A data analysis-based solid-state battery performance testing method and system

Abstract

This invention discloses a data analysis-based method and system for testing the performance of solid-state batteries, relating to the field of battery performance testing. The method includes: establishing a multi-physics coupled digital twin model; collecting strain signals and temperature field distributions of the failure region based on a three-dimensional physical field state map of the failure region; triggering multi-level early warnings when abnormal changes occur; matching physical failure evidence with a historical database; extracting the optimal combination of test parameters through a meta-learning framework; dynamically adjusting the charging and discharging strategy and monitoring frequency; generating a real-time data stream; updating the parameters of the multi-physics coupled digital twin model through a backpropagation algorithm; outputting the corrected multi-physics coupled digital twin model; and generating a test report. This invention dynamically optimizes test parameters through a meta-learning framework, enabling the monitoring and prediction of solid-state battery performance, which helps improve the safety and reliability of solid-state batteries and optimizes charging and discharging strategies to extend battery life.

Description

Technical Field

[0001] This invention relates to the field of battery performance testing, and in particular to a method and system for testing the performance of solid-state batteries based on data analysis. Background Technology

[0002] Solid-state batteries, as a new type of high-performance battery technology, have received widespread attention in the field of new energy in recent years. With the rapid development of electric vehicles, portable electronic devices and large-scale energy storage applications, the requirements for battery energy density, safety, cycle life and fast charge and discharge capabilities are becoming increasingly stringent. Solid-state batteries use solid electrolytes to replace traditional liquid electrolytes, which have higher ionic conductivity, more stable chemical properties and lower interfacial impedance, thus exhibiting significant performance advantages.

[0003] Although there are various technical means for testing the performance of solid-state batteries, existing technologies are limited to the measurement of a single physical field, making it difficult to fully reflect the complex multi-physical field coupling effects inside solid-state batteries. In addition, traditional testing methods usually rely on fixed experimental conditions, lack the ability to dynamically respond to real-time data streams, and cannot adjust themselves according to the actual working state of the battery. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a data analysis-based solid-state battery performance testing method to solve the problem that existing technologies are limited to the measurement of a single physical field and cannot fully reflect the complex multi-physical field coupling effects inside solid-state batteries.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a solid-state battery performance testing method based on data analysis, which includes: establishing a digital twin model of multi-physics coupling based on the material parameters and structural parameters of the target solid-state battery, and outputting a three-dimensional physical field state map of the failure region;

[0008] Based on the three-dimensional physical field state map of the failure area, the strain signal and temperature field distribution of the failure area are collected. When abnormal changes are detected, multi-level early warning is triggered and physical failure evidence is obtained.

[0009] Physical failure evidence is matched with historical databases, the optimal combination of test parameters is extracted through meta-learning framework, the charging and discharging strategy and monitoring frequency are dynamically adjusted, the adjusted test process is executed, and a real-time data stream is generated.

[0010] Based on real-time data streams, the parameters of the multi-physics coupled digital twin model are updated through the backpropagation algorithm, and the corrected multi-physics coupled digital twin model is output.

[0011] Based on the modified multiphysics coupling digital twin model, the time series of solid-state batteries is analyzed in real time, the failure probability curve is obtained, and a test report is generated.

[0012] As a preferred embodiment of the solid-state battery performance testing method based on data analysis described in this invention, the material parameters include: the composition of positive and negative electrode active materials, the type of solid electrolyte, and the ratio of binder to conductive agent;

[0013] The structural parameters include electrode thickness, porosity, particle size distribution, current collector type, and interface layer thickness.

[0014] As a preferred embodiment of the solid-state battery performance testing method based on data analysis described in this invention, the steps for establishing a multi-physics coupled digital twin model and outputting a three-dimensional physical field state map of the failure region are as follows:

[0015] Based on the material and structural parameters of the target solid-state battery, synchrotron radiation X-ray tomography is driven to reconstruct the three-dimensional voxel matrix inside the electrode at nanometer resolution and obtain the three-dimensional microstructure of the electrode.

[0016] Based on the three-dimensional voxel matrix, the three-dimensional microstructure of the electrode is discretized into regular grid units. Material parameters and structural parameters are injected into each grid unit to establish an electrochemical-thermal-mechanical coupling equation.

[0017] Based on the electrochemical-thermodynamic coupling equation set, a multi-physics field coupled digital twin model is constructed to dynamically reflect the ion concentration gradient distribution, stress field distribution, and temperature field distribution;

[0018] Based on the three-dimensional microstructure of the electrode provided by the three-dimensional voxel matrix, a three-dimensional physical field state map of ion concentration gradient distribution, stress distribution field and temperature field distribution is dynamically output through a digital twin model of multi-physics field coupling.

[0019] By using the regional anomaly frequencies in historical failure cases, the failure locations of lithium dendrite growth regions and interface peeling regions are dynamically identified in the three-dimensional physical field state map, forming a three-dimensional physical field state map of the failure region.

[0020] As a preferred embodiment of the solid-state battery performance testing method based on data analysis described in this invention, the steps of collecting strain signals and temperature field distribution in the failure area, triggering multi-level early warnings and obtaining physical failure evidence when abnormal abrupt changes are detected, are as follows:

[0021] Based on the three-dimensional physical field state map of the failure area, distributed fiber optic strain sensors and non-contact infrared thermal imagers are attached to the surface of the dynamically marked lithium dendrite growth area and interface peeling area to collect local strain signals and surface temperature field distribution in real time.

[0022] The fiber optic strain sensor continuously outputs strain time-series data. When a sudden increase in local strain is detected, it is determined to be the critical point of mechanical failure. The non-contact infrared thermal imager outputs a temperature field distribution map. When a local temperature changes abruptly, it is determined to be a precursor to thermal runaway.

[0023] When either a critical point of mechanical failure or a precursor to thermal runaway occurs, a multi-level early warning system is immediately triggered, and a list of early warning records is obtained.

[0024] Based on multi-level early warning and safety control, when the early warning is triggered, physical failure evidence of the lithium dendrite growth region and the interface peeling region is collected by field emission scanning electron microscopy and X-ray photoelectron spectroscopy, respectively.

[0025] As a preferred embodiment of the data analysis-based solid-state battery performance testing method of the present invention, the steps of matching physical failure evidence with historical databases, extracting the optimal combination of test parameters through a meta-learning framework, dynamically adjusting the charge / discharge strategy and monitoring frequency, executing the adjusted test process, and generating a real-time data stream are as follows.

[0026] The historical database includes crack morphology characteristics, monitoring frequency values, decomposition product composition, charge / discharge rate values, pressure loading values, and failure event frequency per unit cycle.

[0027] The physical failure evidence and historical database are transformed into multidimensional vectors and historical database vectors using the feature vectorization method. The matching degree between the multidimensional vectors and historical database vectors is calculated by cosine similarity to filter the failure case subset.

[0028] Based on a subset of failure cases, a covariance matrix is ​​constructed using a radial basis kernel covariance function. The objective function is defined as the negative logarithmic marginal likelihood. The cycle life and polarization voltage of the solid-state battery are used as constraints to construct a Gaussian process regression surrogate model.

[0029] A meta-learning framework is adopted to initialize the search space for charge / discharge rate values ​​and pressure loading values. A Gaussian regression surrogate model is used to predict the probability distribution of charge / discharge rate values ​​and pressure loading values. At the same time, the expectation enhancement criterion is used to screen the optimal solution that satisfies the constraints and obtain the optimal combination of test parameters.

[0030] Based on the predicted charge / discharge rate value in the optimal test parameter combination, adjust the current curve shape of the constant current charge / discharge test, adjust the monitoring frequency value according to the pressure loading value, and generate an adjusted test parameter log.

[0031] After the segmented constant current mode and monitoring frequency are adjusted, constant current charge and discharge tests are performed using battery testing methods. The internal resistance is calculated using the DC internal resistance method, and the temperature is monitored by an infrared thermal imager to generate a real-time data stream.

[0032] As a preferred embodiment of the data analysis-based solid-state battery performance testing method of the present invention, the steps of updating the parameters of the multi-physics coupled digital twin model through the backpropagation algorithm and outputting the corrected multi-physics coupled digital twin model are as follows:

[0033] The real-time data stream is input into the Levenberg-Marquardt optimizer. By minimizing the residual between the actual observed data and the theoretical predictions of the multiphysics coupled digital twin model, the backpropagation algorithm is implemented to correct the solid-phase diffusion coefficient and interface reaction term in the multiphysics coupled digital twin model.

[0034] By combining the actual cycling capacity decay data in the real-time data stream, the degradation equation of the multi-physics coupled digital twin model is backfitted to calibrate the nonlinear cumulative effect of the active material loss rate during cycling, and the corrected multi-physics coupled digital twin model is output.

[0035] As a preferred embodiment of the solid-state battery performance testing method based on data analysis described in this invention, the specific steps for real-time parsing of the solid-state battery's time-series sequence, obtaining the failure probability curve, and generating a test report are as follows.

[0036] The charging and discharging cycle process is dynamically simulated based on the modified multi-physics coupling digital twin model. The time series of voltage, current and temperature generated in real time are analyzed, and the voltage polarization curve shape and capacity decay rate value are extracted.

[0037] Based on the frequency of unit cycle failure events in the historical database, the adjacent difference values ​​of the time series are calculated in real time as the decay rate slope to form the input dataset for the Granger causality test.

[0038] Based on the voltage polarization curve shape and capacity decay rate value, the slope of the capacity decay rate is correlated with the frequency of failure events through Granger causality test, generating a cycle number-failure probability curve and marking the critical threshold.

[0039] Based on the cycle count-failure probability curve, the three-dimensional physical field state map of the failure area, multi-level early warning records, and adjusted test parameter logs, a test report is generated by intelligently aligning standardized templates with multi-source data through the MCP protocol.

[0040] Secondly, the present invention provides a solid-state battery performance testing system based on data analysis, comprising:

[0041] Failure module, early warning module, testing module, correction module, and analysis module.

[0042] The failure module establishes a multi-physics coupled digital twin model based on the material and structural parameters of the target solid-state battery, and outputs a three-dimensional physical field state map of the failure region.

[0043] The early warning module collects strain signals and temperature field distribution in the failure area based on the three-dimensional physical field state map of the failure area. When an abnormal change is detected, it triggers multi-level early warning and obtains physical failure evidence.

[0044] The testing module matches physical failure evidence with historical databases, extracts the optimal combination of test parameters through a meta-learning framework, dynamically adjusts the charging and discharging strategy and monitoring frequency, executes the adjusted test process, and generates real-time data streams.

[0045] The correction module updates the parameters of the multi-physics coupled digital twin model based on real-time data streams using the backpropagation algorithm, and outputs the corrected multi-physics coupled digital twin model.

[0046] The analysis module, based on the modified multi-physics coupled digital twin model, analyzes the time sequence of solid-state batteries in real time, obtains the failure probability curve, and generates a test report.

[0047] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, it implements any step of the solid-state battery performance testing method based on data analysis as described in the first aspect of the present invention.

[0048] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the solid-state battery performance testing method based on data analysis as described in the first aspect of the present invention.

[0049] The beneficial effects of this invention are as follows: By establishing a digital twin model with multi-physics coupling, this invention can collect strain signals and temperature field distribution of solid-state batteries in real time, and dynamically optimize test parameters by combining a meta-learning framework, thereby realizing the monitoring and prediction of solid-state battery performance. This helps to improve the safety and reliability of solid-state batteries and optimize charging and discharging strategies to extend battery life. Attached Figure Description

[0050] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a flowchart of a data analysis-based solid-state battery performance testing method.

[0052] Figure 2 Flowchart for constructing a multiphysics coupled digital twin model.

[0053] Figure 3 Flowchart for real-time data stream generation and model correction.

[0054] Figure 4 A flowchart for generating failure probability analysis and test reports. Detailed Implementation

[0055] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0056] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0057] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0058] Reference Figures 1 to 4 This is one embodiment of the present invention, which provides a solid-state battery performance testing method based on data analysis, including the following steps:

[0059] S1. Based on the material and structural parameters of the target solid-state battery, establish a digital twin model with multi-physics coupling and output a three-dimensional physical field state map of the failure region.

[0060] The specific steps are as follows:

[0061] Material parameters include the composition of positive and negative electrode active materials, solid electrolyte type, and the ratio of binder to conductive agent; structural parameters include electrode thickness, porosity, particle size distribution, current collector type, and interface layer thickness.

[0062] Based on the material and structural parameters of the target solid-state battery, synchrotron X-ray tomography is used to reconstruct the three-dimensional voxel matrix inside the electrode at nanometer resolution, and to obtain the three-dimensional microstructure of the electrode containing the spatial distribution and connectivity of the active material phase, porous phase, and electrolyte phase.

[0063] It should be noted that, based on the material and structural parameters of the target solid-state battery, synchrotron X-ray tomography is used to perform a rotating scan of the battery electrode at nanometer resolution using a parallel beam of hard X-rays, acquiring projection images at different angles. These projection images are then arranged in order according to the rotation angle to form a two-dimensional projection sequence. The two-dimensional projection sequence is then converted into a three-dimensional voxel matrix using a filtered back-projection algorithm, where the voxel gray values ​​reflect the differences in the linear attenuation coefficients of the material to X-rays. A threshold segmentation algorithm is used to distinguish the active material phase, porous phase, and electrolyte phase. Morphological operations are combined to quantify the spatial distribution and connectivity parameters of the active material phase, porous phase, and electrolyte phase, ultimately outputting the three-dimensional microstructure of the electrode, including the phase interface contact area and ion transport paths.

[0064] Based on the three-dimensional voxel matrix, the three-dimensional microstructure of the electrode is discretized into regular grid units. Material parameters and structural parameters are injected into each grid unit to establish an electrochemical-thermal-mechanical coupling equation.

[0065] It should be noted that, based on the three-dimensional voxel matrix, the octree spatial partitioning algorithm is used to discretize the three-dimensional microstructure of the electrode into hexahedral regular mesh units, and material parameters and structural parameters are injected into each mesh unit; in COMSOL Multiphysics, the electrochemical equation, thermal equation and mechanical equation are solved in a fully coupled manner within the discretized hexahedral regular mesh units by the finite volume method;

[0066] Among them, the electrochemical equation and the thermal equation are coupled through the Joule heating term, the thermal equation and the mechanical equation are coupled through the thermal expansion term, and the mechanical equation and the electrochemical equation are coupled through the stress-dependent interface reaction rate term, thus constructing a fully coupled electrochemical-thermal-mechanical control equation set.

[0067] Based on the electrochemical-thermodynamic coupling equation set, a multi-physics field coupled digital twin model is constructed to dynamically reflect the ion concentration gradient distribution, stress field distribution, and temperature field distribution;

[0068] The steps for constructing a multiphysics coupled digital twin model are as follows.

[0069] The electrochemical equations are constructed based on the ion diffusion characteristics in the material parameters. The lithium ion concentration gradient distribution is calculated using the Nernst-Planck equation, where the solid-phase diffusion coefficient is determined by the active material composition in the material parameters.

[0070] The thermal equations are constructed by combining Fourier's law of heat conduction with the porosity in the structural parameters to calculate the temperature field distribution formed by the local heat generation rate. The thermal conductivity is determined by the electrolyte type and the proportion of binder.

[0071] The mechanical equations are constructed based on the particle size distribution of the active particles. Linear elasticity equations are used to quantify the electrochemical expansion stress. The resulting stress distribution is constrained by the particle size and the type of current collector.

[0072] Based on the three-dimensional microstructure of the electrode provided by the three-dimensional voxel matrix, a three-dimensional physical field state map of ion concentration gradient distribution, stress distribution field and temperature field distribution is dynamically output through a digital twin model of multi-physics field coupling.

[0073] By using the regional anomaly frequencies in historical failure cases, the failure locations of lithium dendrite growth regions and interface peeling regions are dynamically identified in the three-dimensional physical field state map, thus forming a three-dimensional physical field state map of the failure region.

[0074] It should be noted that, based on the ion concentration gradient distribution, stress distribution, and temperature field distribution in the three-dimensional physical field state map, the physical characteristic criteria of the lithium dendrite growth region are first extracted: when the local ion concentration gradient exceeds a critical value and the current density exceeds a current threshold, the current threshold is defined according to the space charge theory; for example, the current threshold ranges from [0.8−1.2] mA / cm². 2 When the area is identified as a high-risk zone for dendrite nucleation, the spatial failure probability distribution is calculated by combining the regional abnormal frequency data in the historical failure case library with Gaussian kernel density estimation. The real-time physical field data and the failure probability distribution are then fused in Bayesian mode to dynamically identify failure areas that meet the joint conditions and generate a three-dimensional physical field state map of the failure area.

[0075] S2. Based on the three-dimensional physical field state map of the failure area, collect the strain signal and temperature field distribution of the failure area. When abnormal sudden change is detected, trigger multi-level early warning and obtain physical failure evidence.

[0076] The specific steps are as follows:

[0077] Based on the three-dimensional physical field state map of the failure area, distributed fiber optic strain sensors and non-contact infrared thermal imagers are attached to the surface of the dynamically marked lithium dendrite growth area and interface peeling area to collect local strain signals and surface temperature field distribution in real time.

[0078] It should be noted that in the three-dimensional physical field state map of the failure area, distributed fiber optic strain sensors are attached to the surface of the lithium dendrite growth area and the interface stripping area in an array form to measure the local strain signal of the lithium dendrite growth area caused by lithium metal deposition; a non-contact infrared thermal imager simultaneously collects the surface temperature field distribution of the interface stripping area, and locates the hot spot area by combining the thermal anomaly threshold. The thermal anomaly threshold is based on empirical determination in the battery thermal runaway early warning and is used to distinguish between normal thermal diffusion and abnormal local overheating.

[0079] The non-contact infrared thermal imager locates areas of abnormal temperature gradient by detecting the surface temperature field distribution in the interface stripping zone. It then uses the distributed fiber optic strain sensor to capture stress anomaly signals in the lithium dendrite growth zone, and aligns these signals with timestamps to achieve real-time monitoring through electrochemical-thermal-mechanical dual-field coupling.

[0080] The fiber optic strain sensor continuously outputs strain time-series data. When a sudden increase in local strain is detected, it is determined to be the critical point of mechanical failure. The non-contact infrared thermal imager outputs a temperature field distribution map. When a local temperature changes abruptly, it is determined to be a precursor to thermal runaway.

[0081] When mechanical failure critical points and signs of thermal runaway appear, multi-level early warnings are immediately triggered and a list of early warning records is obtained.

[0082] The multi-level early warning system includes a level 1 audible and visual alarm, a level 2 charging and discharging suspension, and a level 3 safety isolation.

[0083] Based on the multi-level early warning system, physical failure evidence of the interface debonding region and lithium dendrite growth region is collected at the same time as the early warning is triggered by field emission scanning electron microscopy and X-ray photoelectron spectroscopy.

[0084] It should be noted that after triggering the multi-level warning, the lithium dendrite growth region is scanned at low voltage using a field emission scanning electron microscope, and the three-dimensional morphology and stress concentration area of ​​the dendrite growth region are obtained using secondary electron imaging; at the same time, X-ray photoelectron spectroscopy is used to excite photoelectrons in the interface stripping region by Al Kα rays, and the binding energy shift is analyzed to detect the chemical state changes of electrolyte decomposition products.

[0085] After aligning the time stamps of the photoelectrons from the three-dimensional morphology of the lithium dendrite growth region, stress concentration region, and interface delamination region obtained by field emission scanning electron microscopy and X-ray photoelectron spectroscopy, the crack propagation paths in the three-dimensional morphology of the lithium dendrite growth region and stress concentration region obtained by field emission scanning electron microscopy, and the elemental valence state shifts detected by analyzing the binding energy shifts of the photoelectrons in the interface delamination region obtained by exciting Al Kα rays in X-ray photoelectron spectroscopy, together constitute physical failure evidence.

[0086] S3. Match physical failure evidence with historical database, extract the optimal test parameter combination through meta-learning framework, dynamically adjust charging and discharging strategy and monitoring frequency, execute the adjusted test process, and generate real-time data stream.

[0087] The specific steps are as follows:

[0088] The historical database includes crack morphology characteristics, monitoring frequency values, decomposition product composition, charge / discharge rate values, pressure loading values, and the frequency of failure events per unit cycle.

[0089] The physical failure evidence and historical database are transformed into multidimensional vectors and historical database vectors using the feature vectorization method. The matching degree between the multidimensional vectors and historical database vectors is calculated by cosine similarity to filter the failure case subset.

[0090] It should be noted that the physical failure evidence was transformed into a 128-dimensional feature vector after dimensionality reduction through principal component analysis, and the cases in the historical database were processed using the same PCA parameters to generate historical database vectors; and cosine similarity was used to calculate the matching degree between the physical failure evidence feature vector and the historical database vector. The formula is:

[0091] ;

[0092] in, The physical failure evidence feature vector is represented as the first... Dimensional component values, Represented as a historical database vector in the th... Dimensional component values;

[0093] The K-nearest neighbor algorithm is used to sort the physical failure evidence feature vectors and historical database vectors by cosine similarity, and the top K cases with the highest similarity are directly selected as the matching subset.

[0094] The failure case subset includes cycle life, polarization voltage, material parameters, and structural parameters.

[0095] Based on a subset of failure cases, a covariance matrix is ​​constructed using a radial basis kernel covariance function. The objective function is defined as the negative logarithmic marginal likelihood. The cycle life and polarization voltage of the solid-state battery are used as constraints to construct a Gaussian process regression surrogate model.

[0096] It should be noted that the cycle lifetime and polarization voltage in the failure case subset are Z-score standardized, and the maximum value of the ion concentration gradient distribution, stress concentration coefficient, and temperature field abrupt change amplitude are extracted as input features, while cycle lifetime and polarization voltage are used as output targets. Subsequently, the radial basis kernel covariance function is used to describe the nonlinear relationship between the input features, and a covariance matrix is ​​constructed. The objective function is defined as the negative logarithmic marginal likelihood. Then, the lower limit of cycle lifetime and the upper limit of polarization voltage of the solid-state battery are set as constraints, and the hyperparameters of the radial basis kernel covariance function are optimized using the conjugate gradient method to obtain a Gaussian process regression surrogate model.

[0097] A meta-learning framework is adopted to initialize the search space for charge / discharge rate values ​​and pressure loading values. A Gaussian regression surrogate model is used to predict the probability distribution of charge / discharge rate values ​​and pressure loading values. At the same time, the expectation enhancement criterion is used to screen the optimal solution that satisfies the constraints and obtain the optimal combination of test parameters.

[0098] It should be noted that, within the meta-learning framework, the initial parameter combination is generated through Latin hypercube sampling within the range of charge / discharge rate and pressure loading value. The mapping relationship between the initial parameter combination and the battery capacity decay rate is established using a Gaussian process regression model. The output is a predicted distribution containing the mean and variance. The potential improvement space of the new parameter combination compared with the current optimal solution is quantified by the expectation enhancement criterion. The Pareto front is screened by combining the constraint cycle life and polarization voltage to always obtain the optimal test parameter combination.

[0099] Based on the predicted charge / discharge rate value in the optimal test parameter combination, adjust the current curve shape of the constant current charge / discharge test, adjust the monitoring frequency value according to the pressure loading value, and generate an adjusted test parameter log.

[0100] It should be noted that, based on the predicted charge / discharge rate value in the optimal test parameter combination, the current curve shape of the constant current charge / discharge test is adjusted to a segmented constant current mode. During the charging phase, the initial current is used for charging, and when the voltage reaches the upper limit, it is switched to constant current charging until the cutoff voltage.

[0101] Based on the dynamic range of the pressure loading value, the monitoring frequency value is dynamically adjusted according to a linear relationship, and the current, voltage, and pressure time sequence data are completely recorded during the charge and discharge cycle. This generates an adjusted test parameter log containing timestamps, charge and discharge rate values, pressure loading values, operating parameters, monitoring frequency values, and corresponding performance indicators.

[0102] After the segmented constant current mode and monitoring frequency are adjusted, constant current charge and discharge tests are performed using battery testing methods. The internal resistance is calculated using the DC internal resistance method, and the temperature is monitored by an infrared thermal imager to generate a real-time data stream.

[0103] It should be noted that after the charging and discharging strategy and monitoring frequency are adjusted, constant current charging and discharging tests are performed using battery testing methods. The battery internal resistance is calculated in real time using the DC internal resistance method. At the same time, the infrared thermal imager collects temperature distribution data at the adjusted monitoring frequency value, generating a real-time data stream containing timestamps, current, voltage, internal resistance, temperature, pressure loading value, minimized actual observation data, and actual cycle capacity decay data.

[0104] Among them, minimizing the actual observed data is obtained by using the least squares method to optimize the sum of squared residuals between the theoretical and measured values ​​after predicting the capacity decay rate through a Gaussian process regression surrogate model; the actual cycle capacity decay data is calculated by the ratio of the discharge capacity to the initial capacity recorded in the constant current charge and discharge test, and is dynamically calibrated based on temperature and rate correction coefficients.

[0105] S4. Based on real-time data stream, update the parameters of the multi-physics coupled digital twin model through backpropagation algorithm, and output the corrected multi-physics coupled digital twin model.

[0106] The specific steps are as follows:

[0107] The real-time data stream is input into the Levenberg-Marquardt optimizer. By minimizing the residual between the actual observed data and the theoretical predictions of the multiphysics coupled digital twin model, the backpropagation algorithm is implemented to correct the solid-phase diffusion coefficient and interface reaction term in the multiphysics coupled digital twin model.

[0108] It should be noted that after the real-time data stream is input into the Levenberg-Marquardt optimizer, the objective function is constructed by calculating the residual vector between the predicted values ​​of the multiphysics coupled digital twin model and the actual observed data. The parameter correction is then iteratively solved using the Jacobian matrix according to the Levenberg-Marquardt update rule. The parameter correction is achieved by comparing the residuals between the model predictions and the actual observed data, and the damping factor and Jacobian matrix are dynamically adjusted using the Levenberg-Marquardt optimizer. The adjustment values ​​of the solid-phase diffusion coefficient and the interface reaction term of the multiphysics digital twin model are iteratively optimized until the residuals converge.

[0109] By combining the actual cycling capacity decay data in the real-time data stream, the degradation equation of the multi-physics coupled digital twin model is backfitted to calibrate the nonlinear cumulative effect of the active material loss rate during cycling, and the corrected multi-physics coupled digital twin model is output.

[0110] The loss rate of active materials was determined by calibrating the nonlinear cumulative effect parameters in the degradation equation during the cyclic process.

[0111] It should be noted that the actual circulating capacity decay data in the real-time data stream is input into the multiphysics coupled digital twin model. The residual function is constructed by backfitting the degradation equation. The nonlinear cumulative effect parameters of the active material loss rate value are solved iteratively using the Levenberg-Marquardt optimizer. After each iteration, the coupling term of the solid-phase diffusion coefficient and the interface reaction rate in the multiphysics coupled digital twin model is corrected by the backpropagation algorithm until the residual converges. The output is the corrected multiphysics coupled digital twin model containing the calibrated active material loss rate equation and the multiphysics coupling parameters.

[0112] S5. Based on the modified multi-physics coupling digital twin model, the time series sequence of the solid-state battery is analyzed in real time, and the failure probability curve is obtained to generate a test report.

[0113] The specific steps are as follows:

[0114] The charging and discharging cycle process is dynamically simulated based on the modified multi-physics coupling digital twin model. The time series of voltage, current and temperature generated in real time are analyzed, and the voltage polarization curve shape and capacity decay rate value are extracted.

[0115] It should be noted that when dynamically simulating the charge-discharge cycle process based on the modified multiphysics coupled digital twin model, the real-time collected voltage, current and temperature time series are first used as boundary conditions and input into the multiphysics coupled digital twin model. The lithium-ion concentration gradient distribution is obtained by solving the Nernst-Planck equation, the temperature field distribution is calculated by combining Fourier's law of heat conduction, and the stress distribution is updated based on the linear elasticity equation. Subsequently, the voltage, current and temperature time series are analyzed and the voltage polarization curve shape and capacity decay rate value are extracted. The decay rate constant is extracted by the capacity increment analysis method.

[0116] Based on the frequency of unit cycle failure events in the historical database, the adjacent difference values ​​of the time series are calculated in real time as the decay rate slope to form the input dataset for the Granger causality test.

[0117] It should be noted that, based on the frequency of failure events per unit cycle in the historical database, the time series sequence of failure event frequency is first extracted, and the difference between adjacent time points is calculated through difference operation to generate the decay rate slope sequence; then, the stationarity test and lag order selection of the decay rate slope sequence are performed to ensure that the preconditions of Granger causality test are met, and the decay rate slope sequence is combined with the time series sequence of the target variable to obtain a multivariate input dataset for Granger causality test analysis;

[0118] The target variable time series is derived from the direct pairing of the preprocessed decay rate slope sequence and the time series of failure event frequency per unit cycle in the historical database.

[0119] Based on the voltage polarization curve shape and capacity decay rate value, the slope of the capacity decay rate is correlated with the frequency of failure events through Granger causality test, generating a cycle number-failure probability curve and marking the critical threshold.

[0120] It should be noted that feature parameters are extracted from the voltage polarization curve morphology, and the change in capacity decay rate value at adjacent time points is calculated using the difference method to form a slope sequence. The capacity decay rate slope sequence is then compared with the time series sequence of failure event frequency per unit cycle in the historical database using a Granger causality test. Specifically, the capacity decay rate slope sequence is used as the independent variable, and the failure event frequency sequence is used as the dependent variable. The difference in the sum of squared regression residuals with and without the lagged terms of the capacity decay rate slope sequence is used to verify the predictive power of the capacity decay rate slope for failure events. A mapping relationship between the number of cycles and the failure probability is established through logistic regression, and a cycle number-failure probability curve is generated. The cycle number corresponding to the abrupt change point of the capacity decay rate slope is marked on the cycle number-failure probability curve as a critical threshold.

[0121] Based on the cycle count-failure probability curve, the three-dimensional physical field state map of the failure area, multi-level early warning records, and adjusted test parameter logs, a test report is generated by intelligently aligning standardized templates with multi-source data through the MCP protocol.

[0122] It should be noted that the failure probability time series data is extracted based on the cycle number-failure probability curve, and the spatial strain / stress distribution characteristics are analyzed by combining the three-dimensional physical field state map of the failure area. The warning level and trigger timestamp in the three-level warning record are synchronously associated, and the time series is aligned with the working condition parameters in the adjusted test parameter log. Multi-source data cleaning, field mapping and unit standardization are completed through a structured ETL process. Finally, it is integrated into a structured test report that includes a failure probability trend chart, a three-dimensional physical field isosurface rendering map, a warning event time axis and a parameter adjustment comparison table.

[0123] This embodiment also provides a solid-state battery performance testing system based on data analysis, including: a failure module, an early warning module, a testing module, a correction module, and an analysis module.

[0124] The failure module establishes a multi-physics coupled digital twin model based on the material and structural parameters of the target solid-state battery, and outputs a three-dimensional physical field state map of the failure region.

[0125] The early warning module collects strain signals and temperature field distribution in the failure area based on the three-dimensional physical field state map of the failure area. When an abnormal change is detected, it triggers multi-level early warning and obtains physical failure evidence.

[0126] The testing module matches physical failure evidence with historical databases, extracts the optimal combination of test parameters through a meta-learning framework, dynamically adjusts the charging and discharging strategy and monitoring frequency, executes the adjusted test process, and generates real-time data streams.

[0127] The correction module updates the parameters of the multi-physics coupled digital twin model based on real-time data streams using the backpropagation algorithm, and outputs the corrected multi-physics coupled digital twin model.

[0128] The analysis module, based on the modified digital twin model of multi-physics coupling, analyzes the time sequence of solid-state batteries in real time, obtains the failure probability curve, and generates a test report.

[0129] This embodiment also provides a computer device applicable to the data analysis-based solid-state battery performance testing method, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the data analysis-based solid-state battery performance testing method proposed in the above embodiment.

[0130] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0131] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the data analysis-based solid-state battery performance testing method proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0132] In summary, this invention achieves the monitoring and prediction of solid-state battery performance by establishing a digital twin model coupled with multiple physics fields, acquiring strain signals and temperature field distribution of solid-state batteries in real time, and dynamically optimizing test parameters using a meta-learning framework. This helps improve the safety and reliability of solid-state batteries and optimize charging and discharging strategies to extend battery life.

[0133] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A solid-state battery performance testing method based on data analysis, characterized in that: include: Based on the material and structural parameters of the target solid-state battery, a digital twin model with multi-physics coupling is established, and a three-dimensional physical field state map of the failure region is output. Based on the three-dimensional physical field state map of the failure area, the strain signal and temperature field distribution of the failure area are collected. When abnormal changes are detected, multi-level early warning is triggered and physical failure evidence is obtained. Physical failure evidence is matched with historical databases, the optimal combination of test parameters is extracted through meta-learning framework, the charging and discharging strategy and monitoring frequency are dynamically adjusted, the adjusted test process is executed, and a real-time data stream is generated. Based on real-time data streams, the parameters of the multi-physics coupled digital twin model are updated through the backpropagation algorithm, and the corrected multi-physics coupled digital twin model is output. Based on the modified multiphysics coupling digital twin model, the time series of solid-state batteries is analyzed in real time, the failure probability curve is obtained, and a test report is generated.

2. The solid-state battery performance testing method based on data analysis as described in claim 1, characterized in that: The material parameters include the composition of the positive and negative electrode active materials, the type of solid electrolyte, and the ratio of binder to conductive agent; The structural parameters include electrode thickness, porosity, particle size distribution, current collector type, and interface layer thickness.

3. The solid-state battery performance testing method based on data analysis as described in claim 1, characterized in that: The steps for establishing a multi-physics coupled digital twin model and outputting a three-dimensional physical field state map of the failure region are as follows. Based on the material and structural parameters of the target solid-state battery, synchrotron radiation X-ray tomography is driven to reconstruct the three-dimensional voxel matrix inside the electrode at nanometer resolution and obtain the three-dimensional microstructure of the electrode. Based on the three-dimensional voxel matrix, the three-dimensional microstructure of the electrode is discretized into regular grid units. Material parameters and structural parameters are injected into each grid unit to establish an electrochemical-thermal-mechanical coupling equation. Based on the electrochemical-thermodynamic coupling equation set, a multi-physics field coupled digital twin model is constructed to dynamically reflect the ion concentration gradient distribution, stress field distribution, and temperature field distribution; Based on the three-dimensional microstructure of the electrode provided by the three-dimensional voxel matrix, a three-dimensional physical field state map of ion concentration gradient distribution, stress distribution field and temperature field distribution is dynamically output through a digital twin model of multi-physics field coupling. By using the regional anomaly frequencies in historical failure cases, the failure locations of lithium dendrite growth regions and interface peeling regions are dynamically identified in the three-dimensional physical field state map, forming a three-dimensional physical field state map of the failure region.

4. The solid-state battery performance testing method based on data analysis as described in claim 1, characterized in that: The strain signal and temperature field distribution of the acquired failure area are used to trigger multi-level early warning and obtain physical failure evidence when abnormal abrupt changes are detected. The steps are as follows. Based on the three-dimensional physical field state map of the failure area, distributed fiber optic strain sensors and non-contact infrared thermal imagers are attached to the surface of the dynamically marked lithium dendrite growth area and interface peeling area to collect local strain signals and surface temperature field distribution in real time. The fiber optic strain sensor continuously outputs strain time-series data. When a sudden increase in local strain is detected, it is determined to be the critical point of mechanical failure. The non-contact infrared thermal imager outputs a temperature field distribution map. When a local temperature changes abruptly, it is determined to be a precursor to thermal runaway. When either a critical point of mechanical failure or a precursor to thermal runaway occurs, a multi-level early warning system is immediately triggered, and a list of early warning records is obtained. Based on multi-level early warning and safety control, when the early warning is triggered, physical failure evidence of the lithium dendrite growth region and the interface peeling region is collected by field emission scanning electron microscopy and X-ray photoelectron spectroscopy, respectively.

5. The solid-state battery performance testing method based on data analysis as described in claim 1, characterized in that: The steps involved matching physical failure evidence with a historical database, extracting the optimal combination of test parameters using a meta-learning framework, dynamically adjusting the charging / discharging strategy and monitoring frequency, executing the adjusted test process, and generating a real-time data stream. These steps are as follows: The historical database includes crack morphology characteristics, monitoring frequency values, decomposition product composition, charge / discharge rate values, pressure loading values, and failure event frequency per unit cycle. The physical failure evidence and historical database are transformed into multidimensional vectors and historical database vectors using the feature vectorization method. The matching degree between the multidimensional vectors and historical database vectors is calculated by cosine similarity to filter the failure case subset. Based on a subset of failure cases, a covariance matrix is ​​constructed using a radial basis kernel covariance function. The objective function is defined as the negative logarithmic marginal likelihood. The cycle life and polarization voltage of the solid-state battery are used as constraints to construct a Gaussian process regression surrogate model. A meta-learning framework is adopted to initialize the search space for charge / discharge rate values ​​and pressure loading values. A Gaussian regression surrogate model is used to predict the probability distribution of charge / discharge rate values ​​and pressure loading values. At the same time, the expectation enhancement criterion is used to screen the optimal solution that satisfies the constraints and obtain the optimal combination of test parameters. Based on the predicted charge / discharge rate value in the optimal test parameter combination, adjust the current curve shape of the constant current charge / discharge test, adjust the monitoring frequency value according to the pressure loading value, and generate an adjusted test parameter log. After the segmented constant current mode and monitoring frequency are adjusted, constant current charge and discharge tests are performed using battery testing methods. The internal resistance is calculated using the DC internal resistance method, and the temperature is monitored by an infrared thermal imager to generate a real-time data stream.

6. The solid-state battery performance testing method based on data analysis as described in claim 1, characterized in that: The steps for updating the parameters of the multiphysics coupled digital twin model using the backpropagation algorithm and outputting the corrected multiphysics coupled digital twin model are as follows. The real-time data stream is input into the Levenberg-Marquardt optimizer. By minimizing the residual between the actual observed data and the theoretical predictions of the multiphysics coupled digital twin model, the backpropagation algorithm is implemented to correct the solid-phase diffusion coefficient and interface reaction term in the multiphysics coupled digital twin model. By combining the actual cycling capacity decay data in the real-time data stream, the degradation equation of the multi-physics coupled digital twin model is backfitted to calibrate the nonlinear cumulative effect of the active material loss rate during cycling, and the corrected multi-physics coupled digital twin model is output.

7. The solid-state battery performance testing method based on data analysis as described in claim 1, characterized in that: The process of real-time analysis of the solid-state battery's time series sequence, acquisition of the failure probability curve, and generation of a test report involves the following steps: The charging and discharging cycle process is dynamically simulated based on the modified multi-physics coupling digital twin model. The time series of voltage, current and temperature generated in real time are analyzed, and the voltage polarization curve shape and capacity decay rate value are extracted. Based on the frequency of unit cycle failure events in the historical database, the adjacent difference values ​​of the time series are calculated in real time as the decay rate slope to form the input dataset for the Granger causality test. Based on the voltage polarization curve shape and capacity decay rate value, the slope of the capacity decay rate is correlated with the frequency of failure events through Granger causality test, generating a cycle number-failure probability curve and marking the critical threshold. Based on the cycle count-failure probability curve, the three-dimensional physical field state map of the failure area, multi-level early warning records, and adjusted test parameter logs, a test report is generated by intelligently aligning standardized templates with multi-source data through the MCP protocol.

8. A solid-state battery performance testing system based on data analysis, based on the solid-state battery performance testing method based on data analysis according to any one of claims 1 to 7, characterized in that: include: Failure module, early warning module, testing module, correction module, and analysis module. The failure module establishes a multi-physics coupled digital twin model based on the material and structural parameters of the target solid-state battery, and outputs a three-dimensional physical field state map of the failure region. The early warning module collects strain signals and temperature field distribution in the failure area based on the three-dimensional physical field state map of the failure area. When an abnormal change is detected, it triggers multi-level early warning and obtains physical failure evidence. The testing module matches physical failure evidence with historical databases, extracts the optimal combination of test parameters through a meta-learning framework, dynamically adjusts the charging and discharging strategy and monitoring frequency, executes the adjusted test process, and generates real-time data streams. The correction module updates the parameters of the multi-physics coupled digital twin model based on real-time data streams using the backpropagation algorithm, and outputs the corrected multi-physics coupled digital twin model. The analysis module, based on the modified multi-physics coupled digital twin model, analyzes the time sequence of solid-state batteries in real time, obtains the failure probability curve, and generates a test report.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the solid-state battery performance testing method based on data analysis as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the solid-state battery performance testing method based on data analysis as described in any one of claims 1 to 7.

Images