A method and device for estimating the state of health of a lithium-ion battery and an electronic device

Abstract

This invention relates to the field of battery management technology and discloses a method, apparatus, and electronic device for estimating the health state of a lithium-ion battery. The method includes: acquiring real-time terminal voltage, current, and temperature signals of the battery; acquiring a priori estimates of a state vector provided by a nonlinear state estimation algorithm, wherein the state vector includes at least a first parameter characterizing capacity decay and a second parameter characterizing power decay; generating a model-predicted terminal voltage based on the priori estimates, real-time current, temperature, and an electrochemical battery model retaining the solid-phase diffusion and interfacial transport characteristics of the electrodes; comparing the predicted voltage with the measured voltage to obtain a voltage residual, and performing a correlational posterior update on the first and second parameters by combining the residual and the Jacobian matrix corresponding to the model; using the updated state vector as the priori estimate for the next cycle for iterative use, and outputting a multidimensional health state result accordingly.

Description

Technical Field

[0001] This invention relates to the field of battery management technology, and specifically to a method, apparatus, and electronic device for estimating the health status of a lithium-ion battery. Background Technology

[0002] With the rapid development of new energy technologies, lithium-ion batteries have been widely used in electric vehicles, portable electronic products, and other fields due to their advantages such as high energy density and long service life. To ensure the safety, reliability, and efficient operation of battery systems, the battery management system (BMS) must achieve real-time monitoring and precise control of the battery's state. Among these parameters, the battery's state of health (SOH) is one of the most important monitoring parameters in the BMS, directly affecting the battery's capacity, cycle life, and safety.

[0003] Currently, mainstream methods for estimating State of Harm (SOH) include those based on capacity decay, internal resistance growth, and machine learning. However, these methods have some significant drawbacks: capacity-based methods require full charge-discharge cycles to obtain accurate results, internal resistance-based methods have low accuracy, and machine learning algorithms rely on large amounts of data and have weak generalization ability. Furthermore, while electrochemical models can more accurately reflect the internal state of the battery, their calculations are complex and involve numerous parameters, making them difficult to directly apply to automotive BMS systems. Summary of the Invention

[0004] This invention provides a method, apparatus, and electronic device for estimating the health status of lithium-ion batteries, in order to solve the problem of difficulty in accurately and in real-time estimating the aging status of batteries under vehicle conditions.

[0005] In a first aspect, the present invention provides a method for estimating the health state of a lithium-ion battery, comprising the following steps: acquiring real-time terminal voltage signals, real-time current signals, and real-time temperature signals of the battery; acquiring prior estimates of a state vector determined by a nonlinear state estimation algorithm, wherein the state vector includes at least a first parameter characterizing battery capacity decay and a second parameter characterizing battery power decay; generating a model-predicted terminal voltage based on the prior estimates of the state vector, the real-time current signal, the real-time temperature signal, and an electrochemical battery model characterizing the internal mass transport and polarization characteristics of the battery; comparing the model-predicted terminal voltage with the real-time terminal voltage signal to obtain a voltage residual, and using the voltage residual and the Jacobian matrix calculated based on the electrochemical battery model to perform a posterior update of the first and second parameters; iterating the updated state vector as the prior estimate for the next cycle, and obtaining a multidimensional health state result based on the updated state vector.

[0006] This invention integrates a physically meaningful electrochemical battery model with a nonlinear state estimation algorithm. It enables simultaneous, online, and high-precision joint estimation of battery capacity and power degradation, relying solely on voltage, current, and temperature signals collected by a conventional BMS, without requiring full charge-discharge cycles or additional hardware support. The employed electrochemical battery model retains key kinetic characteristics of electrode solid-phase diffusion and electrolyte interface transport, making the model's predicted voltage exhibit clear physical sensitivity to aging parameters. Combined with a Jacobian matrix-guided correlation update mechanism, it effectively improves the consistency, robustness, and adaptability of multidimensional health state estimation, thereby outputting multidimensional health state results reflecting the aging characteristics of the battery's entire life cycle in real time under any dynamic operating conditions.

[0007] In one alternative implementation, the nonlinear state estimation algorithm is an extended Kalman filter; the electrochemical cell model is a single-particle model, and the single-particle model is configured with a kinetic sub-model characterizing the charge transfer and solid-phase diffusion properties at the electrolyte-electrode interface.

[0008] This implementation combines an extended Kalman filter with a single-event model, which significantly reduces the computational burden while retaining the physical characterization capability of key battery aging mechanisms, and achieves coordinated, high-precision online estimation of capacity and power decay parameters.

[0009] In one alternative implementation, the first parameter is the maximum reversible lithium-ion concentration at the positive electrode, and the second parameter is the ohmic internal resistance.

[0010] This implementation uses the maximum reversible lithium-ion concentration and ohmic internal resistance of the positive electrode to characterize capacity decay and power decay, respectively. This facilitates direct mapping to multidimensional health indicators, effectively reflecting the aging mechanism of active lithium loss and interface impedance growth throughout the battery's life cycle, and improving the interpretability and engineering practicality of SOH estimation.

[0011] In one optional implementation, generating the model-predicted terminal voltage based on the prior estimate of the state vector, the real-time current signal, the real-time temperature signal, and the electrochemical battery model characterizing the internal mass transport and polarization properties of the battery includes: reading the updated lithium-ion concentration on the surface of the cathode particles from the electrochemical battery model; calculating the current lithium-ion concentration on the surface of the cathode particles based on the real-time current signal and the updated lithium-ion concentration on the surface of the cathode particles from the previous cycle, using the solid-phase diffusion kinetics equation contained in the electrochemical battery model; calculating the cathode open-circuit potential based on the current cathode open-circuit potential and the first parameter in the state vector; calculating the cathode overpotential based on the real-time current signal, the current cathode open-circuit potential, and the real-time temperature signal, combined with the reaction kinetics relationship contained in the electrochemical battery model; determining the ohmic voltage drop term according to the second parameter in the state vector and the real-time current signal; and adding the cathode open-circuit potential, the cathode overpotential, and the ohmic voltage drop term to generate the model-predicted terminal voltage.

[0012] This implementation method constructs a terminal voltage prediction model step by step, sequentially integrating solid-phase lithium-ion diffusion dynamics, electrode / electrolyte interface electrochemical reaction kinetics, and ohmic impedance effects. This fully preserves the contribution of key internal physical processes to the terminal voltage. Real-time updates of the lithium-ion concentration on the cathode particle surface ensure the accuracy of open-circuit potential and overpotential calculations, while the ohmic voltage drop term is directly related to aging-related internal resistance changes. This structured modeling approach not only ensures a high degree of matching between the model-predicted voltage and the measured signal but also allows the voltage residual to effectively decouple and sensitively reflect the independent aging mechanisms of capacity and power decay. This provides a solid physical observation foundation for subsequent extended Kalman filters to achieve high-precision, robust, multi-dimensional joint estimation of health status.

[0013] In one optional implementation, the posterior state update of the first and second parameters using the voltage residual and the Jacobian matrix calculated based on the electrochemical cell model includes: determining the voltage sensitivity of the electrochemical cell model to the first parameter and using the voltage sensitivity of the first parameter as the first row element of the Jacobian matrix; determining the voltage sensitivity of the electrochemical cell model to the second parameter and using the voltage sensitivity of the second parameter as the second row element of the Jacobian matrix; calculating the Kalman gain matrix based on the Jacobian matrix, the prior error covariance matrix obtained by iterative update of the nonlinear state estimation algorithm, and the pre-calibrated measurement noise covariance; obtaining the state correction amount by multiplying the Kalman gain matrix and the voltage residual, and superimposing the state correction amount onto the prior estimate of the state vector to complete the update of the prior estimate.

[0014] This implementation calculates the Kalman gain based on the Jacobian matrix and covariance information, enabling the first and second parameters to share the aging information carried by the voltage residual during the update process. This achieves dynamic coupling and mutual correction between the two during the estimation process, effectively improving the consistency, convergence speed, and anti-interference capability of multidimensional health state estimation.

[0015] In one optional implementation, obtaining the multidimensional health status result based on the updated state vector includes: calculating the capacity health based on the first parameter in the state vector and the initial calibration value of the first parameter; calculating the power health based on the second parameter in the state vector, the initial calibration value of the second parameter and the lifetime termination threshold of the second parameter; and using the capacity health and power health as the multidimensional health status result.

[0016] This implementation method integrates capacity health and power health with multiple parameters, reflecting a multi-dimensional SOH decision-making and health index construction method based on physical mechanisms, which effectively supports the battery management system in accurate assessment and proactive safety control under complex operating conditions.

[0017] Secondly, the present invention also provides a device for estimating the health status of a lithium-ion battery. The device includes a first acquisition module, a second acquisition module, a terminal voltage prediction module, a parameter update module, and a health status prediction module. The first acquisition module is used to acquire the real-time terminal voltage signal, real-time current signal, and real-time temperature signal of the battery. The second acquisition module is used to acquire the prior estimate of the state vector determined by a nonlinear state estimation algorithm. The state vector includes at least a first parameter representing the battery capacity decay and a second parameter representing the battery power decay. The terminal voltage prediction module is used to generate a model-predicted terminal voltage based on the prior estimate of the state vector, the real-time current signal, the real-time temperature signal, and an electrochemical battery model representing the internal mass transport and polarization characteristics of the battery. The parameter update module is used to compare the model-predicted terminal voltage with the real-time terminal voltage signal to obtain a voltage residual, and to perform a posterior update of the first and second parameters using the voltage residual and the Jacobian matrix calculated based on the electrochemical battery model. The health status determination module is used to iterate the updated state vector as the prior estimate for the next cycle and obtain a multidimensional health status result based on the updated state vector.

[0018] Thirdly, the present invention also provides an electronic device, including a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to perform the lithium-ion battery health state estimation method described in the first aspect or any corresponding embodiment.

[0019] Fourthly, the present invention also provides a computer-readable storage medium storing computer instructions for causing a computer to execute the method for estimating the state of health of a lithium-ion battery according to the first aspect or any corresponding embodiment thereof.

[0020] Fifthly, the present invention also provides a computer program product, including computer instructions for causing a computer to execute the method for estimating the state of health of a lithium-ion battery according to the first aspect or any corresponding embodiment thereof. Attached Figure Description

[0021] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of an application scenario according to an embodiment of the present invention; Figure 2 This is a flowchart of the first method for estimating the health status of a lithium-ion battery according to an embodiment of the present invention; Figure 3 This is a second flowchart of the lithium-ion battery health status estimation method according to an embodiment of the present invention; Figure 4 This is a schematic diagram of a low-pass filter model according to an embodiment of the present invention; Figure 5 This is an internal logic diagram of a low-pass filter according to an embodiment of the present invention; Figure 6 This is a flowchart of the extended Kalman filter according to an embodiment of the present invention; Figure 7 This is a schematic diagram of a single-particle logic model according to an embodiment of the present invention; Figure 8 This is a schematic diagram illustrating an example of a lithium-ion battery health status estimation method according to an embodiment of the present invention; Figure 9 This is a schematic diagram verifying the convergence process of different state estimations according to an embodiment of the present invention; Figure 10 This is a schematic diagram illustrating the full lifecycle SOH tracking accuracy according to an embodiment of the present invention; Figure 11 This is a structural block diagram of a lithium-ion battery health status estimation device according to an embodiment of the present invention; Figure 12 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0025] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0026] As an optional application scenario of this invention, such as Figure 1 As shown, the lithium-ion battery health status estimation system may include at least one terminal device and at least one server. Figure 1 The system is illustrated in the example, which includes a computer 101, a mobile terminal 102, and a server 103, and the terminal devices such as the computer 101 and the mobile terminal 102 are connected to the server 103 through a network 110.

[0027] Specifically, the terminal device can be a smartphone, tablet, laptop, PDA, desktop computer, game console, smart TV, smart wearable device, in-vehicle terminal, VR (Virtual Reality) device, AR (Augmented Reality) device, etc. Server 103 can be a standalone physical server, a server cluster, a distributed system, or a cloud server providing cloud services. Network 110 can be a wired or wireless network, examples of which include, but are not limited to, the Internet, corporate intranet, local area network, wide area network, mobile communication network, and combinations thereof.

[0028] According to an embodiment of the present invention, an embodiment of a method for estimating the health status of a lithium-ion battery is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0029] This embodiment provides a method for estimating the health status of a lithium-ion battery, which can be used in the aforementioned mobile terminals, such as mobile phones and tablets. Figure 2 This is a flowchart of the first method for estimating the state of health of a lithium-ion battery according to an embodiment of the present invention, as follows: Figure 2 As shown, the process includes the following steps: Step S201: Acquire the real-time terminal voltage signal, real-time current signal, and real-time temperature signal of the battery.

[0030] Specifically, the real-time terminal voltage signal, real-time current signal, and real-time temperature signal can be synchronously acquired by the battery management system (BMS) at a fixed frequency and obtained through first-order RC low-pass filtering and short-time averaging preprocessing.

[0031] Step S202: Obtain the prior estimate of the state vector determined by the nonlinear state estimation algorithm. The state vector includes at least a first parameter characterizing battery capacity decay and a second parameter characterizing battery power decay.

[0032] The prior estimate of the state vector is directly inherited from the posterior estimate of the previous cycle at the initial moment of the current cycle.

[0033] Capacity decay and power decay are characterized by the maximum reversible lithium-ion concentration and ohmic internal resistance of the positive electrode, respectively. This allows for direct mapping to multidimensional health indicators, effectively reflecting the aging mechanism of active lithium loss and interface impedance growth throughout the battery's life cycle, and improving the interpretability and engineering practicality of SOH estimation.

[0034] Step S203: Based on the prior estimate of the state vector, the real-time current signal, the real-time temperature signal, and the electrochemical cell model characterizing the internal mass transport and polarization characteristics of the battery, generate the model-predicted terminal voltage.

[0035] Specifically, the electrochemical cell model is a single-particle model, and the single-particle model is equipped with a kinetic sub-model that characterizes the charge transfer and solid-phase diffusion properties at the electrolyte-electrode interface.

[0036] This kinetic sub-model can explicitly describe the charge transfer process and concentration-dependent exchange current behavior of lithium ions at the electrode / electrolyte interface, while coupling the dynamic changes in surface concentration caused by solid-phase diffusion, thus accurately reflecting the nonlinear overpotential dominated by interface kinetics in the polarization voltage. This allows the model to not only reflect the thermodynamic equilibrium potential, but also capture the voltage deviation caused by interface impedance and lithium insertion / extraction kinetics limitations under high-rate or transient conditions, significantly improving the identifiability and estimation robustness of aging-sensitive parameters.

[0037] Step S204: Compare the model-predicted terminal voltage with the real-time terminal voltage signal to obtain the voltage residual, and use the voltage residual and the Jacobian matrix corresponding to the electrochemical cell model to perform a correlation posterior update on the first and second parameters.

[0038] The Jacobian matrix can be the observation Jacobian matrix, which consists of the partial derivatives of the model predicted voltage with respect to the first and second parameters. The Kalman gain is calculated based on the Jacobian matrix, the prior covariance, and the measurement noise covariance, so that the update amounts of the two parameters can be mutually corrected through the cross terms in the gain matrix.

[0039] Step S205: Iterate the updated state vector as the prior estimate for the next cycle, and output the multidimensional health status result based on the updated state vector.

[0040] Specifically, the multidimensional health status results include capacity health (SOH_C) and power health (SOH_R), which are calculated by normalizing the updated first and second parameters, respectively.

[0041] The lithium-ion battery health state estimation method provided in this embodiment integrates a simplified electrochemical model with physical meaning and a nonlinear state estimation algorithm. This enables simultaneous, online, and high-precision joint estimation of battery capacity and power degradation, relying solely on voltage, current, and temperature signals collected by a conventional BMS, without requiring full charge-discharge cycles or additional hardware support. The electrochemical model retains key kinetic characteristics of electrode solid-phase diffusion and electrolyte interface transport, making the model's predicted voltage exhibit clear physical sensitivity to aging parameters. Combined with a Jacobian matrix-guided correlation update mechanism, it effectively improves the consistency, robustness, and adaptability of multidimensional health state estimation. Thus, it outputs multidimensional health state results reflecting the aging characteristics of the battery's entire life cycle in real time under any dynamic operating conditions, significantly outperforming traditional single-index SOH methods and black-box data-driven models.

[0042] This embodiment provides a method for estimating the health status of a lithium-ion battery, which can be used in the aforementioned mobile terminals, such as mobile phones and tablets. Figure 3This is a second flowchart of the lithium-ion battery health status estimation method according to an embodiment of the present invention, as follows: Figure 3 As shown, the process includes the following steps: Step S301: Acquire the real-time terminal voltage signal, real-time current signal, and real-time temperature signal of the battery.

[0043] Specifically, at the beginning of each estimation cycle, the battery's raw sensor signals, including terminal voltage Vmeas, load current Imeas, and battery surface or internal temperature Tmeas, are synchronously acquired by the battery management system (BMS) at a fixed sampling frequency (e.g., 1 kHz). To suppress the adverse effects of high-frequency electromagnetic interference and sensor noise on model predictions, the raw signals are subjected to first-order RC low-pass filtering. Subsequently, the moving average of the filtered signals is calculated within a short time window (e.g., 10–50 ms) to obtain the denoised observation vector used for updating the electrochemical model. This preprocessing mechanism effectively avoids state estimation divergence caused by transient glitches or measurement jitter, while preserving key electrical characteristics under dynamic operating conditions, providing high signal-to-noise ratio input data for subsequent voltage prediction and parameter updates.

[0044] For example, a first-order RC low-pass filter can be performed using a low-pass filter model, such as... Figure 4 As shown, in the low-pass filter model, Vmeas, Imeas, and Tmeas are the input values, representing the measured voltage, current, and temperature, respectively; Vk, Ik, and Tk are the output values ​​after filtering, representing the filtered voltage, current, and temperature, respectively. The low-pass filter model aims to avoid random error fluctuations at a single sampling point by smoothing continuously sampled data to output stable values; the sampling time Ts is 1 second.

[0045] Figure 5 The internal logic diagram of the low-pass filter illustrates the specific algorithm implementation of the above model, which includes three parallel filtering branches, corresponding to voltage, current, and temperature signals, respectively. In each branch, the difference between the current measured value and the output value of the previous cycle is first obtained, and this difference is then passed through a gain stage (such as...). Figure 5 The output value (with a coefficient of 20) is amplified and then added to the output value of the previous cycle (obtained through a unit delay module 1 / z) to calculate the filtered output value of the current cycle. This structure utilizes information from historical data to achieve high-frequency noise suppression of the signal.

[0046] Step S302: Obtain the prior estimate of the state vector determined by the nonlinear state estimation algorithm. The state vector includes at least a first parameter characterizing battery capacity decay and a second parameter characterizing battery power decay.

[0047] The nonlinear state estimation algorithm is an extended Kalman filter algorithm. The first parameter is the maximum reversible lithium-ion concentration at the positive electrode, and the second parameter is the ohmic internal resistance.

[0048] like Figure 6 As shown, the Extended Kalman Filter (EKF) algorithm linearizes the nonlinear equations by performing a first-order Taylor expansion at the estimated state points. The state-space model of the nonlinear discrete system is as follows: x k+1 =f(x k ,u k )+w k y k = g(x k ,u k )+v k Where, x k U is the state variable at time k. k It is the system's control input, y k This is the output value at time k. f(x) k ,u k ) is the state transition function, g(x) k ,u k ) is the measurement function. k and v k These are process noise and measurement noise, respectively, both with a mean of 0, and they are uncorrelated. k ~ N(0,Q k ),v k ~ N(0,R k ), where Q k Let R be the covariance matrix of the process noise. k This is the covariance matrix for measuring noise.

[0049] When the system runs for the first time, the initialization of the state estimator is completed first: defining the state vector - The first parameter The maximum reversible lithium-ion concentration at the positive electrode is used to characterize the capacity decay caused by the loss of active lithium; the second parameter R0 is the ohmic internal resistance, used to reflect the power decay caused by contact degradation, electrolyte drying, etc.

[0050] Will The initial values ​​are set to the calibration values ​​of the battery in its brand-new state (obtained by GITT or full charge / discharge tests). The initial value of R0 is determined by HPPC pulse testing, and the initial state vector X0 is set accordingly. At the same time, based on the assessment of the uncertainty of the initial state, a reasonable initial error covariance matrix P0 is configured, and the process noise covariance matrix Q and the measurement noise covariance matrix R are set according to the dynamic characteristics of the system and the sensor noise level.

[0051] In each subsequent estimation cycle, the prior estimate of the state vector is generated from the posterior state of the previous cycle through a prediction step. Since the estimated aging parameters change slowly, a random walk model is used in this embodiment. Therefore, the prior estimate is directly taken as the posterior estimate of the previous cycle. This prior estimate serves as the starting point for the current cycle's electrochemical model voltage prediction and Kalman update, ensuring that the state estimation algorithm continuously tracks the evolution of the battery's health state throughout its entire life cycle.

[0052] Step S303: Based on the prior estimate of the state vector, the real-time current signal, the real-time temperature signal, and the electrochemical cell model characterizing the internal mass transport and polarization characteristics of the battery, generate the model-predicted terminal voltage.

[0053] The electrochemical cell model is a single-particle model.

[0054] For example, such as Figure 7 As shown, in the Single Particle Model (SPM), Cs_surf is the lithium concentration on the solid particle surface, Cs_avg is the average lithium concentration of the solid particle, j is the reaction current density, F is the Faraday constant, R_s is the particle radius, D_s is the solid diffusion coefficient, and A is the solid conductive area. The calculation logic in the SPM is as follows: using the input reaction current density j, solid conductive area A, and Faraday constant F, the average lithium concentration Cs_avg is obtained by accumulating the data through an integrator after gain calculation; combining parameters such as particle radius Rs and solid diffusion coefficient Ds, the average concentration is corrected and subjected to a second integral calculation to solve for the solid surface lithium concentration Cs_surf, which reflects the polarization characteristics; finally, the normalized ratio of the surface concentration to the maximum lithium concentration is calculated and input into the one-dimensional lookup table module 1-DT(u) to obtain the thermodynamic potential OCP under the current state. Based on this, before generating the model prediction terminal voltage using the prior estimates based on the state vector, the real-time current signal, the real-time temperature signal, and the electrochemical battery model that characterizes the internal mass transport and polarization properties of the battery, the parameters of the electrochemical battery model are initialized and calibrated to ensure the accuracy of the model calculation.

[0055] In the battery's brand-new state, the three core parameters required for the electrochemical model were calibrated using testing methods such as HPPC (hybrid pulse power characteristics) and GITT (galvanostatic titration), and a parameter database was established. The three core parameters include material parameters, configuration parameters, and thermodynamic parameters.

[0056] Among them, the material parameters reflect the intrinsic physicochemical properties of the battery active materials, including the particle radius Rs of the positive and negative electrodes, the theoretical maximum lithium-ion concentration Cs,max of the positive electrode (characterizing the material's ultimate lithium storage capacity), the solid-phase diffusion coefficient Ds, the electrochemical reaction rate constant k, the activation energy Ea, the electrolyte lithium-ion concentration Ce (considered as a constant), and the Faraday constant F.

[0057] Configuration parameters describe the geometric and interfacial structural features of the battery, including the electrochemical active area Apos of the positive electrode and the electrochemical active area Aneg of the negative electrode.

[0058] Thermodynamic parameters are used to characterize the relationship between open circuit potential (OCP) and the lithium intercalation state of the electrode and temperature.

[0059] Specifically, the equilibrium voltage of the battery at different lithium insertion depths was obtained through GITT testing, and the corresponding degree of lithium insertion θ = Cs,surf / Cs,max was calculated. Based on the OCP–θ data, a third-order polynomial fitting was used to obtain the analytical expression for OCP.

[0060] Among them, A, B, C, and D are polynomial coefficients obtained by fitting experimental data, which are used to accurately describe the nonlinear relationship between the electrode open circuit potential (OCP) and the degree of lithium intercalation θ.

[0061] Furthermore, the temperature dependence of OCP was obtained by conducting static or GITT tests at multiple temperature points, and a linear temperature compensation model was established.

[0062] Among them, T ref For reference temperature, k temp This is the temperature compensation coefficient.

[0063] The above calibration results together form the basis of a simplified electrochemical model with physical consistency, providing accurate and reliable mechanistic support for subsequent online health status estimation.

[0064] In one optional implementation, generating the model-predicted terminal voltage based on the prior estimate of the state vector, the real-time current signal, the real-time temperature signal, and the electrochemical cell model characterizing the internal mass transport and polarization properties of the battery includes the following steps S3031 to S3036.

[0065] Step S3031: Read the lithium ion concentration on the surface of the cathode particles after the previous cycle update from the electrochemical cell model.

[0066] Specifically, at the start of the current estimation cycle, the lithium-ion concentration on the surface of the cathode particles, obtained from the previous cycle and updated via EKF posterior, is read from the state memory unit. This concentration value is the output of the solid-phase diffusion process at the previous moment. It serves as the initial condition for solving the dynamic evolution of surface concentration in the current cycle, ensuring the continuity of the concentration state in time and its physical consistency.

[0067] Step S3032: Based on the real-time current signal and the updated lithium-ion concentration on the surface of the cathode particles in the previous cycle, calculate the lithium-ion concentration on the surface of the cathode particles in the current cycle using the solid-phase diffusion kinetics equation contained in the electrochemical battery model.

[0068] Specifically, the real-time current signal I of the current period k Converted to local current density And substitute it into the solid-state diffusion dynamics equation of the single-particle model:

[0069] in The average particle concentration (which can be obtained from historical data or simplified assumptions) is given. The ordinary differential equation is numerically integrated using a polynomial approximation method, incorporating the first parameter from the prior state. Compared to the previous cycle The lithium-ion concentration on the cathode surface in the current cycle is calculated. .

[0070] Step S3033: Calculate the open-circuit potential of the positive electrode based on the lithium ion concentration on the surface of the positive electrode particles in the current cycle and the first parameter in the state vector.

[0071] Specifically, according to the definition of lithium intercalation degree Substitute θpos into the third-order polynomial thermodynamic relation determined in step 101: And combine the real-time temperature Tk with the application of the temperature compensation formula Finally, the temperature-corrected positive open-circuit potential is obtained.

[0072] Step S3034: Based on the real-time current signal, the lithium ion concentration on the surface of the cathode particles in the current cycle, and the real-time temperature signal, and combined with the reaction kinetic relationship built into the electrochemical battery model, calculate the cathode overpotential.

[0073] First, according to , Given the electrolyte concentration Ce, calculate the exchange current density:

[0074] The reaction rate constant k(Tk) is expressed through the Arrhenius relation. Temperature compensation is performed, where R is the molar gas constant in the Arrhenius relation. Subsequently, the known local current densities j and i0 are substituted into the Butler-Wolmer equation, and the nonlinear equation for the overpotential ηpos is numerically solved using the Newton-Raphson method: In the Butler-Wolmer equation, R is the molar gas constant and F is the Faraday constant, thus obtaining the overpotential caused by positive electrode electrochemical polarization. 。

[0075] Step S3035: Determine the ohmic voltage drop term based on the second parameter in the state vector and the real-time current signal.

[0076] Specifically, the second parameter (i.e., the ohmic resistance) is extracted from the prior estimate of the current state vector. and with real-time current signal Multiplying them together, we get the Ohm drop term: This voltage drop term characterizes the purely resistive voltage loss caused by factors such as current collector contact resistance and electrolyte resistance.

[0077] Step S3036: Add the positive open circuit potential, positive overpotential, and ohmic voltage drop term to generate the model prediction terminal voltage.

[0078] Specifically, the difference in open-circuit potential between the positive and negative terminals... The difference between positive and negative overpotentials And ohmic voltage drop term By superimposing the data, we obtain the complete model-predicted terminal voltage:

[0079] This voltage comprehensively reflects the battery's thermodynamic equilibrium potential, kinetic polarization, and ohmic losses, providing a high-fidelity observation and prediction benchmark for subsequent EKF updates.

[0080] Step S304: Compare the model-predicted terminal voltage with the real-time terminal voltage signal to obtain the voltage residual.

[0081] Specifically, the model-predicted terminal voltage Vpred generated in step S3036 is compared with the preprocessed real-time terminal voltage measurement value VK in step S301, and the voltage residual is calculated. .

[0082] The residual reflects the prediction bias of the current electrochemical model under a given state prior and is the core observational information driving the state update. In this embodiment, if the absolute value of the residual is less than a preset threshold (e.g., ...), If the value is less than 0.01, the estimation is considered to have converged, and subsequent updates will continue; otherwise, data resampling or anomaly diagnosis mechanisms may be triggered.

[0083] Step S305: Use the voltage residual and the Jacobian matrix calculated based on the electrochemical cell model to perform a post-hoc update on the first and second parameters.

[0084] In one optional implementation, the state posterior update of the first parameter and the second parameter using the voltage residual and the Jacobian matrix calculated based on the electrochemical cell model includes the following steps S3051 to S3053.

[0085] Step S3051: Determine the voltage sensitivity of the electrochemical cell model to the first parameter, and use the voltage sensitivity of the first parameter as the first row element of the Jacobian matrix; determine the voltage sensitivity of the electrochemical cell model to the second parameter, and use the voltage sensitivity of the second parameter as the second row element of the Jacobian matrix.

[0086] Constructing the Jacobian matrix .

[0087] The first row of elements Calculated using the perturbation method: in the prior state The first parameter Apply small increments Keeping other inputs unchanged, repeat step S303 to obtain the predicted voltage after disturbance. And the partial derivatives are approximately calculated as follows: .

[0088] Second row elements Ohm voltage drop term Direct analysis reveals a strong linear coupling relationship between internal resistance and terminal voltage.

[0089] The Jacobian matrix quantitatively describes the sensitivity of two aging parameters to the output voltage and is a key bridge for achieving parameter correlation updates.

[0090] Step S3052: Based on the Jacobian matrix, the prior error covariance matrix obtained by iterative updating of the nonlinear state estimation algorithm, and the pre-calibrated measurement noise covariance, calculate the Kalman gain matrix.

[0091] Specifically, the prior error covariance matrix passed in step S303 is used. The Jacobian matrix obtained in step S3051 and the pre-defined measurement noise covariance Calculate the Kalman gain matrix according to the standard formula for the extended Kalman filter:

[0092] The gain matrix dynamic trade-off model predicts reliability versus measurement reliability: when voltage observation uncertainty ( When the first parameter is smaller, the gain is larger, indicating greater trust in the residual information; conversely, it relies more on prior estimates. Its structure implicitly includes cross-update weights between the first and second parameters.

[0093] Step S3053: Obtain the state correction by multiplying the Kalman gain matrix and the voltage residual, and then superimpose the state correction onto the prior estimate of the state vector to complete the update of the prior estimate.

[0094] Specifically, the Kalman gain Kk is multiplied by the voltage residual ek to obtain the state correction, and the state vector is updated synchronously. .

[0095] because Contains cross terms of Together, they determine the interaction between the update amounts of the first parameter (capacity decay) and the second parameter (power decay). For example, under high current conditions, R The high sensitivity of 0 will enhance its ability to The updated indirect constraints enable collaborative identification and consistency correction of multi-dimensional aging states at the data level.

[0096] Step S306: Iterate the updated state vector as the prior estimate for the next cycle, and output the multidimensional health status result based on the updated state vector.

[0097] In one optional implementation, outputting multidimensional health status results based on the updated state vector includes: calculating capacity health based on a first parameter in the state vector and its initial calibration value; calculating power health based on a second parameter in the state vector, its initial calibration value, and its lifespan termination threshold; outputting capacity health and power health as multidimensional health status results; and generating a comprehensive health status index for safety decision-making in the battery management system based on the multidimensional health status results.

[0098] Specifically, the posterior state vector The updated state vector is stored in the state buffer as the prior initial value for the next estimation cycle, enabling closed-loop iteration. Simultaneously, multidimensional health indicators are calculated based on this updated state vector.

[0099] Capacity health:

[0100] Power health status:

[0101] Adopt a conservative fusion strategy (such as taking) It generates comprehensive health status indicators for BMS safety decisions, enabling real-time, multi-dimensional, and interpretable battery health assessment.

[0102] Figure 8 This is a schematic diagram illustrating an example of a lithium-ion battery health status estimation method according to an embodiment of the present invention, as shown below. Figure 8 As shown, the overall process of this method consists of five stages: system initialization and parameter calibration, real-time parameter acquisition and preprocessing, extended Kalman filter prior prediction, single-event model calculation of the voltage, and extended Kalman filter update. A convergence judgment node is set between stages four and five, and the end of the process includes result output and state iteration steps, thus forming a closed-loop real-time estimation loop.

[0103] The specific details of each stage and key step are as follows: 1. Phase 1: System Initialization and Parameter Calibration. First, step 11 is executed to calibrate the electrochemical model parameters (such as solid-phase diffusion coefficient, particle radius, maximum lithium concentration, etc.); then step 12 is executed to complete the initialization of the state estimator (extended Kalman filter), setting the initial state vector and error covariance matrix to provide a benchmark for subsequent real-time estimation.

[0104] 2. Phase Two: Real-time Parameter Acquisition and Preprocessing. After initialization, the real-time main loop begins. This phase includes signal acquisition in step 21 and signal preprocessing (filtered averaging) in step 22, aiming to acquire and clean the raw data to eliminate noise interference and provide reliable input for subsequent calculations.

[0105] 3. Stage Three: Extended Kalman Filter Prior Prediction. After data preprocessing, the prediction stage begins. The state vector prior prediction from step 31 and the error covariance prior prediction from step 32 are executed sequentially. Based on the state estimate from the previous time step and the system dynamic model, a preliminary prediction of the current state is made.

[0106] 4. Stage Four: Solving the Voltage Using a Single-Particle Model. Starting with the calculation of the local current density in step 41, the process proceeds to solving the solid-state diffusion equation in step 42, calculating the thermodynamic potential OCP in step 43, calculating the kinetic overpotential η in step 44, and finally obtaining the model-predicted voltage Vpred in step 45.

[0107] Convergence Judgment (Step 51): A judgment node is set after Stage 4. If the judgment result is "No" (not converged), the process rolls back to Step 22 of Stage 2 for reprocessing; if the result is "Yes" (converged), the process continues to the next step.

[0108] 5. Stage Five: Update of the Extended Kalman Filter. This stage is entered after the convergence condition is met. First, the Jacobian matrix H is calculated using step 52. k This is used to linearize the observation equation, and then in step 53, the Kalman gain is calculated, and the state variable Xk and error covariance are updated accordingly to complete the correction of the prior prediction value.

[0109] Results Output and Iteration (Steps 6 and 7): At the end of the process, Step 6 is executed sequentially to output the health status. Finally, Step 7 is used to iterate the status, passing the updated status from the current moment to the next moment, and returning to Phase 2 to start a new cycle.

[0110] Figure 9 This is a schematic diagram verifying the convergence process of different state estimations. Figure 9 Based on the true initial error, the algorithm uses the incorrect estimates of R0 and c_s_max_set as the starting input. The algorithm converges the state vector quickly within 50 laps and keeps close to the true value during the 200-lap cycle.

[0111] Figure 10 A schematic diagram of SOH tracking accuracy throughout the entire lifecycle, as shown below. Figure 10 As shown, based on 3000 cycles of LFP battery data, the model is used to predict the state of health. Over the entire lifespan, the predicted SOHc and SOHr values ​​have errors of 0.29% and 0.17% respectively compared to the actual values. This demonstrates the model's good robustness to SOH.

[0112] The method for estimating the health status of lithium-ion batteries provided in this embodiment has the following beneficial effects: 1. By establishing a highly simplified but still physically meaningful electrochemical model and combining it with extended Kalman filter (EKF) technology, the goal of online real-time estimation of lithium battery state of health (SOH) can be achieved without additional hardware, which significantly reduces computational complexity and hardware cost, and is particularly suitable for real-time monitoring of vehicle BMS systems; 2. By adopting a single-particle model (SPM) and combining it with electrolyte kinetics improvements, the internal aging mechanism of the battery can be accurately reflected by precisely tracking the lithium-ion concentration on the positive and negative electrode surfaces, thus overcoming the low accuracy problem of traditional capacity and internal resistance methods. 3. By simultaneously estimating the maximum lithium-ion concentration and internal resistance of the cathode using an extended Kalman filter, the multi-dimensional variation characteristics of SOH can be effectively characterized, providing more comprehensive battery health information and overcoming the weakness of machine learning algorithms in generalization. 4. SOH estimation can be performed without full charge-discharge cycles, breaking through the limitations of traditional capacity methods. It can monitor SOH in real time under actual dynamic battery conditions, improving battery performance and safety. 5. The physical relationships between key parameters during battery aging were established, including active lithium ion loss and internal resistance growth, which can more accurately reflect the battery's life cycle characteristics and improve the reliability and accuracy of the estimation results.

[0113] This embodiment also provides a lithium-ion battery health status estimation device, which is used to implement the above embodiments and preferred embodiments, and will not be repeated as already described. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0114] This embodiment provides a device for estimating the health status of a lithium-ion battery, such as... Figure 11 As shown, it includes a first acquisition module 1101, a second acquisition module 1102, a terminal voltage prediction module 1103, a parameter update module 1104, and a health status determination module 1105.

[0115] The first acquisition module 1101 is used to acquire the real-time terminal voltage signal, real-time current signal and real-time temperature signal of the battery.

[0116] The second acquisition module 1102 is used to acquire the prior estimate of the state vector determined by the nonlinear state estimation algorithm. The state vector includes at least a first parameter characterizing the battery capacity decay and a second parameter characterizing the battery power decay.

[0117] The terminal voltage prediction module 1103 is used to generate a model-predicted terminal voltage based on prior estimates of the state vector, real-time current signals, real-time temperature signals, and an electrochemical cell model characterizing the internal mass transport and polarization properties of the battery.

[0118] The parameter update module 1104 is used to compare the model-predicted terminal voltage with the real-time terminal voltage signal to obtain the voltage residual, and to perform state posterior update on the first and second parameters using the voltage residual and the Jacobian matrix calculated based on the electrochemical cell model.

[0119] The health status determination module 1105 is used to iterate the updated state vector as the prior estimate for the next period and obtain the multidimensional health status result based on the updated state vector.

[0120] In some optional implementations, the nonlinear state estimation algorithm is an extended Kalman filter; the electrochemical cell model is a single-particle model, and the single-particle model is configured with a kinetic sub-model characterizing the charge transfer and solid-phase diffusion properties at the electrolyte-electrode interface.

[0121] In some alternative implementations, the first parameter is the maximum reversible lithium-ion concentration at the positive electrode, and the second parameter is the ohmic internal resistance.

[0122] In some optional implementations, the terminal voltage prediction module 1103 is specifically used for: reading the updated lithium-ion concentration on the surface of the cathode particles from the electrochemical cell model; calculating the lithium-ion concentration on the surface of the cathode particles in the current cycle based on the real-time current signal and the updated lithium-ion concentration on the surface of the cathode particles in the previous cycle, using the solid-phase diffusion kinetics equation contained in the electrochemical cell model; calculating the cathode open-circuit potential based on the current cathode open-circuit lithium-ion concentration and the first parameter in the state vector; calculating the cathode overpotential based on the real-time current signal, the current cathode open-circuit lithium-ion concentration, and the real-time temperature signal, combined with the reaction kinetics relationship contained in the electrochemical cell model; determining the ohmic voltage drop term according to the second parameter in the state vector and the real-time current signal; and adding the cathode open-circuit potential, the cathode overpotential, and the ohmic voltage drop term to generate the model-predicted terminal voltage.

[0123] In some optional implementations, the parameter update module 1104 is specifically used to: determine the voltage sensitivity of the electrochemical cell model to a first parameter, and use the voltage sensitivity of the first parameter as the first row element of the Jacobian matrix; determine the voltage sensitivity of the electrochemical cell model to a second parameter, and use the voltage sensitivity of the second parameter as the second row element of the Jacobian matrix; calculate the Kalman gain matrix based on the Jacobian matrix, the prior error covariance matrix obtained by iterative updating of the nonlinear state estimation algorithm, and the pre-calibrated measurement noise covariance; obtain the state correction amount by multiplying the Kalman gain matrix and the voltage residual, and superimpose the state correction amount onto the prior estimate of the state vector to complete the update of the prior estimate.

[0124] In some optional implementations, the health status determination module 1105 is specifically used to: calculate the capacity health based on the first parameter in the state vector and the initial calibration value of the first parameter; calculate the power health based on the second parameter in the state vector, the initial calibration value of the second parameter and the lifetime termination threshold of the second parameter; and output the capacity health and power health as a multidimensional health status result.

[0125] The lithium-ion battery health status estimation device provided in this embodiment of the invention can execute the lithium-ion battery health status estimation method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the above modules and units are the same as in the corresponding embodiments described above, and will not be repeated here.

[0126] Figure 12 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0127] The following is a detailed reference. Figure 12The diagram illustrates a structural schematic suitable for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 1201, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 1202 or a program loaded from memory 1208 into random access memory (RAM) 1203. The RAM 1203 also stores various programs and data required for the operation of the electronic device. The processor 1201, ROM 1202, and RAM 1203 are interconnected via a bus 1204. An input / output (I / O) interface 1205 is also connected to the bus 1204.

[0128] Typically, the following devices can be connected to I / O interface 1205: input devices 1206 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 1207 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 1208 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1209. Communication device 1209 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 12 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0129] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 1209, or installed from a memory 1208, or installed from a ROM 1202. When the computer program is executed by the processor 1201, it performs the functions defined in the lithium-ion battery health state estimation method of the embodiments of the present invention.

[0130] Figure 12 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0131] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the method for estimating the state of health of a lithium-ion battery shown in the above embodiments is implemented.

[0132] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0133] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and all such modifications and variations fall within the scope defined by the appended claims.

Claims

1. A method for estimating the state of health of a lithium-ion battery, characterized in that, The method includes: Acquire the real-time terminal voltage signal, real-time current signal, and real-time temperature signal of the battery; Obtain a priori estimate of the state vector determined by a nonlinear state estimation algorithm, wherein the state vector includes at least a first parameter characterizing battery capacity decay and a second parameter characterizing battery power decay; Based on the prior estimate of the state vector, the real-time current signal, the real-time temperature signal, and the electrochemical battery model characterizing the internal mass transport and polarization properties of the battery, the model predicts the terminal voltage. The predicted terminal voltage of the model is compared with the real-time terminal voltage signal to obtain the voltage residual. The first parameter and the second parameter are then updated posteriorly using the voltage residual and the Jacobian matrix calculated based on the electrochemical cell model. The updated state vector is used as the prior estimate for the next cycle for iteration, and a multidimensional health status result is obtained based on the updated state vector.

2. The method according to claim 1, characterized in that, The nonlinear state estimation algorithm is an extended Kalman filter algorithm; the electrochemical cell model is a single-particle model, and the single-particle model is equipped with a kinetic sub-model characterizing the charge transfer and solid-phase diffusion characteristics at the electrolyte-electrode interface.

3. The method according to claim 1 or 2, characterized in that, The first parameter is the maximum reversible lithium-ion concentration at the positive electrode, and the second parameter is the ohmic internal resistance.

4. The method according to claim 1, characterized in that, The model-predicted terminal voltage is generated based on the prior estimate of the state vector, the real-time current signal, the real-time temperature signal, and the electrochemical cell model characterizing the internal mass transport and polarization properties of the battery, including: Read the lithium ion concentration on the surface of the cathode particles after the previous cycle update from the electrochemical cell model; Based on the real-time current signal and the lithium-ion concentration on the surface of the cathode particles updated in the previous cycle, the lithium-ion concentration on the surface of the cathode particles in the current cycle is calculated using the solid-phase diffusion kinetics equation contained in the electrochemical cell model. The open-circuit potential of the positive electrode is calculated based on the lithium ion concentration on the surface of the positive electrode particles in the current cycle and the first parameter in the state vector. Based on the real-time current signal, the lithium ion concentration on the surface of the cathode particles in the current cycle, and the real-time temperature signal, combined with the reaction kinetics relationship contained in the electrochemical battery model, the cathode overpotential is calculated. The ohmic voltage drop term is determined based on the second parameter in the state vector and the real-time current signal; The positive open-circuit potential, the positive overpotential, and the ohmic voltage drop term are added together to generate the model's predicted terminal voltage.

5. The method according to claim 1, characterized in that, The step of updating the first and second parameters a posteriori using the voltage residual and the Jacobian matrix calculated based on the electrochemical cell model includes: Determine the voltage sensitivity of the electrochemical cell model to the first parameter, and use the voltage sensitivity of the first parameter as the first row element of the Jacobian matrix; Determine the voltage sensitivity of the electrochemical cell model to the second parameter, and use the voltage sensitivity of the second parameter as the second row element of the Jacobian matrix; Based on the Jacobian matrix, the prior error covariance matrix obtained by iterative updating of the nonlinear state estimation algorithm, and the pre-calibrated measurement noise covariance, the Kalman gain matrix is ​​calculated. The state correction is obtained by multiplying the Kalman gain matrix and the voltage residual. The state correction is then superimposed on the prior estimate of the state vector to update the prior estimate.

6. The method according to claim 1, characterized in that, The process of obtaining the multidimensional health status result based on the updated state vector includes: The capacity health is calculated based on the first parameter in the state vector and the initial calibration value of the first parameter. The power health is calculated based on the second parameter in the state vector, the initial calibration value of the second parameter, and the lifetime termination threshold of the second parameter. The capacity health and the power health are used as the results of the multidimensional health status.

7. A device for estimating the health status of a lithium-ion battery, characterized in that, The device includes: The first acquisition module is used to acquire the real-time terminal voltage signal, real-time current signal and real-time temperature signal of the battery; The second acquisition module is used to acquire the prior estimate of the state vector determined by the nonlinear state estimation algorithm, wherein the state vector includes at least a first parameter characterizing battery capacity decay and a second parameter characterizing battery power decay. The terminal voltage prediction module is used to generate a model-predicted terminal voltage based on the prior estimate of the state vector, the real-time current signal, the real-time temperature signal, and an electrochemical cell model characterizing the internal mass transport and polarization properties of the battery. The parameter update module is used to compare the model-predicted terminal voltage with the real-time terminal voltage signal to obtain the voltage residual, and use the voltage residual and the Jacobian matrix calculated based on the electrochemical cell model to perform state posterior update on the first parameter and the second parameter. The health status determination module is used to iterate the updated state vector as the prior estimate for the next period and obtain a multidimensional health status result based on the updated state vector.

8. An electronic device, characterized in that, include: A memory and a processor are communicatively connected, the memory storing computer instructions, and the processor executing the computer instructions to perform the method for estimating the state of health of a lithium-ion battery as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute the method for estimating the state of health of a lithium-ion battery according to any one of claims 1 to 6.

10. A computer program product, characterized in that, Includes computer instructions for causing a computer to execute the method for estimating the state of health of a lithium-ion battery as described in any one of claims 1 to 6.

Images