A battery detection method, electronic equipment and storage medium

Abstract

The embodiment of the application provides a battery detection method, an electronic device and a storage medium, through deep analysis of macroscopic original operation data into microcosmic interface peeling state information specific to a sulfide solid-state battery, a quantitative mapping system from an external electrical signal to internal solid-solid contact quality is successfully established, a problem of a detection blind area of early microcosmic damage without significant abnormality in macroscopic dimensions such as voltage and current caused by the "soft mechanics" characteristics of the sulfide electrolyte is fundamentally solved, early accurate perception of battery interface failure and transparent evaluation of health degree are realized, and therefore, a scientific decision basis is provided for timely maintenance before missing the best intervention opportunity.

Description

Technical Field

[0001] This invention relates to the field of battery testing technology, and in particular to a battery testing method, an electronic device, and a computer-readable storage medium. Background Technology

[0002] With the rapid development of the new energy vehicle industry, sulfide solid-state batteries have become an important development direction in the all-solid-state battery technology route due to their high ionic conductivity, good mechanical ductility, and potential cost advantages. During the use of sulfide solid-state batteries, the solid-solid physical contact interface formed between the electrode active material particles and the sulfide solid electrolyte layer is the key channel for lithium-ion transport.

[0003] Battery testing technologies primarily rely on battery management systems (BMS) to collect macroscopic operating parameters such as voltage, current, and temperature, constructing equivalent circuit models or data-driven models to assess battery health. However, due to the low Young's modulus of sulfide solid electrolytes, they exhibit significant "soft" mechanical properties. During battery cycling, the volume expansion of electrode active material particles due to lithium insertion / deintercalation can easily lead to microscopic delamination or contact loss at the interface. In the early stages, this microscopic interface delamination typically does not cause obvious anomalies in macroscopic parameters such as voltage, current, or temperature, making it difficult for existing detection methods to effectively perceive the physical contact state between the active material particles and the sulfide solid electrolyte layer, resulting in a significant detection blind spot. Only when interface delamination develops to a certain extent does it manifest as increased battery internal resistance or capacity decay; by this time, the optimal early intervention opportunity has often been missed, affecting battery safety and lifespan. Summary of the Invention

[0004] The present invention provides a battery detection method, apparatus, electronic device, and computer-readable storage medium to overcome or at least partially solve the above-mentioned problems.

[0005] This invention discloses a battery detection method, characterized by comprising: Obtain raw operating data for the battery; Based on the original operating data, the peeling status information of the battery interface is determined, where the interface is the physical contact surface formed between the highly active material particles of the battery and the sulfide solid electrolyte layer. The detection results for the battery are generated based on the stripping status information.

[0006] 2. The method according to claim 1, characterized in that the method is applied to a cloud-based diagnostic ranking platform, the cloud-based diagnostic ranking platform being configured with corresponding vehicles, and the vehicles being configured as follows: The electrical, thermal, and state parameters of the vehicle are obtained when it is in a stationary state or in the relaxation phase after charging is completed; the electrical parameters include the individual cell terminal voltage sequence and charging cut-off voltage of the battery. The voltage relaxation curve after the vehicle charging is completed is generated by using the preprocessed single-cell terminal voltage sequence and the charging cutoff voltage, as well as the decay time constant of the voltage relaxation curve. The voltage relaxation curve, the decay time constant, the electrical parameters, the thermal parameters, and the state parameters are determined as the original operating data; The raw operating data is sent to the cloud-based diagnostic platform; the step of obtaining the raw operating data for the battery includes: Receive raw operating data sent by the vehicle.

[0007] 3. The method according to claim 2, characterized in that the method is applied to a cloud-based diagnostic platform equipped with a sulfide material parameter library, and the step of determining the peeling status information of the battery interface based on the original operating data includes: Assign a unique identifier to the battery of the vehicle; When the unique identifier is used to determine that the battery type is a sulfide solid-state battery system, geometric parameters, electrochemical parameters, mechanical parameters, interface parameters and creep parameters that match the battery are extracted from the sulfide material parameter library; The electrochemical parameters are corrected by the decay time constant, and the lithium-ion concentration distribution is calculated based on the decay time constant, the corrected electrochemical parameters, the thermal parameters, and the state parameters; the lithium-ion concentration distribution is used to characterize the spatial distribution of lithium ions inside the electrode active particles at the target charge and discharge time. When the lithium ion concentration distribution is determined to meet the preset conditions by the voltage relaxation curve, the interfacial contact pressure distribution used to characterize the degree of adhesion of the interface is calculated by the lithium ion concentration distribution, the electrochemical parameters, and the mechanical parameters. When the interface contact pressure distribution is less than or equal to zero, the interface is determined to be in a microscopic peeling state, and the interface separation displacement result, which reflects the degree of physical separation of the interface, is determined. Based on the interface separation displacement results and the interface parameters, the peeling state information is determined.

[0008] 4. The method according to claim 3, characterized in that the electrochemical parameter includes a solid-phase diffusion coefficient, the state parameter includes current state of charge information, and the step of calculating the lithium-ion concentration distribution using the decay time constant, the electrochemical parameter, the thermal parameter, and the state parameter includes: The decay time constant is determined as a reference feature; Substitute the solid-phase diffusion coefficient into the modified Arrhenius equation formula constructed based on the reference feature, use the thermal parameters as temperature compensation, and combine the reference feature to perform gain correction on the solid-phase diffusion coefficient to obtain the corrected diffusion coefficient. The modified diffusion coefficient and the current state of charge information are used as initial boundary conditions and substituted into the Fick second law diffusion equation to solve for the lithium ion concentration distribution.

[0009] 5. The method according to claim 3, characterized in that the electrochemical parameters include an initial equilibrium lithium concentration, the mechanical parameters include a partial molar volume and a stiffness matrix, and the step of calculating the interfacial contact pressure distribution characterizing the degree of adhesion of the interface using the lithium ion concentration distribution, the electrochemical parameters, and the mechanical parameters when the lithium ion concentration distribution is determined to meet the preset conditions by the voltage relaxation curve includes: Generate a simulated voltage based on the lithium ion concentration distribution state, and calculate the residual between the simulated voltage and the voltage relaxation curve; When the residual is less than a preset threshold, the partial molar volume is called, and combined with the lithium ion concentration distribution inside the electrode and the initial equilibrium lithium concentration, the micro-strain data generated by chemical expansion is calculated according to the constitutive equation. Using the stiffness matrix and the micro-strain data, a multi-physics coupling solution is performed based on the mechanical equilibrium equation and the total stress relationship to reconstruct the three-dimensional stress tensor inside the battery. Determine the unit normal vector of the interface, perform a dot product operation on the unit normal vector to generate a traction force vector, and perform a second dot product projection on the unit normal vector and the traction force vector to generate a normal stress perpendicular to the interface. The normal stress is converted into the interface contact pressure distribution.

[0010] 6. The method according to claim 3, wherein the interface parameters include damage initiation displacement and complete failure displacement parameters, and the step of determining the peeling state information based on the interface separation displacement result and the interface parameters includes: The damage initiation displacement parameter, the complete failure displacement parameter, and the interface separation displacement result are substituted into the bilinear cohesive model to calculate the damage variable describing the degree of degradation of the micro-interface properties. The damage variable is determined as the peeling state information.

[0011] 7. The method according to claim 3, wherein the geometric parameters include the total interface area, and the step of generating a detection result for the battery using the peeling state information includes: The percentage of the interface failure area is calculated using the total area of ​​the interface and the damage variable. The contact failure rate of the interface is calculated based on the proportion of the failed area, and the fault level is determined based on the contact failure rate. The failure rate of the interface is used to determine the fault level and generate the detection result.

[0012] 8. The method according to claim 3, wherein the geometric parameters include the electrolyte layer thickness, the creep parameters include the creep constant, the stress exponent, and the activation energy coefficient, and the method further comprises: The average crack width is determined based on the interface separation displacement results, and the target repair strain is calculated in conjunction with the electrolyte layer thickness. The target repair strain is used to characterize the degree of plastic deformation that the sulfide electrolyte material of the battery must undergo in order to close the crack. Substituting the creep constant, the stress exponent, the activation energy coefficient, and the normal stress into the Norton power law equation, the viscous flow rate of the sulfide electrolyte material under hot-pressing conditions is calculated. An optimization function is constructed with the goal of minimizing repair time. The function is then solved in reverse under the constraints of temperature and pressure that meet the material safety threshold to generate a target maintenance parameter set. The target maintenance parameter set includes at least the target temperature, target pressure, and duration. The target temperature, target pressure, and duration are encapsulated in an electronic work order and sent to the offline maintenance service terminal corresponding to the cloud-based diagnostic platform.

[0013] This invention also discloses an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; The memory is used to store computer programs; When the processor executes a program stored in the memory, it implements the method described in the embodiments of the present invention.

[0014] This invention also discloses a computer-readable storage medium storing instructions that, when executed by one or more processors, cause the processors to perform the methods described in this invention.

[0015] The embodiments of the present invention have the following advantages: By deeply analyzing macroscopic raw operating data into microscopic interface delamination state information unique to sulfide solid-state batteries, this invention successfully established a quantitative mapping system from external electrical signals to internal solid-solid contact quality. This fundamentally solves the problem of detection blind spots caused by the "soft mechanical" characteristics of sulfide electrolytes, where early microscopic damage shows no significant abnormalities in macroscopic dimensions such as voltage and current. It enables early and accurate perception of battery interface failure and transparent assessment of battery health, thus providing a scientific basis for timely maintenance before the optimal intervention time is missed. Attached Figure Description

[0016] Figure 1 This is a flowchart of the steps of a battery detection method provided in an embodiment of the present invention; Figure 2 This is a schematic flowchart of a battery detection method provided in an embodiment of the present invention; Figure 3 This is a hardware structure block diagram of an electronic device provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of a computer-readable medium provided in an embodiment of the present invention. Detailed Implementation

[0017] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] Reference Figure 1 The diagram illustrates a flowchart of a battery detection method provided in an embodiment of the present invention, which may specifically include the following steps: Step 101: Obtain raw operating data for the battery; Step 102: Determine the peeling status information of the battery interface based on the original operating data. The interface is the physical contact surface formed between the highly active material particles of the battery and the sulfide solid electrolyte layer. Step 103: Generate detection results for the battery using the stripping status information.

[0019] The embodiments of the present invention can be applied to a cloud-based diagnostic testing platform, which can be a cloud server specifically designed for processing battery testing data. The cloud-based diagnostic testing platform can interact with vehicles for data exchange. In specific implementation, the embodiments of the present invention can acquire the original operating data of the battery, so as to obtain macroscopic dynamic signals that reflect the real-time internal state of the battery from the physical entity through non-invasive means, providing the most basic input source and boundary conditions for subsequent microscopic physical field reconstruction.

[0020] The battery can be a non-vehicle battery or a vehicle battery. Optionally, it can be a solid-state energy storage unit consisting of highly active material particles and a sulfide solid electrolyte.

[0021] Raw operating data refers to the collection of physical quantities generated by the battery during actual operation or rest, without extensive processing, including terminal voltage, current, temperature, state of charge (SOC), and state of health (SOH).

[0022] In a specific implementation, embodiments of the present invention can determine the peeling state information of the battery interface based on the original operating data. The interface is the physical contact surface formed between the highly active material particles of the battery and the sulfide solid electrolyte layer. By utilizing cross-scale physical modeling, macroscopic electrical signals are converted into microscopic mechanical responses. By calculating the invisible interface forces and deformations, the degree of solid-solid contact failure can be accurately identified and quantified.

[0023] The interface refers to the solid-solid physical contact surface formed between the highly active material particles and the sulfide solid electrolyte layer, which is the microscopic channel for lithium-ion transport.

[0024] Highly active material particles are microscopic material particles in the positive or negative electrode of a battery that are responsible for storing and releasing lithium ions.

[0025] The sulfide solid electrolyte layer is located between the positive and negative electrodes or fills the spaces between particles. It is a solid dielectric layer made of sulfide materials that serves to isolate electrons and conduct lithium ions.

[0026] Peeling status information refers to quantitative data characterizing the degree of interface adhesion or separation, including interface contact pressure distribution, separation displacement, damage variables, and failure rate.

[0027] For example, the acquired raw operating data can be input into a pre-built cloud-based digital twin model. By extracting the differential features of the terminal voltage sequence (such as the dV / dQ curve), a feature map is generated. A deep learning model (such as a convolutional neural network CNN) is then used to compare the feature map with a pre-stored sulfide interface damage database, directly mapping and outputting the peeling state information of the battery interface.

[0028] The purpose of inputting the acquired raw operational data into a pre-built cloud-based digital twin model is to transform fragmented sensor data into a feature set with a physical and logical context. As a virtual "digital mirror," the digital twin model can simulate the response patterns of a battery under different health conditions, thereby providing a high-quality training and comparison benchmark for subsequent AI feature extraction.

[0029] Raw operating data refers to unfiltered battery physical parameters, such as voltage, current, temperature, and their raw sequences of changes over time, collected in real time from the vehicle.

[0030] A cloud-based digital twin model refers to a digital simulation model built on a cloud server that closely matches the physical characteristics of a physical battery, used to simulate the electrochemical behavior inside the battery.

[0031] The purpose of generating feature maps by extracting differential features from the terminal voltage sequence is to achieve "feature dimensionality reduction" and "signal amplification". By performing differential operations on the voltage sequence, minute and imperceptible interfacial contact fluctuations are transformed into significant peak changes, thereby exposing characteristic signals that reflect the physical state of the interface.

[0032] A terminal voltage sequence refers to an ordered set of data showing the voltage values ​​observed externally during the charging and discharging process of a battery, arranged over time or by capacity. Differential characteristics refer to physical features reflecting the rate of voltage change obtained through mathematical differentiation (such as dV / dQ or dV / dt), which typically correspond to phase transitions or kinetic inhibition within the material.

[0033] Feature mapping refers to arranging extracted features in the form of images or multi-dimensional vectors to form a fingerprint-like identification sample, which is then used by computers for pattern recognition.

[0034] The purpose of using a deep learning model to compare this feature map with a pre-stored sulfide interface damage database is to find the correlation between the current battery state and known failure modes through nonlinear pattern matching. By comparing large datasets, the degree of interface damage can be quickly located, bypassing complex mechanical derivations.

[0035] Deep learning models refer to multi-layered artificial neural networks (such as CNNs or Transformers) that have the ability to automatically extract and learn abstract feature patterns from complex graphs.

[0036] The sulfide interface damage database refers to a large sample library accumulated in advance through experiments or high-precision simulations, which records different degrees of sulfide interface delamination (cracks, voids) and corresponding electrical signal characteristics.

[0037] The purpose of directly mapping the stripping status information of the battery interface is to complete the final jump from the "signal space" to the "physical state space", output quantitative or qualitative diagnostic data, and inform the user interface of the specific degree of detachment.

[0038] The stripping status information refers to an indicator describing the degree of separation between highly active material particles and sulfide electrolytes. In this example, it is represented by the "stripping percentage" or "contact degradation level" predicted by the AI ​​model.

[0039] By combining deep learning and digital twin technology, the real-time reconstruction process of the mechanical field, which requires extremely high computing power inside the battery, is avoided. It can leverage the processing advantages of big data in the cloud to achieve rapid, batch, and automatic identification of the interface peeling state of sulfide solid batteries.

[0040] Of course, the above examples are merely illustrative. Those skilled in the art can use any other methods to determine the peeling status information of the battery interface based on the original operating data. In this regard, the embodiments of the present invention do not impose any limitations.

[0041] In a specific implementation, the embodiments of the present invention generate detection results for the battery through the stripping state information, so as to transform complex microscopic physical parameters into intuitive diagnostic conclusions, realize qualitative evaluation and classification of the internal health status of solid-state batteries, and thus guide subsequent maintenance decisions.

[0042] The test results are the final diagnostic conclusions based on the stripping status information, including whether the interface has malfunctioned, the severity of the malfunction, and the corresponding malfunction level information.

[0043] This invention, through in-depth analysis of macroscopic raw operating data into microscopic interface delamination state information unique to sulfide solid-state batteries, successfully establishes a quantitative mapping system from external electrical signals to internal solid-solid contact quality. This fundamentally solves the problem of detection blind spots caused by the "soft mechanical" characteristics of sulfide electrolytes, where early microscopic damage shows no significant abnormalities in macroscopic dimensions such as voltage and current. It enables early and accurate perception of battery interface failure and transparent assessment of battery health, thus providing a scientific basis for timely maintenance before the optimal intervention time is missed.

[0044] Based on the above embodiments, modified embodiments of the above embodiments are proposed. It should be noted that, in order to keep the description brief, only the differences from the above embodiments are described in the modified embodiments.

[0045] In an optional embodiment of the present invention, the method is applied to a cloud-based diagnostic ranking platform, which is configured with a corresponding vehicle, the vehicle being configured as follows: The electrical, thermal, and state parameters of the vehicle are obtained when it is in a stationary state or in the relaxation phase after charging is completed; the electrical parameters include the individual cell terminal voltage sequence and charging cut-off voltage of the battery. The voltage relaxation curve after the vehicle charging is completed is generated by using the preprocessed single-cell terminal voltage sequence and the charging cutoff voltage, as well as the decay time constant of the voltage relaxation curve. The voltage relaxation curve, the decay time constant, the electrical parameters, the thermal parameters, and the state parameters are determined as the original operating data; The raw operating data is sent to the cloud-based diagnostic platform; the step of obtaining the raw operating data for the battery includes: Receive raw operating data sent by the vehicle.

[0046] In a specific implementation, this optional embodiment can be applied to a vehicle, which can interact with a cloud-based diagnostic platform. The vehicle is configured to acquire electrical, thermal, and state parameters when it is in a static state or in the relaxation phase after charging is completed, so as to achieve high-precision acquisition of data when the vehicle is in a static or relaxation phase. This allows for the acquisition of macroscopic dynamic signals that reflect the internal state of the battery from the physical entity, providing an accurate input source and calibration benchmark for subsequent cloud-based microscopic interface diagnostics.

[0047] Electrical parameters: battery cell terminal voltage sequence V(t), charging cutoff voltage Vcutoff, and charging / discharging circuit current sequence I(t).

[0048] Thermal parameter: Internal temperature of a single battery cell, Tcell.

[0049] State parameters: Current state of charge (SOC), initial battery health state (SOH0).

[0050] In a specific implementation, the embodiments of the present invention generate a voltage relaxation curve after the vehicle charging is completed by using a preprocessed single-unit terminal voltage sequence and a charging cutoff voltage, as well as the decay time constant of the voltage relaxation curve. The voltage relaxation curve, decay time constant, electrical parameters, thermal parameters, and state parameters are then determined as raw operating data and sent to a cloud diagnostic platform to provide a highly reliable calibration basis for the cloud model by utilizing the dynamic characteristics of the relaxation process.

[0051] Preprocessing operations may include filtering to eliminate high-frequency noise, etc. The voltage relaxation curve refers to the decay curve of the battery voltage from the excited state back to the equilibrium state after charging is cut off, Vrelax(t) = V(t) - Vcutoff.

[0052] The decay time constant is a key characteristic parameter that characterizes the rate of decay of the voltage relaxation curve and is used for subsequent correction of electrochemical parameters.

[0053] This invention achieves reliable acquisition and efficient transmission of macroscopic operational data through a collaborative approach of lightweight vehicle-side data acquisition and cloud-based reception. This provides the necessary data foundation for subsequent microscopic interface peeling state diagnosis based on physical mechanisms and solves the problem of insufficient vehicle-side computing power making it difficult to directly perform complex modeling.

[0054] In an optional embodiment of the present invention, the method is applied to a cloud-based diagnostic platform equipped with a sulfide material parameter library, and the step of determining the peeling status information of the battery interface based on the original operating data includes: Assign a unique identifier to the battery of the vehicle; When the unique identifier is used to determine that the battery type is a sulfide solid-state battery system, geometric parameters, electrochemical parameters, mechanical parameters, interface parameters and creep parameters that match the battery are extracted from the sulfide material parameter library; The electrochemical parameters are corrected by the decay time constant, and the lithium-ion concentration distribution is calculated based on the decay time constant, the corrected electrochemical parameters, the thermal parameters, and the state parameters; the lithium-ion concentration distribution is used to characterize the spatial distribution of lithium ions inside the electrode active particles at the target charge and discharge time. When the lithium ion concentration distribution is determined to meet the preset conditions by the voltage relaxation curve, the interfacial contact pressure distribution used to characterize the degree of adhesion of the interface is calculated by the lithium ion concentration distribution, the electrochemical parameters, and the mechanical parameters. When the interface contact pressure distribution is less than or equal to zero, the interface is determined to be in a microscopic peeling state, and the interface separation displacement result, which reflects the degree of physical separation of the interface, is determined. Based on the interface separation displacement results and the interface parameters, the peeling state information is determined.

[0055] In this embodiment of the invention, a unique identifier can be assigned to the battery of the vehicle to establish a one-to-one digital mapping relationship with each physical battery entity on the cloud diagnostic platform, ensuring that the material parameters and calculation models called in the subsequent calls are targeted and unique.

[0056] A unique identifier is a unique numerical code (such as a VIN or battery ID) assigned to a specific battery cell or battery pack, used to retrieve and index its unique physical characteristic parameters.

[0057] In a specific implementation, when the unique identifier is used to determine that the battery type is a sulfide solid-state battery system, the embodiments of the present invention can extract geometric parameters, electrochemical parameters, mechanical parameters, interface parameters and creep parameters that match the battery from the sulfide material parameter library, so as to automatically identify the technical route of the battery, and call up the various basic physical properties required to support multi-physics coupling simulation from the preset database, so as to provide a data benchmark for high-precision microstate restoration.

[0058] The sulfide material parameter library is a structured database stored on a cloud-based diagnostic platform, containing the physical properties of various sulfide solid electrolytes and their matching electrode materials.

[0059] Geometric parameters describe the dimensional information of the battery's internal microstructure, such as electrode thickness and active particle radius.

[0060] Electrochemical parameters characterize the electrochemical kinetic properties of materials, such as solid-phase diffusion coefficient and initial equilibrium lithium concentration.

[0061] Mechanical parameters are constants that describe the deformation behavior of materials under stress, such as stiffness matrix and partial molar volume.

[0062] Interface parameters specifically define the numerical values ​​of interface mechanical failure characteristics, such as damage initiation displacement and complete failure displacement.

[0063] Creep parameters are characteristic parameters that describe the viscous flow of sulfide materials over time under stress, such as creep constant and stress exponent.

[0064] Geometric parameters: electrode particle radius Rp, electrode thickness Lelec, current collector thickness.

[0065] Electrochemical parameters: solid-phase diffusion coefficient Ds, maximum lithium concentration cmax, initial equilibrium lithium concentration c0.

[0066] Mechanical parameters: a) Sulfide electrolytes (such as LGPS and LPS) typically have a low modulus (approximately 10-30 GPa), which is much lower than that of oxide electrolytes, exhibiting certain "soft" characteristics.

[0067] b) Poisson's ratio ν, partial molar volume Ω.

[0068] Interface parameters: Interfacial bonding strength Tmax and critical separation displacement δc. The interface between sulfide electrolytes and electrodes is typically a solid-solid physical contact and is sensitive to interfacial stress.

[0069] Creep parameters: Creep constant A, stress exponent n, activation energy Q.

[0070] In a specific implementation, embodiments of the present invention can correct the electrochemical parameters using the decay time constant, and calculate the lithium-ion concentration distribution based on the decay time constant, the corrected electrochemical parameters, the thermal parameters, and the state parameters. This allows for online calibration of the theoretical model using measured dynamic characteristics, eliminating deviations caused by individual battery aging differences, and thus accurately reconstructing the true spatial distribution of lithium ions within the active particles.

[0071] The decay time constant is a time characteristic quantity extracted from the terminal voltage relaxation process that reflects the speed of electrochemical reaction.

[0072] Thermal parameters are real-time collected data on the internal temperature of the battery or the ambient temperature.

[0073] State parameters are variables that describe the current operating state of a battery, such as state of charge (SOC).

[0074] Lithium-ion concentration distribution is a characterization of the spatial density distribution of lithium ions at various locations within the active particles of an electrode at a specific charge and discharge time.

[0075] In a specific implementation, when the lithium ion concentration distribution is determined to meet the preset conditions by the voltage relaxation curve, the interface contact pressure distribution used to characterize the degree of adhesion of the interface is calculated by the lithium ion concentration distribution, the electrochemical parameters and the mechanical parameters, so as to establish a physical mapping from the "concentration field" to the "stress field", and verify the accuracy of the simulation results by the measured curve, thereby deriving the most realistic microscopic stress state at the interface.

[0076] The voltage relaxation curve is the evolution curve of the battery terminal voltage as it recovers equilibrium over time after the battery current is cut off. It is used to verify the fitting accuracy of the simulation model.

[0077] The preset condition can refer to the residual between the simulated voltage sequence and the measured voltage relaxation curve being within the allowable error range.

[0078] The interfacial contact pressure distribution characterizes the magnitude of the compressive force at each point on the contact surface between the highly active material particles and the sulfide solid electrolyte. Pressure loss is a prerequisite for determining delamination.

[0079] When the interface contact pressure distribution is less than or equal to zero, it signifies the following three levels of physical conditions: 1. At the physical level, the transition from "compression" to "separation" between electrode particles and the sulfide electrolyte can be expressed as follows: Under normal conditions, in practical applications, the electrode particles and the sulfide electrolyte are tightly compressed together, and the interface should exhibit positive contact pressure (i.e., interface contact pressure distribution Pc>0), ensuring a tight bond between the solid and solid interfaces. When Pc=0, it means that the compressive force at the interface completely disappears, and the electrode particles and electrolyte are in a critical state of "just in contact but not exerting force on each other." When Pc<0, in physical and mechanical calculations, negative pressure is equivalent to tensile stress. Since the electrode and electrolyte are connected only by weak intermolecular forces (cohesive forces), a negative value means that due to the volume contraction of the electrode particles, an outward "pulling" force is generated, causing the originally tightly connected solid-solid interface to be physically pulled apart.

[0080] 2. At the electrochemical level, the "fracture" of ion transport channels in sulfide solid-state batteries relies on direct solid-solid contact to conduct lithium ions. This means the formation of "microscopic cracks." When Pc ≤ 0, the interface is no longer an atomic-level contact, but rather a nanometer or even micrometer-level gap. Furthermore, this also means a surge in local resistance, preventing lithium ions from hopping through air or vacuum gaps. Therefore, the region with pressure less than zero effectively becomes a dead zone for ion transport. This leads to extremely uneven current distribution at the interface, potentially causing localized overheating or lithium dendrite growth.

[0081] 3. At the diagnostic logic level, this means that the starting point for determining a "stripping fault" is within the algorithm flow; it's the switch that jumps from "normal operation" to "fault diagnosis." Once the system detects that point Pc≤0, the program will initiate subsequent interface separation displacement δ calculations to measure the extent of the gap opening. Damage quantification: The system will calculate the area of ​​all regions with Pc≤0. If the sum of these areas exceeds a certain proportion (i.e., the failure rate η is too high), a formal fault warning or maintenance work order will be issued in the cloud.

[0082] In a specific implementation, embodiments of the present invention can determine that the interface is in a microscopic peeling state when the interface contact pressure distribution is less than or equal to zero, and determine the interface separation displacement result that reflects the degree of physical separation of the interface, so as to identify the loss of interface contact due to volume shrinkage and quantify the specific spatial geometric dimensions of crack opening.

[0083] Microscopic separation refers to the physical detachment of active particles from electrolytes at the microscopic level due to uneven stress or volume changes.

[0084] The interface separation displacement result refers to the width of the physical gap that forms between the particle surface and the electrolyte surface after the interface peels off.

[0085] In specific implementation, embodiments of the present invention can determine the peeling state information based on the interface separation displacement result and the interface parameters, so as to calculate the final damage index by combining the geometric separation amount and the mechanical failure limit of the material itself, and complete a comprehensive quantitative evaluation of the interface failure.

[0086] Stripped state information refers to a collection of data such as displacement and damage variables, used to accurately describe the nature, extent, and degree of degradation of interface damage.

[0087] In this embodiment of the invention, through online calibration of cloud-based parameter library and measured attenuation characteristics, combined with a fitting verification mechanism based on voltage relaxation curve, a layer-by-layer accurate mapping of the micro-interface of sulfide solid-state batteries from electrochemical concentration distribution to mechanical peeling displacement is achieved. This significantly improves the accuracy and physical authenticity of the determination of internal solid-solid contact failure under weak fluctuations in macroscopic signals.

[0088] Optionally, the electrochemical parameters include the solid-phase diffusion coefficient, the state parameters include current state of charge information, and the step of calculating the lithium-ion concentration distribution using the decay time constant, the electrochemical parameters, the thermal parameters, and the state parameters includes: The decay time constant is determined as a reference feature; Substitute the solid-phase diffusion coefficient into the modified Arrhenius equation formula constructed based on the reference feature, use the thermal parameters as temperature compensation, and combine the reference feature to perform gain correction on the solid-phase diffusion coefficient to obtain the corrected diffusion coefficient. The modified diffusion coefficient and the current state of charge information are used as initial boundary conditions and substituted into the Fick second law diffusion equation to solve for the lithium ion concentration distribution.

[0089] In this embodiment of the invention, the decay time constant can be determined as a reference feature to transform the dynamic time index extracted from the macroscopic voltage signal into a key physical feature that can characterize the current dynamic decay degree of the battery, providing a reference operator for subsequent online calibration of dynamic parameters.

[0090] The decay time constant refers to the characteristic time it takes for the terminal voltage of a battery to recover to an equilibrium state during the voltage relaxation phase after the current is cut off, reflecting the speed at which internal polarization is eliminated.

[0091] The baseline characteristic is a characteristic indicator used as a calibration reference to indicate the current electrochemical reactivity of the battery.

[0092] In this embodiment of the invention, the solid-phase diffusion coefficient can be substituted into the modified Arrhenius equation constructed based on the reference features. The thermal parameters are used as temperature compensation, and the solid-phase diffusion coefficient is modified by combining the reference features to obtain the modified diffusion coefficient. This comprehensively considers the influence of temperature on ion migration and the attenuation differences of individual batteries, and performs dual calibration of the theoretical diffusion model to obtain the kinetic parameters that best match the actual physical state of the current battery.

[0093] The solid-phase diffusion coefficient characterizes the ability of lithium ions to diffuse in the solid phase within active material particles and is a core physical constant that determines the concentration gradient distribution.

[0094] The Arrhenius equation correction formula is a mathematical model built on the Arrhenius empirical formula to describe the exponential relationship between diffusion capacity and temperature, and a benchmark feature is introduced to correct for deviations.

[0095] Thermal parameters are the battery temperature collected in real time by the sensor, used to determine the basic thermodynamic performance of the diffusion coefficient under the current environment.

[0096] Temperature compensation utilizes thermal parameters to eliminate the interference caused by ambient temperature fluctuations in the calculation of the diffusion coefficient.

[0097] Gain correction is achieved by proportionally adjusting the diffusion coefficient using a baseline characteristic to compensate for changes in kinetic performance caused by battery aging or batch variations.

[0098] The corrected diffusion coefficient is the final calculated value that accurately reflects the ion migration capability inside the battery at the current moment after temperature compensation and aging gain correction.

[0099] In this embodiment of the invention, the modified diffusion coefficient and the current state of charge information can be determined as initial boundary conditions and substituted into the Fick second law diffusion equation to solve for the lithium ion concentration distribution. Based on the known mass migration capacity and total amount of material, the distribution details of lithium ions inside the active particles can be restored in the spatial dimension by solving the mass transport differential equation, providing input for subsequent mechanical deformation calculations.

[0100] The current state of charge (SOC) information represents the current macroscopic level of charge in the battery, and mathematically corresponds to the total mass constraint of lithium ions inside the particles.

[0101] Initial boundary conditions are a necessary prerequisite for mathematical solutions, including the ion flux on the particle surface and the initial concentration state.

[0102] Fick's second law diffusion equation is a partial differential equation describing the change of substance concentration with time and spatial location, and is a fundamental physical law for simulating the lithium-ion migration process.

[0103] The lithium-ion concentration distribution is the final spatial function of lithium-ion concentration along the radius of the active particles, which directly determines the amount of microscopic chemical expansion of the material.

[0104] For example, in a specific implementation, the decay time constant τ is used as a reference characteristic, which can be used to calibrate or determine the value of Dref (i.e., to perform gain correction or calibration on Dref) so that the reference diffusion coefficient is more consistent with the actual state of the current battery. Since the solid-phase diffusion coefficient Ds is affected by temperature T, the Arrhenius equation correction formula is as follows:

[0105] D s : Solid diffusion coefficient, representing the actual diffusion coefficient at the current temperature T.

[0106] Dref: Reference diffusion coefficient, which is the diffusion coefficient at a reference temperature Tref. It is a known reference value and is calibrated or determined by the decay time constant τ, which is used as a reference characteristic.

[0107] Ea: Activation energy, measured in J / mol. It represents the energy barrier that lithium-ion diffusion needs to overcome; the higher the value, the more significant the effect of temperature on the diffusion coefficient.

[0108] R: Universal gas constant, R = 8.314 J / (mol·K).

[0109] Tref: Reference temperature, in Kelvin (K), usually taken as a standard temperature (e.g., 298K).

[0110] T: Current temperature, i.e., the actual operating temperature of the battery, in Kelvin (K).

[0111] exp[·]: natural exponential function.

[0112] Furthermore, the lithium ion concentration distribution cs(r,t) inside the electrode particles can be solved using Fick's second law. The diffusion equation for the second law is as follows: Cs: Solid phase lithium concentration, usually expressed in mol / m³. It represents the lithium ion concentration at a specific location within the electrode active material particles.

[0113] t: time, in seconds (s).

[0114] Cs / t: The rate of change of lithium ion concentration over time (partial derivative of concentration with respect to time), reflecting the evolution of concentration over time.

[0115] The Nabla operator (gradient / divergence operator) represents spatial derivative operations in three-dimensional space.

[0116] D sSolid phase diffusion coefficient, measured in m² / s. It represents the diffusion capacity of lithium ions within the active material particles of the electrode and is one of the most important material parameters in this formula.

[0117] c s The spatial gradient of lithium-ion concentration reflects the degree of non-uniformity of concentration in space.

[0118] ·(D s c s ): Divergence of diffusion flux, representing the net inflow / outflow of lithium ions per unit volume due to diffusion.

[0119] cs(r, t): The lithium-ion concentration distribution function inside the electrode particle, where r represents the radial coordinate inside the particle and t is time.

[0120] In this embodiment of the invention, by introducing the measured decay time constant as a calibration operator and combining it with a temperature compensation mechanism to perform real-time gain correction on the solid-phase diffusion coefficient, the problem of concentration reconstruction deviation caused by the fixed diffusion parameters in the traditional model is effectively solved, and high-precision, dynamic identification of the lithium ion concentration distribution inside the active particles is achieved.

[0121] Optionally, the electrochemical parameters include the initial equilibrium lithium concentration, and the mechanical parameters include the partial molar volume and stiffness matrix. The step of calculating the interfacial contact pressure distribution, characterizing the degree of adhesion of the interface, based on the lithium ion concentration distribution, the electrochemical parameters, and the mechanical parameters, when the lithium ion concentration distribution is determined to meet preset conditions by the voltage relaxation curve, includes: Generate a simulated voltage based on the lithium ion concentration distribution state, and calculate the residual between the simulated voltage and the voltage relaxation curve; When the residual is less than a preset threshold, the partial molar volume is called, and combined with the lithium ion concentration distribution inside the electrode and the initial equilibrium lithium concentration, the micro-strain data generated by chemical expansion is calculated according to the constitutive equation. Using the stiffness matrix and the micro-strain data, a multi-physics coupling solution is performed based on the mechanical equilibrium equation and the total stress relationship to reconstruct the three-dimensional stress tensor inside the battery. Determine the unit normal vector of the interface, perform a dot product operation on the unit normal vector to generate a traction force vector, and perform a second dot product projection on the unit normal vector and the traction force vector to generate a normal stress perpendicular to the interface. The normal stress is converted into the interface contact pressure distribution.

[0122] In a specific implementation, embodiments of the present invention can generate a simulated voltage based on the lithium ion concentration distribution state, and calculate the residual between the simulated voltage and the voltage relaxation curve to establish a feedback mechanism between virtual simulation and the real physical world. By comparing the simulated value with the measured value, the physical accuracy of the aforementioned electrochemical calculation results can be verified.

[0123] The simulated voltage is the theoretical voltage response derived by back-calculating the calculated lithium-ion concentration distribution using electrochemical kinetic equations (such as the Butler-Volmer equation).

[0124] The voltage relaxation curve is a sequence of real-world observational data showing how the battery's terminal voltage recovers to equilibrium over time after the current is cut off.

[0125] The residual is the set of differences between the simulated voltage sequence and the measured voltage relaxation curve at corresponding time points, reflecting the prediction error of the model.

[0126] In a specific implementation, when the residual is less than a preset threshold, the partial molar volume can be invoked, and combined with the lithium ion concentration distribution inside the electrode and the initial equilibrium lithium concentration, the microscopic strain data generated by chemical expansion can be calculated according to the constitutive equation. Based on confirming the accuracy of the electrochemical field, the "concentration field" can be mapped to the "strain field", quantifying the microscopic geometric size changes of the electrode particles caused by lithium insertion / extraction.

[0127] The preset threshold is the maximum allowable error limit set and used to determine whether the model passes validation.

[0128] The initial equilibrium lithium concentration is the baseline lithium-ion concentration of the electrode material under strain-free, fully balanced conditions.

[0129] Partial molar volume is the change in macroscopic volume of a substance caused by the entry of lithium ions of a unit molar concentration into the active material.

[0130] Constitutive equations are equations that describe the physical relationship between strain and concentration changes in a material.

[0131] Microscopic strain data are used to characterize the relative deformation of a material due to lattice expansion or contraction.

[0132] In a specific implementation, the embodiments of the present invention can use the stiffness matrix and the micro-strain data to perform multi-physics field coupling solution based on the mechanical equilibrium equation and the total stress relationship, reconstruct the three-dimensional stress tensor inside the battery, so as to comprehensively consider the mechanical strength and deformation constraints of the material, solve the complex stress state inside the battery, and realize the transformation from deformation to internal force.

[0133] The stiffness matrix is ​​a set of multidimensional constants used to characterize the elastic properties of a material, defining the proportional relationship between stress and strain.

[0134] The mechanical equilibrium equation is a differential equation that describes the stress distribution inside an object and requires it to satisfy static equilibrium (i.e., the resultant force is zero).

[0135] The total stress formula is a complete stress expression formula that integrates factors such as elastic strain and chemical strain.

[0136] The three-dimensional stress tensor is a complete set of quantities that describes the mechanical states (including pushing, pulling, shearing, etc.) of a point in space in all directions.

[0137] In a specific implementation, the embodiments of the present invention can determine the unit normal vector of the interface, perform a dot product operation on the unit normal vector to generate a traction force vector, and perform a second dot product projection on the unit normal vector and the traction force vector to generate a normal stress perpendicular to the interface, so as to accurately peel out the key force component that directly acts on the contact surface and has a peeling tendency from the complex stress field in three-dimensional space.

[0138] The unit normal vector is a vector perpendicular to the contact surface with a length of 1 unit, used to define the spatial orientation of the interface.

[0139] The dot product is an algebraic operation between vectors used to calculate the effect of the stress tensor on a specific plane.

[0140] The traction force vector is the total force vector acting on a specific interface.

[0141] The second dot product projection extracts the projection value of the traction force in the normal direction through mathematical operations.

[0142] Normal stress is the mechanical component that acts perpendicularly on the interface and is the core physical quantity that determines whether the interface adheres or separates.

[0143] In specific implementations, embodiments of the present invention can transform the normal stress into the interface contact pressure distribution, thereby converting the normal force in mechanical calculations into the intuitive engineering concept of "pressure" to quantitatively evaluate the physical bonding quality of the interface.

[0144] Interface contact pressure distribution is the numerical representation of normal stress on the contact surface. It is usually defined as compressive stress as positive and tensile stress as negative, and is used to visually represent the tightness of the contact surface.

[0145] For example, 1. Analog voltage calculation and residual verification (model validation phase): Simulated voltage Usim=f(cs(r,t),SOC,T) (Specifically, the open-circuit voltage (OCV) is calculated using the Nernst equation or Butler-Volmer dynamics combined with concentration distribution, and then the analog terminal voltage is obtained by superimposing polarization, etc.) in, Usim: Simulated voltage, the theoretical battery voltage calculated based on the current lithium-ion concentration distribution.

[0146] cs(r,t): Lithium ion concentration distribution inside the electrode particles.

[0147] SOC: State of charge.

[0148] T: Temperature.

[0149] Residual = |Usim - Urelax| (Urelax is the measured voltage value of the voltage relaxation curve).

[0150] 2. Chemical expansion strain calculation (constitutive equation): εchem=Ω(c-c0) in, εchem: chemical expansion strain (eigenstrain), the strain caused by volume changes resulting from lithium intercalation.

[0151] Ω: Partial molar volume, in m³ / mol, represents the volume expansion caused by the insertion of one mole of lithium ion into the active material. It is a key coupling parameter connecting electrochemistry and mechanics.

[0152] c: Current lithium-ion concentration (local lithium concentration).

[0153] c0: Initial / reference lithium concentration, corresponding to the concentration under reference conditions.

[0154] 3. Equilibrium equations in mechanics: ·σ+f=0 in, •: Divergence operator.

[0155] σ: Stress tensor, which describes the stress state at various points inside the battery (3×3 matrix).

[0156] f: Body force, which is usually ignored or zero in battery models.

[0157] 4. Total stress relation (constitutive relation) equation: σ=C:(εtotal-εchem) in, σ: Total stress tensor.

[0158] C: Stiffness matrix (elasticity tensor), a fourth-order tensor characterizing a material's resistance to deformation, related to Young's modulus and Poisson's ratio. For sulfide electrolytes, due to their "soft" characteristics, this matrix value is relatively small.

[0159] :: Double dot product (double contraction operation).

[0160] εtotal: Total strain, determined by boundary constraints and overall deformation.

[0161] εchem: Chemical expansion strain (same as above).

[0162] 5. Calculation of interface traction force vector equation: t=σ·n in, t: Traction vector, the force vector acting on the interface.

[0163] σ: Stress tensor (same as above).

[0164] •: Dot product (multiplication of a matrix and a vector).

[0165] n: The unit normal vector of the interface, perpendicular to the interface and pointing towards the electrolyte side.

[0166] 6. Calculation of normal stress (quadratic dot product projection) equation: σn=n·t or σn = n·(σ·n) in, σn: Normal stress, the stress component perpendicular to the interface (positive values ​​usually represent tensile stress).

[0167] 7. Calculation of interfacial contact pressure distribution equation: Pc=-σn in, Pc: Interface contact pressure. In engineering, pressure is defined as a positive value, so the negative value of the normal stress is taken.

[0168] When Pc≤0, it is determined that the interface has been peeled off.

[0169] The embodiments of the present invention ensure the accuracy of the premise of mechanical reconstruction by verifying the simulated residual and the measured residual, and extract the normal force from the three-dimensional stress field by using the precise secondary dot product projection technology. This enables the high-precision reconstruction of the microscopic contact pressure between the highly active material particles and the electrolyte, providing highly realistic physical data support for the subsequent determination of early peeling failure.

[0170] Optionally, the interface parameters include damage initiation displacement and complete failure displacement parameters, and the step of determining the peeling state information based on the interface separation displacement result and the interface parameters includes: The damage initiation displacement parameter, the complete failure displacement parameter, and the interface separation displacement result are substituted into the bilinear cohesive model to calculate the damage variable describing the degree of degradation of the micro-interface properties. The damage variable is determined as the peeling state information.

[0171] In this embodiment of the invention, the damage initiation displacement parameter, the complete failure displacement parameter, and the interface separation displacement result can be substituted into a bilinear cohesive model to calculate damage variables describing the degree of degradation of micro-interface properties. This establishes a nonlinear mapping relationship from "geometric separation amount" to "material physical damage". The mechanical model is used to quantitatively evaluate the degree of performance degradation of the interface material during the peeling process, thereby accurately determining the stage of interface failure.

[0172] The damage initiation displacement parameter refers to the critical displacement threshold at which the highly reactive material particles and the electrolyte are pulled apart when irreversible physical property degradation (such as microcrack initiation) begins at the interface, marking the beginning of damage.

[0173] The complete failure displacement parameter refers to the displacement when the interfacial bonding force is completely lost and the material completely fractures to form macroscopic cracks, marking the end of the material's function.

[0174] The interface separation displacement result is the actual physical opening distance between the two ends of the interface under the current working condition, calculated through the aforementioned steps.

[0175] The bilinear cohesive model is a constitutive model that describes the relationship between force and displacement at a material interface. Its curve characteristics are usually represented by a bilinear shape consisting of an ascending segment (elastic loading) and a descending segment (damage evolution), and it is often used to simulate the delamination process of solid-solid interfaces.

[0176] The damage variable receives a scalar value between 0 (intact) and 1 (complete failure), which is used to quantitatively describe the proportion of the material’s internal microstructure that is damaged.

[0177] The degree of degradation of micro-interface properties is used to characterize the physical process by which the interface's ability to transfer lithium ions and bear mechanical loads gradually weakens with increasing damage.

[0178] In a specific implementation, embodiments of the present invention can determine the damage variable as the peeling state information, so as to encapsulate the complex mechanical evolution calculation results into standard diagnostic information, providing a core basis with high physical generalization for the final generation of detection results.

[0179] The stripping status information refers to a data set centered on damage variables, which directly reflects the quality of contact between highly active material particles and sulfide solid electrolytes.

[0180] For example, if the contact pressure is The cohesive force model is used to quantitatively assess interface damage.

[0181] Detailed calculation formula: a) Interface separation determination and displacement calculation: If a certain area is under contact pressure If the value is less than or equal to 0, the region is determined to have separated, and the separation displacement δ of the interface is obtained through iterative calculation.

[0182] b) Calculation of damage variables: The damage variable D is calculated based on a bilinear cohesive model to describe the process of the interface from intact to fractured:

[0183] in: The maximum separation displacement in the loading history; This is the damage initiation displacement (the critical point at which the interface begins to soften). This is the displacement of complete failure (complete interface breakage).

[0184] Damage variable range: D=0 indicates intact, D=1 indicates complete fracture.

[0185] In this embodiment of the invention, by introducing a bilinear cohesive force model (CZM), simple displacement measurement is elevated to in-depth material damage evaluation. This model can accurately identify the nonlinear dynamic evolution process of the interface from microcrack initiation to complete failure, providing a highly reliable mechanical criterion for quantitatively assessing the structural integrity of the internal interface of sulfide solid-state batteries.

[0186] Optionally, the geometric parameters include the total interface area, and the step of generating a detection result for the battery using the peeling state information includes: The percentage of the interface failure area is calculated using the total area of ​​the interface and the damage variable. The contact failure rate of the interface is calculated based on the proportion of the failed area, and the fault level is determined based on the contact failure rate. The failure rate of the interface is used to determine the fault level and generate the detection result.

[0187] In specific implementations, embodiments of the present invention can use the total area of ​​the interface and the damage variable to statistically determine the proportion of the failed area of ​​the interface, thereby spatially integrating point-like damage data at the micro level, quantifying the physical scale of interface damage, and thus achieving a leap from the evaluation dimension of "point" to "surface".

[0188] The total interface area refers to the theoretically designed total contact area between the highly active material particles and the sulfide solid electrolyte.

[0189] The failure area ratio refers to the ratio of the sum of the areas of tiny units on the interface whose damage variables exceed a preset threshold (determined to have undergone substantial peeling) to the total area of ​​the interface.

[0190] In a specific implementation, embodiments of the present invention can calculate the contact failure rate of the interface based on the proportion of the failure area, and determine the fault level based on the contact failure rate, so as to convert the proportion of physical damage into an engineering reliability indicator, and classify the sub-health or fault state of the battery according to a preset risk threshold, so as to take different levels of intervention measures.

[0191] Contact failure rate is a quantitative indicator used to define the severity of the loss of lithium-ion transport function at the interface due to peeling.

[0192] The fault level is a preset series of health status evaluation gradients (e.g., healthy, mild contact failure, moderate peeling warning, severe failure, etc.).

[0193] In a specific implementation, the embodiments of the present invention can use the contact failure rate to determine the fault level and generate detection results to complete the final step of the diagnostic process, solidify the complex calculation data into an executable decision report, and send it to the management end or user end.

[0194] The test results are the final diagnostic report, which includes the current battery fault level, performance degradation prediction, and instructions on whether to proceed with the repair process.

[0195] For example, the interface contact failure rate η is calculated as follows: To quantify the overall failure rate, the interface contact failure rate η is defined as follows:

[0196] in The total area of ​​the interface. This is the damage threshold (e.g., set to 0.5). This formula calculates the percentage of the area whose damage exceeds the threshold.

[0197] Fault classification: If η>10%, it is judged as a moderate fault (Level L2); If η > 30%, it is judged as a serious fault (Level L3).

[0198] This invention achieves macroscopic quantification of the solid-solid interface contact quality of sulfide solid-state batteries by integrating and statistically analyzing micro-damage variables in the interface space dimension. It can accurately classify the severity level of faults based on the contact failure rate, solving the technical problem that traditional methods cannot qualitatively describe the impact of micro-contact failures on the overall performance of the battery pack.

[0199] In an optional embodiment of the present invention, the geometric parameters include the electrolyte layer thickness, the creep parameters include the creep constant, the stress exponent, and the activation energy coefficient, and the method further includes: The average crack width is determined based on the interface separation displacement results, and the target repair strain is calculated in conjunction with the electrolyte layer thickness. The target repair strain is used to characterize the degree of plastic deformation that the sulfide electrolyte material of the battery must undergo in order to close the crack. Substituting the creep constant, the stress exponent, the activation energy coefficient, and the normal stress into the Norton power law equation, the viscous flow rate of the sulfide electrolyte material under hot-pressing conditions is calculated. An optimization function is constructed with the goal of minimizing repair time. The function is then solved in reverse under the constraints of temperature and pressure that meet the material safety threshold to generate a target maintenance parameter set. The target maintenance parameter set includes at least the target temperature, target pressure, and duration. The target temperature, target pressure, and duration are encapsulated in an electronic work order and sent to the offline maintenance service terminal corresponding to the cloud-based diagnostic platform.

[0200] In practical applications, embodiments of the present invention can determine the average crack width based on the interface separation displacement results, and calculate the target repair strain in combination with the electrolyte layer thickness, so as to transform the diagnosed geometric damage into mechanical repair task indicators. By calculating the relative deformation of the gaps to be filled by the material, a quantitative benchmark is provided for the subsequent precise setting of the repair intensity.

[0201] The average crack width is the crack space width obtained by statistically averaging multiple separation displacement points in a local area.

[0202] The electrolyte layer thickness is the nominal thickness of the sulfide solid electrolyte layer in the structural design, and it belongs to the geometric parameters.

[0203] The target repair strain is a key repair indicator defined by the definition, which characterizes the relative displacement ratio that the electrolyte material needs to produce on a macroscopic scale in order to physically close the crack.

[0204] The degree of plastic deformation is the amount of permanent, irreversible deformation that a material undergoes after being subjected to stress; here, it specifically refers to the ability to fill cracks through viscous flow.

[0205] In practical applications, embodiments of the present invention can substitute the creep constant, the stress exponent, the activation energy coefficient, and the normal stress into the Norton power law equation to calculate the viscous flow rate of the sulfide electrolyte material under hot and pressure conditions. This allows the use of the unique "soft" mechanical properties (creep) of sulfide materials to establish a functional relationship between stress, temperature, and material flow rate, thereby enabling the prediction of the rate of deformation of the material under heat and pressure.

[0206] Creep constant, stress exponent, and activation energy coefficient are the core constitutive parameters that describe the slow permanent deformation (creep) behavior of materials under high temperature and high pressure.

[0207] Normal stress is the normal pressure acting on the interface and is the mechanical force driving the creep flow of materials.

[0208] Norton's power law is a mathematical equation that describes the change in creep rate of a solid material with temperature and pressure under constant stress.

[0209] Viscous flow rate refers to the rate at which a material undergoes plastic deformation per unit time under hot pressing conditions.

[0210] In practical applications, embodiments of the present invention can construct an optimization function with the goal of minimizing repair time, and perform inverse solution under the temperature and pressure constraints that meet the material safety threshold to generate a target maintenance parameter set. This allows for finding the most efficient maintenance solution without damaging the battery (not exceeding the safety threshold), thus achieving an intelligent leap from "knowing how to repair" to "calculating the optimal maintenance solution".

[0211] Minimizing repair time is the objective function of the optimization algorithm, aiming to shorten the repair cycle as much as possible.

[0212] Material safety thresholds refer to the maximum temperature and pressure that a battery can withstand, preventing thermal runaway or breakage of active particles during maintenance.

[0213] Inverse solving is a mathematical calculation process that derives the conditions (temperature, pressure, time) from the result (target repair strain).

[0214] The target maintenance parameter set is a calculated set of specific execution instructions, including target temperature, target pressure, and duration.

[0215] In practical applications, embodiments of the present invention can encapsulate the target maintenance parameter group into an electronic work order and send it to the offline maintenance service terminal corresponding to the cloud diagnostic platform, so as to achieve seamless connection between cloud decision-making and offline execution, transforming abstract calculation parameters into standardized maintenance operation instructions to guide the implementation of professional equipment.

[0216] Electronic work orders, which carry maintenance instructions in their data messages, serve as the carrier of digital maintenance.

[0217] Offline repair service terminals are physical terminals located in factories or repair stations that can receive cloud commands and drive hot-press repair equipment.

[0218] In this embodiment of the invention, by introducing the creep mechanics mechanism of sulfide materials, a fully automated reverse closed loop from microscopic peeling diagnosis to repair instruction generation is realized. It can provide the optimal repair parameter scheme within safety constraints according to the actual damage degree (target repair strain) of each individual battery, thereby achieving precise and non-destructive repair of interface failure without disassembling the battery, and significantly extending the service life of the battery.

[0219] To enable those skilled in the art to better understand the embodiments of the present invention, an example is used below to illustrate the embodiments of the present invention, with reference to... Figure 2 , Figure 2 This is a schematic flowchart of a battery detection method provided in an embodiment of the present invention; Step S1: Vehicle-side operation data collection and preprocessing Specific operations: When the vehicle is in a static or relaxation phase after charging is completed, the on-board BMS controller starts a high-precision acquisition mode to record the process of the battery returning from the excited state to the equilibrium state.

[0220] Required parameters: Electrical parameters: battery cell terminal voltage sequence V(t), charge / discharge circuit current sequence I(t).

[0221] Thermal parameter: Internal temperature of a single battery cell, Tcell.

[0222] State parameters: Current state of charge (SOC), initial battery health state (SOH0).

[0223] Calculation method: The original signal is low-pass filtered to eliminate high-frequency noise. The voltage relaxation curve Vrelax(t) after charging is cut off is extracted: Vrelax(t) = V(t) - Vcutoff, where Vcutoff is the charging cutoff voltage. The decay time constant of this curve is extracted as a baseline feature for subsequent electrochemical model parameter calibration, used to correct the diffusion coefficient of the cloud model.

[0224] Step S2: Cloud model initialization and parameter library call Specific operations: The cloud server receives data packets uploaded from the vehicle, parses the battery's unique identifier, identifies the battery as a sulfide solid-state battery system, and loads the corresponding physical property parameters from the materials database. Because sulfide electrolytes have unique mechanical properties, the parameters need to be adjusted accordingly.

[0225] Required parameters: Geometric parameters: electrode particle radius Rp, electrode thickness Lelec, current collector thickness.

[0226] Electrochemical parameters: solid-phase diffusion coefficient Ds, maximum lithium concentration cmax, initial equilibrium lithium concentration c0.

[0227] Mechanical parameters: a) Sulfide electrolytes (such as LGPS and LPS) typically have a low modulus (approximately 10-30 GPa), which is much lower than that of oxide electrolytes, exhibiting certain "soft" characteristics.

[0228] b) Poisson's ratio ν, partial molar volume Ω.

[0229] Interface parameters: Interfacial bonding strength Tmax and critical separation displacement δc. The interface between sulfide electrolytes and electrodes is typically a solid-solid physical contact and is sensitive to interfacial stress.

[0230] Creep parameters: Creep constant A, stress exponent n, activation energy Q. Sulfide materials can exhibit significant creep behavior at relatively low temperatures, which is the key physical basis for the repair function implemented in this invention.

[0231] Step S3: Microscopic stress reconstruction based on multiphysics coupling Specific operations: A cloud-based digital twin engine constructs a three-dimensional finite element model of the solid-state battery, and the sequential coupling method is used to solve the physical fields. The core logic of this step is to transform the electrochemical process into a mechanical process, ultimately deriving the interfacial contact pressure Pc.

[0232] Detailed derivation and calculation formulas: 1. Calculation of lithium-ion concentration field (diffusion equation): The lithium concentration distribution cs(r,t) inside the electrode particles is solved using Fick's second law, and the formula is:

[0233] The solid-phase diffusion coefficient Ds is affected by temperature T, and the corrected formula is the Arrhenius equation:

[0234] This step allows us to obtain the lithium-ion concentration gradient at any location inside the battery.

[0235] 2. Chemical-mechanical strain transformation (constitutive equation): The calculated lithium concentration distribution is converted into volume expansion strain. This is the source of internal stress, and the formula is: .in The partial molar volume characterizes the volume expansion caused by the insertion of one mole of lithium ions. For silicon-carbon anodes, this expansion effect is significant and can easily cause compression of the sulfide electrolyte layer.

[0236] 3. Stress field solution (elasticity equations): Solve the mechanical equilibrium equations to calculate the internal stress tensor. The equilibrium equations are: Based on the constitutive relation, the total stress Determined by the difference between total strain and chemical expansion strain: .in Here is the stiffness matrix. The total strain is determined by the boundary constraints.

[0237] Because sulfide electrolytes are relatively soft, they are prone to deformation under stress, making the interfacial contact pressure distribution extremely sensitive to volume expansion.

[0238] 4. Rigorous derivation of the interfacial contact pressure Pc (key steps): To quantify the interface state, it is necessary to use the stress tensor The pressure component perpendicular to the interface is extracted. The derivation process is as follows: a) Define the normal vector: Let the unit normal vector of the electrode-electrolyte interface be n (pointing to the electrolyte side); b) Calculate the traction force vector: According to the Cauchy stress principle, the traction force vector acting on the interface is... The dot product of the stress tensor and the normal vector: Interface normal stress The projection (dot product) of the traction force vector onto the normal direction: This value represents the stress state in the vertical direction of the interface (positive value represents tensile stress, negative value represents compressive stress).

[0239] c) Define contact pressure: According to the conventions of contact mechanics (pressure is positive), the interface contact pressure is... Defined as the negative of the normal stress:

[0240] Output result: Obtain the interface contact pressure distribution cloud map .when When >0, the interface is under good pressure and contact; when When the value is ≤0, the interface separates and is identified as a potential crack region.

[0241] Step S4: Quantitative Diagnosis of Interface Faults Specific operations: Based on the contact pressure calculated in step S3 The cohesive force model is used to quantitatively assess interface damage.

[0242] Detailed calculation formula: a) Interface separation determination and displacement calculation: If a certain area is under contact pressure If the value is ≤0, it is determined that separation has occurred in the region, and the separation displacement δ of the interface is obtained through iterative calculation.

[0243] b) Calculation of damage variables: The damage variable D is calculated based on a bilinear cohesive model to describe the process of the interface from intact to fractured:

[0244] in: The maximum separation displacement in the loading history; This is the damage initiation displacement (the critical point at which the interface begins to soften). This is the displacement of complete failure (complete interface breakage).

[0245] Damage variable range: D=0 indicates intact, D=1 indicates complete fracture.

[0246] c) Calculation of interface contact failure rate η: To quantify the overall failure rate, the interface contact failure rate η is defined as follows:

[0247] in The total area of ​​the interface. This is the damage threshold (e.g., set to 0.5). This formula calculates the percentage of the area whose damage exceeds the threshold.

[0248] Fault classification: If η>10%, it is judged as a moderate fault (Level L2); If η > 30%, it is judged as a serious fault (Level L3).

[0249] Step S5: Reverse solution of maintenance parameters (core algorithm) Specific operations: If the fault level is determined to be repairable (such as L1 or L2), the cloud-based optimization algorithm is activated to calculate the thermo-mechanical parameters required for repair based on the material's creep characteristics.

[0250] Detailed calculation formula: a) Define the repair objectives: Based on the average width of the crack (Calculated from the δ data in step S4), the required creep closure strain is calculated.

[0251] in The feature length scale.

[0252] b) Creep rate calculation (Norton power-law equation): Calculating the repair rate using the creep properties of solid electrolytes

[0253] in: n is a material constant; Creep activation energy; R is the gas constant (8.314 J / (mol·K)); T is the absolute temperature (K); The equivalent stress applied.

[0254] c) Optimize the solution model: Build to fix time An optimization problem with minimization as the objective.

[0255] Objective function:

[0256] Constraints: Temperature constraint Tmin≤T≤Tmaxsafe (to prevent thermal aging); Pressure constraint min≤ ≤ crush (to prevent structural collapse).

[0257] The cloud uses an iterative algorithm to solve for the optimal temperature Tset and pressure. Set and calculate the corresponding time tset.

[0258] Output result: Generate precise maintenance parameter sets: (Tset, `set,tset` is used to guide offline maintenance equipment to perform hot-press repair.

[0259] Step S6: Electronic work order generation and issuance Specific operations: The cloud encapsulates the diagnostic results and maintenance parameters into an encrypted electronic work order.

[0260] Data content: Fault information: Faulty cell number, interface failure rate η.

[0261] Maintenance instructions: Target temperature Tset, target pressure Fset (by...) (set conversion), duration tset.

[0262] Work orders are sent to the user's APP and the 4S store's maintenance management system via an encrypted channel.

[0263] Step S7: Offline Execution and Closed-Loop Verification Specific operations: The offline repair terminal executes the repair procedure and provides feedback data.

[0264] Detailed steps: Repair execution: The equipment reads the work order parameters and performs heating and pressurization actions.

[0265] Real-time monitoring: The device uploads actual temperature (Treal) and pressure (Freal) data in real time. If the deviation exceeds ±5%, a correction command is sent to the cloud.

[0266] Results Verification: After the repair was completed, the vehicle was powered on again. The system collected data again and uploaded it to the cloud (go to step S1).

[0267] Closed-loop calculation: The cloud-based model recalculates the damage variable Dnew and the failure rate ηnew. If ηnew < ηthreshold, the repair is considered successful, and the battery health record is updated.

[0268] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of the present invention are not limited to the described order of actions, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.

[0269] In addition, this application also provides an electronic device 30, please refer to... Figure 3 It includes a processor 310 and a memory 320, wherein the memory 310 is used to store computer programs; the processor 320 is used to execute the programs stored in the memory 310 to implement the battery detection method described in any embodiment of this application.

[0270] like Figure 4 As shown, in another embodiment of the present invention, a computer-readable storage medium 401 is also provided, which stores instructions that, when executed on a computer, cause the computer to perform the battery detection method described in the above embodiment.

[0271] In this application, "multiple" refers to two or more.

[0272] In this application, unless otherwise expressly defined, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0273] The terms “first,” “second,” “third,” “fourth,” etc., in this application (if present) are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0274] In this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, in this application, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0275] Unless otherwise specified, all steps in this application may be performed sequentially or randomly. For example, if the method includes steps A and B, it means that the method may include steps A and B performed sequentially, or it may include steps B and A performed sequentially. For example, if the method may also include step C, it means that step C may be added to the method in any order. For example, the method may include steps A, B, and C, or it may include steps A, C, and B, or it may include steps C, A, and B, etc.

[0276] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A battery testing method, characterized in that, include: Obtain raw operating data for the battery; Based on the original operating data, the peeling status information of the battery interface is determined, where the interface is the physical contact surface formed between the highly active material particles of the battery and the sulfide solid electrolyte layer. The detection results for the battery are generated based on the stripping status information.

2. The method according to claim 1, characterized in that, The method is applied to a cloud-based diagnostic ranking platform, which is configured with vehicles, and the vehicles are configured as follows: The electrical, thermal, and state parameters of the vehicle are obtained when it is in a stationary state or in the relaxation phase after charging is completed; the electrical parameters include the individual cell terminal voltage sequence and charging cut-off voltage of the battery. The voltage relaxation curve after the vehicle charging is completed is generated by using the preprocessed single-cell terminal voltage sequence and the charging cutoff voltage, as well as the decay time constant of the voltage relaxation curve. The voltage relaxation curve, the decay time constant, the electrical parameters, the thermal parameters, and the state parameters are determined as the original operating data; The raw operating data is sent to the cloud-based diagnostic platform; the step of obtaining the raw operating data for the battery includes: Receive raw operating data sent by the vehicle.

3. The method according to claim 2, characterized in that, The method is applied to a cloud-based diagnostic platform with a sulfide material parameter library. The step of determining the peeling status information of the battery interface based on the original operating data includes: Assign a unique identifier to the battery of the vehicle; When the unique identifier is used to determine that the battery type is a sulfide solid-state battery system, geometric parameters, electrochemical parameters, mechanical parameters, interface parameters and creep parameters that match the battery are extracted from the sulfide material parameter library; The electrochemical parameters are corrected by the decay time constant, and the lithium-ion concentration distribution is calculated based on the decay time constant, the corrected electrochemical parameters, the thermal parameters, and the state parameters; the lithium-ion concentration distribution is used to characterize the spatial distribution of lithium ions inside the electrode active particles at the target charge and discharge time. When the lithium ion concentration distribution is determined to meet the preset conditions by the voltage relaxation curve, the interfacial contact pressure distribution used to characterize the degree of adhesion of the interface is calculated by the lithium ion concentration distribution, the electrochemical parameters, and the mechanical parameters. When the interface contact pressure distribution is less than or equal to zero, the interface is determined to be in a microscopic peeling state, and the interface separation displacement result, which reflects the degree of physical separation of the interface, is determined. Based on the interface separation displacement results and the interface parameters, the peeling state information is determined.

4. The method according to claim 3, characterized in that, The electrochemical parameters include the solid-phase diffusion coefficient, and the state parameters include current state of charge information. The step of calculating the lithium-ion concentration distribution using the decay time constant, the electrochemical parameters, the thermal parameters, and the state parameters includes: The decay time constant is determined as a reference feature; Substitute the solid-phase diffusion coefficient into the modified Arrhenius equation formula constructed based on the reference feature, use the thermal parameters as temperature compensation, and combine the reference feature to perform gain correction on the solid-phase diffusion coefficient to obtain the corrected diffusion coefficient. The modified diffusion coefficient and the current state of charge information are used as initial boundary conditions and substituted into the second law diffusion equation to solve for the lithium ion concentration distribution.

5. The method according to claim 3, characterized in that, The electrochemical parameters include the initial equilibrium lithium concentration, and the mechanical parameters include the partial molar volume and stiffness matrix. The step of calculating the interfacial contact pressure distribution, characterizing the degree of adhesion of the interface, based on the lithium ion concentration distribution, the electrochemical parameters, and the mechanical parameters, when the lithium ion concentration distribution is determined to meet preset conditions by the voltage relaxation curve, includes: Generate a simulated voltage based on the lithium ion concentration distribution state, and calculate the residual between the simulated voltage and the voltage relaxation curve; When the residual is less than a preset threshold, the partial molar volume is called, and combined with the lithium ion concentration distribution inside the electrode and the initial equilibrium lithium concentration, the micro-strain data generated by chemical expansion is calculated according to the constitutive equation. Using the stiffness matrix and the micro-strain data, a multi-physics coupling solution is performed based on the mechanical equilibrium equation and the total stress relationship to reconstruct the three-dimensional stress tensor inside the battery. Determine the unit normal vector of the interface, perform a dot product operation on the unit normal vector to generate a traction force vector, and perform a second dot product projection on the unit normal vector and the traction force vector to generate a normal stress perpendicular to the interface. The normal stress is converted into the interface contact pressure distribution.

6. The method according to claim 3, characterized in that, The interface parameters include damage initiation displacement and complete failure displacement parameters. The step of determining the peeling state information based on the interface separation displacement result and the interface parameters includes: The damage initiation displacement parameter, the complete failure displacement parameter, and the interface separation displacement result are substituted into the bilinear cohesive model to calculate the damage variable describing the degree of degradation of the micro-interface properties. The damage variable is determined as the peeling state information.

7. The method according to claim 3, characterized in that, The geometric parameters include the total interface area, and the step of generating the detection result for the battery using the peeling state information includes: The percentage of the interface failure area is calculated using the total area of ​​the interface and the damage variable. The contact failure rate of the interface is calculated based on the proportion of the failed area, and the fault level is determined based on the contact failure rate. The failure rate of the interface is used to determine the fault level and generate the detection result.

8. The method according to claim 3, characterized in that, The geometric parameters include the electrolyte layer thickness, and the creep parameters include the creep constant, stress exponent, and activation energy coefficient. The method further includes: The average crack width is determined based on the interface separation displacement results, and the target repair strain is calculated in conjunction with the electrolyte layer thickness. The target repair strain is used to characterize the degree of plastic deformation that the sulfide electrolyte material of the battery must undergo in order to close the crack. Substituting the creep constant, the stress exponent, the activation energy coefficient, and the normal stress into the Norton power law equation, the viscous flow rate of the sulfide electrolyte material under hot-pressing conditions is calculated. An optimization function is constructed with the goal of minimizing repair time. The function is then solved in reverse under the constraints of temperature and pressure that meet the material safety threshold to generate a target maintenance parameter set. The target maintenance parameter set includes at least the target temperature, target pressure, and duration. The target temperature, target pressure, and duration are encapsulated in an electronic work order and sent to the offline maintenance service terminal corresponding to the cloud-based diagnostic platform.

9. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; The memory is used to store computer programs; When the processor executes a program stored in the memory, it implements the method as described in any one of claims 1-8.

10. A computer-readable storage medium having instructions stored thereon that, when executed by one or more processors, cause the processors to perform the method as described in any one of claims 1-8.

Images