failure diagnosis and rating module is used to perform structure-material correlation analysis on the structural health vector and local dielectric property spectrum to obtain failure mode identification results; based on the failure mode identification results, the battery energy storage health status is mapped to obtain the local health index distribution; based on the local health index distribution, the tiered utilization rating is comprehensively performed to obtain the tiered utilization evaluation index.

[0050] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for monitoring and evaluating the state of battery energy storage, characterized in that, Includes the following steps: Step S1: Collect the structural vibration response of the battery to obtain the frequency response characteristic set and vibration mode distribution map of the measurement points; calculate the mechanical impedance spectrum of the vibration mode distribution map based on the frequency response characteristic set of the measurement points to obtain the battery vibration response spectrum; Step S2: Identify the resonance feature table from the battery vibration response spectrum; calculate the modal consistency index of the battery vibration response spectrum based on the resonance feature table to obtain the modal consistency index; construct the structural health vector based on the modal consistency index; Step S3: Optimize dielectric detection parameters based on the structural health vector to obtain a regionalized detection scheme; collect multi-point dielectric response data based on the regionalized detection scheme to obtain a dielectric response dataset; perform dielectric parameter inversion calculation on the dielectric response dataset to obtain a regional dielectric parameter set; perform dielectric property spatial mapping and anomaly identification on the regional dielectric parameter set to obtain a local dielectric property map; wherein, step S3, dielectric detection parameter optimization includes: Based on the structural health vector, abnormal areas of the battery are located and prioritized to obtain a priority table of detection areas. Structural anomaly classification analysis is performed on the structural health vector to obtain an anomaly type distribution map; Based on the anomaly type distribution map, the material property correspondence is established to obtain the material-frequency sensitivity table; Generate a regional frequency scheme based on the material-frequency sensitivity table; Based on the regional frequency scheme, the shell penetration parameters are calculated to obtain the power configuration table; Determine the spatial resolution configuration based on the anomaly type distribution map; A regionalized detection scheme is generated based on the regional frequency scheme, power configuration table, and spatial resolution configuration. Step S4: Perform structure-material correlation analysis on the structural health vector and local dielectric property spectrum to obtain failure mode identification results; perform battery energy storage health state mapping based on failure mode identification results to obtain local health index distribution; perform tiered utilization rating comprehensive based on local health index distribution to obtain tiered utilization evaluation index.

2. The battery energy storage state monitoring and evaluation method according to claim 1, characterized in that, The structural vibration response acquisition in step S1 includes: Design and generate a wideband vibration excitation signal with a frequency range of 10Hz to 5000Hz and an amplitude controlled below 0.1g to obtain the vibration excitation signal; The vibration excitation signal is controlled by the excitation signal to obtain the calibrated vibration excitation; Spatial grid response acquisition of the battery yields a time-domain dataset of vibration response. The vibration response time-domain dataset is preprocessed to obtain vibration response data; Frequency domain transformation and feature extraction are performed on the vibration response data to obtain the frequency response characteristic set of the measurement points; By performing response spectrum spatial mapping based on the frequency response characteristic set of the measurement points, the vibration mode distribution map is obtained.

3. The battery energy storage state monitoring and evaluation method according to claim 1, characterized in that, The mechanical impedance spectrum calculation in step S1 includes: The frequency response characteristic set of the measuring points is converted from acceleration to displacement to obtain the frequency domain displacement response; Calculate the point impedance spectrum based on the frequency domain displacement response; Extract the impedance characteristic parameter set of the potential impedance spectrum; Based on the set of impedance characteristic parameters, the spatial impedance distribution of the vibration mode distribution map is calculated to obtain the spatial impedance distribution map. Calculate the impedance anomaly distribution map of the impedance spatial distribution map; The battery vibration response spectrum is generated based on the impedance anomaly distribution diagram and the frequency domain displacement response.

4. The battery energy storage state monitoring and evaluation method according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Identify the peak characteristics of the battery vibration response spectrum to obtain a resonance characteristic table; Step S22: Compare the resonance feature table with the pre-established healthy battery reference feature library to obtain the resonance offset index set; Step S23: Extract the damping parameters of the battery based on the resonance characteristic table and the battery vibration response spectrum to obtain the structural damping characteristic set; Step S24: Calculate the modal consistency index of the battery vibration response spectrum based on the resonance characteristic table to obtain the modal consistency index; Step S25: Perform local anomaly detection on the battery based on the modal consistency index and resonance shift index set to obtain a local anomaly map; Step S26: Calculate the spectral integrity index of the local anomaly map; Step S27: Construct a structural health vector from the resonance offset index set and the spectral integrity index.

5. The battery energy storage state monitoring and evaluation method according to claim 1, characterized in that, Step S3, multi-point dielectric response data acquisition, includes: Based on the regionalized detection scheme, the dielectric probe array is configured to obtain an optimized probe configuration; An adaptive frequency sweep signal set is generated based on optimized probe configuration and regionalized detection scheme; Synchronous multi-point dielectric response acquisition is performed based on the adaptive frequency sweep signal set to obtain the dielectric response dataset.

6. The battery energy storage state monitoring and evaluation method according to claim 1, characterized in that, Step S3, the inversion calculation of dielectric parameters, includes: The scattering parameters of the dielectric response dataset are preprocessed to obtain the calibrated scattering parameters; An electromagnetic propagation model was constructed using the calibration scattering parameters to obtain a layered propagation model. Initial parameter estimation is performed on the layered propagation model to obtain initial parameter estimates. The theoretical scattering parameters are obtained by performing forward propagation calculations on the initial parameter estimates and the layered propagation model. The theoretical scattering parameters are optimized through iterative parameter optimization to obtain the optimized dielectric parameters; Frequency response characteristics are obtained by fitting the optimized dielectric parameters with frequency characteristics. Based on the frequency response characteristics and optimized dielectric parameters, the layer thickness parameters of the layered propagation model are corrected to obtain the corrected layer thickness parameters. Material state analysis was performed on the modified layer thickness parameters and optimized dielectric parameters to obtain the regional dielectric parameter set.

7. The battery energy storage state monitoring and evaluation method according to claim 6, characterized in that, Layer thickness parameter correction includes: Extract the phase frequency slope from frequency response characteristics and optimized dielectric parameters; Calculate the propagation velocity within the layer based on optimized dielectric parameters; The initial estimated layer thickness is obtained by estimating the base thickness based on the propagation velocity and phase frequency slope within the layer. Based on the initial estimated layer thickness, the initial estimated layer thickness is verified by multiple frequencies to obtain the multi-frequency thickness set and thickness dispersion. The interface transition layer parameters are obtained by optimizing the dielectric parameters, multi-frequency thickness set, and thickness dispersion. The abnormal region is refined based on the transition layer parameters to obtain the refined thickness distribution; The corrected layer thickness parameters are generated based on the refined thickness distribution, transition layer parameters, and intralayer propagation velocity.

8. The battery energy storage state monitoring and evaluation method according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: Perform a unified feature space transformation on the structural health vector and the local dielectric property map to obtain a unified feature dataset; Step S42: Perform structure-material correlation analysis on the unified feature dataset to obtain failure mode identification results; Step S43: Optimize the health feature weights based on the failure mode identification results and the unified feature dataset to obtain the optimized feature weight set; Step S44: Calculate the local health index distribution based on the optimized feature weight set and the unified feature dataset; Step S45: Conduct a key area impact assessment based on the local health index distribution to obtain the key area impact score; Step S46: Based on the impact score of key areas and the distribution of local health index, predict performance parameters to obtain performance prediction results; Step S47: Combine the distribution of local health index, the impact score of key areas and the performance prediction results to obtain the tiered utilization evaluation index.

9. A battery energy storage state monitoring and evaluation system, characterized in that, For performing the battery energy storage state monitoring and evaluation method as described in claim 1, the battery energy storage state monitoring and evaluation system includes: The structural vibration response module is used to acquire the structural vibration response of the battery, obtain the frequency response characteristic set of the measurement points and the vibration mode distribution map; based on the frequency response characteristic set of the measurement points and the vibration mode distribution map, the mechanical impedance spectrum is calculated to obtain the battery vibration response spectrum; The modal health assessment module is used to identify resonance feature tables from the battery vibration response spectrum; calculate modal consistency index based on the resonance feature tables to obtain the modal consistency index; and construct a structural health vector based on the modal consistency index. The dielectric property detection module is used to optimize dielectric detection parameters based on the structural health vector to obtain a regional detection scheme; to collect multi-point dielectric response data based on the regional detection scheme to obtain a dielectric response dataset; to perform dielectric parameter inversion calculation on the dielectric response dataset to obtain a regional dielectric parameter set; and to perform dielectric property spatial mapping and anomaly identification on the regional dielectric parameter set to obtain a local dielectric property map. The failure diagnosis and rating module is used to perform structure-material correlation analysis on the structural health vector and local dielectric property spectrum to obtain failure mode identification results; based on the failure mode identification results, the battery energy storage health status is mapped to obtain the local health index distribution; based on the local health index distribution, the tiered utilization rating is comprehensively performed to obtain the tiered utilization evaluation index.

Images