Method and system for establishing battery status warning model in electrochemical energy storage

Abstract

The present application provides a method and system for establishing a battery status early warning model in electrochemical energy storage, which determines the equivalent circuit model corresponding to the target battery according to the target battery, and determines the state equation and observation equation according to the equivalent circuit model corresponding to the target battery to determine the Kalman filter state model corresponding to the target battery. According to the electrochemical principle of the target battery, the diffusion equation of electrons in the electrode and the transmission equation in the electrolyte in the target battery are determined to determine the P2D electrochemical model corresponding to the target battery; based on the Kalman filter state model and the P2D electrochemical model, a dual-algorithm fusion model of the battery status corresponding to the target battery is obtained. This improves the accuracy of battery status estimation, can better adapt to the changes and complexity of batteries in different working environments, and thus respond to rapid changes in batteries in a timely manner.

Description

Technical Field

[0001] The present application relates to energy storage safety monitoring technology, and in particular to a method and system for establishing a battery status early warning model in electrochemical energy storage. Background Art

[0002] With the continuous increase in the installed capacity of electrochemical energy storage power stations, their safety and control issues have received great attention. Energy storage power stations generally adopt a regular maintenance strategy. However, the maintenance cycle and plan in the regular maintenance strategy are fixed, making it difficult to detect safety hazards in a timely manner.

[0003] At present, the state variables of electrochemical energy storage systems (such as battery voltage, temperature, current, etc.) are estimated through Kalman filtering technology, which overcomes the sensitivity to noise and uncertainty in traditional methods, thereby improving monitoring accuracy and early warning reliability.

[0004] However, traditional Kalman filtering methods usually only consider the state of the battery system for estimation, lack of analysis of the electrochemical characteristics of the battery, and it is difficult to comprehensively evaluate the performance of the electrochemical energy storage system, thus affecting the effectiveness of battery energy storage early warning. Summary of the Invention

[0005] The present application provides a method and system for establishing a battery status early warning model in electrochemical energy storage, which is used to solve the technical problems mentioned in the background technology.

[0006] In a first aspect, the present application provides a method for establishing a battery status early warning model in electrochemical energy storage, comprising:

[0007] Determining an equivalent circuit model corresponding to the target battery according to the target battery, and determining a state equation and an observation equation according to the equivalent circuit model corresponding to the target battery;

[0008] Determining a Kalman filter state model corresponding to the target battery according to the state equation and the observation equation;

[0009] Determining the diffusion equation of electrons in the electrode and the transport equation of electrons in the electrolyte in the target battery according to the electrochemical principle of the target battery;

[0010] Determining a P2D electrochemical model corresponding to the target battery according to the electron diffusion equation in the electrode and the electron transport equation in the electrolyte;

[0011] According to the Kalman filter state model and the P2D electrochemical model, a dual-algorithm fusion model of the battery state corresponding to the target battery is obtained, the dual-algorithm fusion model of the battery state is used to predict the electrochemical energy storage state prediction value corresponding to the target moment according to the charge and discharge state parameters and electrochemical state parameters corresponding to the target battery at the first moment, the electrochemical energy storage state prediction value includes: an energy storage state prediction value and an electrochemical state prediction value, the energy storage state prediction value corresponding to the target moment is used to compare with the energy storage state actual value in the electrochemical energy storage state actual value corresponding to the target moment, and the electrochemical state prediction value is used to compare with the electrochemical state actual value in the electrochemical energy storage state actual value, so as to provide an early warning of the electrochemical energy storage state of the target battery according to any comparison result;

[0012] The charge and discharge state parameters include: charge and discharge voltage, charge and discharge current, charge and discharge resistance, remaining capacity SOC, and battery remaining energy state SOE; the electrochemical state parameters include: the diffusion coefficient of the electron in the electrode, the diffusion coefficient of the electron in the electrolyte, the radial coordinates of the electron particles, the first-order partial derivative and the second-order partial derivative of the electron concentration in the electrode with respect to the radial coordinate, the first-order partial derivative and the second-order partial derivative of the electron concentration in the electrolyte with respect to the radial coordinate, and the reaction current density.

[0013] In a second aspect, the present application provides a system for establishing a battery status early warning model in electrochemical energy storage, comprising:

[0014] a first model determination module, configured to determine an equivalent circuit model corresponding to a target battery according to the target battery, and to determine a state equation and an observation equation according to the equivalent circuit model corresponding to the target battery; and further configured to determine a Kalman filter state model corresponding to the target battery according to the state equation and the observation equation;

[0015] a second model determination module, configured to determine, based on the electrochemical principle of the target battery, a diffusion equation for electrons in the electrode and a transport equation for electrons in the electrolyte in the target battery; and further configured to determine, based on the diffusion equation for electrons in the electrode and the transport equation for electrons in the electrolyte, a P2D electrochemical model corresponding to the target battery;

[0016] a model fusion module for obtaining a dual-algorithm fusion model of the battery state corresponding to the target battery based on the Kalman filter state model and the P2D electrochemical model, wherein the dual-algorithm fusion model of the battery state is used to predict the electrochemical energy storage state prediction value corresponding to the target moment based on the charge and discharge state parameters and electrochemical state parameters corresponding to the target battery at the first moment, wherein the electrochemical energy storage state prediction value includes: an energy storage state prediction value and an electrochemical state prediction value, wherein the energy storage state prediction value corresponding to the target moment is used to compare with the energy storage state actual value in the electrochemical energy storage state actual value corresponding to the target moment, and the electrochemical state prediction value is used to compare with the electrochemical state actual value in the electrochemical energy storage state actual value, so as to provide an early warning of the electrochemical energy storage state of the target battery based on any comparison result;

[0017] The charge and discharge state parameters include: charge and discharge voltage, charge and discharge current, charge and discharge resistance, remaining capacity SOC, and battery remaining energy state SOE; the electrochemical state parameters include: the diffusion coefficient of the electron in the electrode, the diffusion coefficient of the electron in the electrolyte, the radial coordinates of the electron particles, the first-order partial derivative and the second-order partial derivative of the electron concentration in the electrode with respect to the radial coordinate, the first-order partial derivative and the second-order partial derivative of the electron concentration in the electrolyte with respect to the radial coordinate, and the reaction current density.

[0018] In a third aspect, the present application provides an electronic device, comprising: a processor and a memory;

[0019] Memory stores computer-executable instructions;

[0020] The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method as described in any one of the first aspects.

[0021] In a fourth aspect, an embodiment of the present application provides a readable storage medium, including a program or instruction. When the program or instruction runs on a computer, the method described in any one of the above-mentioned first aspects is executed.

[0022] In a fifth aspect, an embodiment of the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the method as described in any one of the first aspects.

[0023] The present application provides a method and system for establishing a battery status early warning model in electrochemical energy storage, which determines an equivalent circuit model corresponding to the target battery according to the target battery, and determines a state equation and an observation equation according to the equivalent circuit model corresponding to the target battery, determines a Kalman filter state model corresponding to the target battery according to the state equation and the observation equation, and determines the diffusion equation of electrons in the electrode and the transport equation in the electrolyte in the target battery according to the electrochemical principle of the target battery; determines a P2D electrochemical model corresponding to the target battery according to the diffusion equation of electrons in the electrode and the transport equation in the electrolyte; obtains a battery status dual-algorithm fusion model corresponding to the target battery according to the Kalman filter state model and the P2D electrochemical model, and the battery status dual-algorithm fusion model is used to predict the electrochemical energy storage status prediction value corresponding to the target moment according to the charge and discharge state parameters and electrochemical state parameters corresponding to the target battery at the first moment, and the electrochemical energy storage status prediction value corresponding to the target moment is used to compare with the actual value of the electrochemical energy storage state corresponding to the target moment, so as to issue an early warning of the electrochemical energy storage state of the target battery according to the comparison result. The real-time performance of the Kalman filter and the high precision of the P2D model are combined to complement each other, thereby improving the accuracy of battery state estimation and better adapting to the changes and complexity of the battery in different working environments, thereby responding to rapid changes in the battery in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0025] Figure 1 A flow chart of a method for establishing a battery status warning model in electrochemical energy storage provided in one embodiment of the present application;

[0026] Figure 2 A schematic diagram of the structure of a system for establishing a battery status early warning model in an electrochemical energy storage system provided in an embodiment of the present application;

[0027] Figure 3 A schematic diagram of the structure of an electronic device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0028] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application are clearly and completely described below. Obviously, the described embodiments are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts also fall within the scope of protection of this application.

[0029] Existing technologies use Kalman filtering to assess the battery status of electrochemical energy storage systems, relying solely on battery state variables (such as voltage, temperature, and current). While this overcomes the sensitivity to noise and uncertainty inherent in traditional methods, thereby improving monitoring accuracy and early warning reliability, traditional Kalman filtering methods typically only consider the battery system's status for estimation, lacking analysis of the battery's internal electrochemical characteristics. This makes it difficult to comprehensively assess the performance of the electrochemical energy storage system, thus impacting the effectiveness of battery energy storage early warnings.

[0030] Therefore, in order to solve the technical problems existing in the prior art, the present application proposes a method and system for establishing a battery status early warning model in electrochemical energy storage, by constructing a Kalman filter state model for the equivalent circuit model of the battery, and constructing a P2D electrochemical model based on the electrochemical principle of the battery, and obtaining a battery status dual-algorithm fusion model based on the Kalman filter state model and the P2D electrochemical model. The battery status dual-algorithm fusion model can simultaneously analyze the state variables and internal electrochemical characteristics of the battery, thereby providing more scientific decision support for the battery management system (BMS) and improving the accuracy of the early warning.

[0031] Figure 1 This is a flow chart of a method for establishing a battery status warning model in an electrochemical energy storage system according to an embodiment of the present application. Figure 1 As shown, the execution subject of the method may be, for example, a BMS, wherein the method includes:

[0032] S101. Determine an equivalent circuit model corresponding to a target battery according to the target battery, and determine a state equation and an observation equation according to the equivalent circuit model corresponding to the target battery.

[0033] In this step, when constructing the equivalent circuit model, the Thevenin equivalent circuit model is selected. This model consists of an ideal voltage source, an internal resistor R0, and an RC parallel circuit consisting of a resistor R1 and a capacitor C1. The ideal voltage source is used to simulate the open-circuit voltage of the battery. Its voltage value is related to the battery's SOC. The voltage-SOC curve data and the parameters of the internal resistor R0, resistor R1, and capacitor C1 can be obtained by consulting the battery specification. The battery is tested in the frequency range of 10mHz-100kHz using electrochemical impedance spectroscopy (EIS) testing equipment (such as the Solartron 1260 frequency response analyzer with a 1287 electrochemical interface). The parameter values ​​are determined by fitting the test data to obtain the equivalent circuit model.

[0034] And through the equivalent circuit model, the state equation and observation equation are determined.

[0035] S102: Determine a Kalman filter state model corresponding to the target battery according to the state equation and the observation equation.

[0036] In this step, a Kalman filter is applied to the equivalent circuit model based on the state equation and observation equation to obtain a Kalman filter state model. Kalman filtering is a recursive optimal estimation algorithm that estimates the system state through two steps: prediction and update. Prediction is achieved through a state prediction equation, which is obtained through the state equation and is used to predict the charge and discharge state parameters at the target time based on the current charge and discharge state parameters.

[0037] The update is achieved through the state update equation, which is obtained through the observation equation and is used to update the charge and discharge state parameters at the predicted target moment, so that the charge and discharge state parameters at the target moment finally obtained are more accurate.

[0038] Therefore, the Kalman filter state model includes the state prediction equation and the state update equation.

[0039] S103. Determine the diffusion equation of electrons in the electrode and the transport equation of electrons in the electrolyte in the target battery according to the electrochemical principle of the target battery.

[0040] In this step, taking lithium-ion batteries as an example, the corresponding electrons are lithium ions. For electrochemical principles, based on battery materials science, electrochemical kinetics, etc., the lithium ion embedding / de-embedding reaction mechanism in the positive and negative electrode materials, the migration process in the electrolyte, and the electrochemical reaction kinetics principles are sorted out. Lithium iron phosphate (LiFePO4) is selected as the positive electrode material, graphite is selected as the negative electrode material, and the electrolyte is lithium hexafluorophosphate (LiPF6) organic solvent solution. The corresponding P2D (two-dimensional quasi-solid-state) electrochemical model of the battery is constructed. Using COMSOL Multiphysics software, a two-dimensional axisymmetric geometric model is established in the software to simulate the electrode-electrolyte structure of the battery, set the positive electrode, negative electrode, separator and electrolyte regions, and determine the diffusion equation of lithium ions in the electrode.

[0041] and the transport equation in the electrolyte .

[0042] Among them, for the diffusion equation of lithium ions in the electrode:

[0043] Formula 1

[0044] c s is the concentration of lithium ions in the electrode, D s is the diffusion coefficient of lithium ions in the electrode, r is the radial coordinate of the electrode particle, t It's time, The concentration of lithium ions in the electrode changes with time The rate of change, Represents the concentration of lithium ions in the electrode versus the radial coordinate r The first-order partial derivative of Indicates the concentration of lithium ions in the electrode For radial coordinates r The second-order partial derivative of .

[0045] It should be noted that c s 、 D s 、 r 、 t 、 as well as It can be obtained by the method in the prior art. c s 、 D s 、 r 、 t 、 as well as When it is known, when it is substituted into formula 1, we can get , so according to the current moment c s And the target time difference between the target time and the current time, get the target time corresponding c s .

[0046] For the transport equation of lithium ions in the electrolyte:

[0047] Formula 2

[0048] c e is the concentration of lithium ions in the electrolyte, is the gradient operator, D e is the diffusion coefficient of lithium ions in the electrolyte, is the gradient of lithium ion concentration, It is the migration number of lithium ions, which represents the ratio between the current transmission capacity of lithium ions in the electrolyte and the total current transmission capacity. F is the Faraday constant, j is the reaction current density, The lithium ion concentration in the electrolyte changes with time t The rate of change, represents the divergence of the lithium ion diffusion flux.

[0049] The reaction current density is calculated using the electrode reaction kinetics equation, wherein the electrode reaction kinetics equation is specifically calculated as Formula 3:

[0050] Formula 3

[0051] Where, j is the reaction current density, j 0 is the exchange current density, exp is an exponential function, is the charge transfer coefficient of the anode, is the charge transfer coefficient of the cathode, F is the Faraday constant, is the overpotential, S is the gas constant, T is the temperature, where , , represents the transfer coefficient, dimensionless and usually takes 0.5, n It represents the number of electrons transferred per mole of active material by lithium ions. Substituting it into the equation can calculate the change of lithium ion concentration in the electrolyte over time, and finally construct the corresponding lithium ion diffusion equation and lithium ion transport equation.

[0052] It should be noted thatc e 、 、 D e 、 、 、 F as well as It can be obtained by existing technology or is a constant. In addition, j Determined by formula 3, therefore, when c e 、 、 D e 、 、 、 F 、 j as well as When it is known, formula 2 can be used to obtain , so according to the current moment c e And the target time difference between the target time and the current time, get the target time corresponding c e .

[0053] S104. Determine a P2D electrochemical model corresponding to the target battery based on the electron diffusion equation in the electrode and the electron transport equation in the electrolyte.

[0054] In this step, according to the diffusion equation and the transport equation, the diffusion equation and the transport equation are added to the model as partial differential equations, and the corresponding boundary conditions and initial conditions are set. In terms of boundary conditions, electrochemical reaction boundary conditions are set at the electrode-electrolyte interface, and the relationship between lithium ion concentration and current density at the interface is determined according to the actual reaction mechanism of the battery; insulation and concentration fixed boundary conditions are set at the outer boundary of the model, and the initial conditions are set according to the initial state of the battery, and the initial concentration distribution of lithium ions in the electrode and electrolyte is set. The model is solved by the numerical calculation engine of the software, and the spatial area is divided into multiple finite element grids. The equations in each grid are discretized, and the iterative algorithm is used to solve the equation group to obtain the spatiotemporal distribution of lithium ion concentration in the electrode and electrolyte, thereby completing the construction of the battery P2D electrochemical model. The model can accurately describe the physical and chemical processes inside the battery and provide an accurate battery state simulation basis for electrochemical energy storage early warning based on Kalman filtering.

[0055] S105. Obtain a dual-algorithm fusion model of the battery state corresponding to the target battery based on the Kalman filter state model and the P2D electrochemical model. The dual-algorithm fusion model of the battery state is used to predict the electrochemical energy storage state prediction value corresponding to the target moment based on the charge and discharge state parameters and electrochemical state parameters corresponding to the target battery at the first moment. The electrochemical energy storage state prediction value includes: an energy storage state prediction value and an electrochemical state prediction value. The energy storage state prediction value corresponding to the target moment is used to compare with the energy storage state actual value in the electrochemical energy storage state actual value corresponding to the target moment. The electrochemical state prediction value is used to compare with the electrochemical state actual value in the electrochemical energy storage state actual value, so as to provide an early warning of the electrochemical energy storage state of the target battery based on any comparison result.

[0056] The charge and discharge state parameters include: charge and discharge voltage, charge and discharge current, charge and discharge resistance, SOC, SOE; the electrochemical state parameters include: diffusion coefficient of electrons in electrodes, diffusion coefficient of electrons in electrolytes, radial coordinates of electron particles, first-order and second-order partial derivatives of electron concentration in electrodes with respect to radial coordinates, first-order and second-order partial derivatives of electron concentration in electrolyte with respect to radial coordinates, and reaction current density.

[0057] In this step, a battery state dual-algorithm fusion model is constructed based on the collaborative fusion of the Kalman filter state model and the P2D electrochemical model corresponding to the target battery. Among them, the Kalman filter state model and the P2D electrochemical model serve as submodules of the battery state dual-algorithm fusion model. For the Kalman filter state model, its input is the charge and discharge state parameters at the current moment, such as charge and discharge voltage, charge and discharge current, charge and discharge resistance, SOC, SOE. For data, since this application is used to provide early warning of the battery state, the output data can be the SOC and SOE corresponding to the target moment.

[0058] For the P2D electrochemical model, its input is the diffusion coefficient of lithium ions in the electrode and electrolyte corresponding to the current moment, the radial coordinates of the lithium ion particles, the first-order partial derivative and second-order partial derivative of the lithium ion concentration in the electrode with respect to the radial coordinate, the first-order partial derivative and second-order partial derivative of the lithium ion concentration in the electrolyte with respect to the radial coordinate, the reaction current density and other parameters, and the output is the concentration of lithium ions in the electrode and electrolyte corresponding to the target moment.

[0059] Correspondingly, the electrochemical energy storage state prediction value corresponding to the target time includes: the energy storage state prediction value, that is, the SOC and SOE corresponding to the target time, and the electrochemical state prediction value, that is, the concentration of lithium ions in the electrode and the electrolyte.

[0060] Since the predicted value of the electrochemical energy storage state is obtained by prediction and the actual value of the electrochemical energy storage state is obtained by actual measurement, if the error between the predicted value of the electrochemical energy storage state and the actual value of the electrochemical energy storage state is large, it may also indicate that there are potential problems with the battery, such as abnormal chemical reactions inside the battery, accelerated battery aging, measurement equipment failure, etc. Therefore, through the comparison results between the predicted value of the electrochemical energy storage state and the actual value of the electrochemical energy storage state, an early warning of the status of the target battery is issued, allowing relevant personnel to conduct inspections and maintenance in a timely manner to avoid further deterioration of the problem, resulting in safety accidents or affecting the battery's service life.

[0061] Optional, in Figure 1 Based on the embodiment shown, the method further includes:

[0062] S106: Obtain the charge and discharge state parameters and electrochemical state parameters of the target battery at the current moment, as well as the actual value of the electrochemical energy storage state at the target moment.

[0063] In this step, at the current moment, a high-precision voltage sensor and current sensor are used to measure the voltage and current of the target battery in real time. For the measurement of the resistance of the target battery, the AC impedance method can be used. By applying a small-amplitude AC signal to the target battery and measuring the AC response of the target battery, the resistance of the target battery can be calculated.

[0064] For example, for the current SOC and SOE, the ampere-hour integration method combined with the open-circuit voltage method can be used to calculate the SOC. The ampere-hour integration method integrates the battery charge and discharge current to calculate the change in battery charge, thereby obtaining an approximate value of the state of charge. The open-circuit voltage method measures the battery's open-circuit voltage and determines the battery's state of charge based on the battery's open-circuit voltage-state of charge curve. Combining the two methods can improve the accuracy of the state of charge calculation. The calculation of SOE needs to consider the battery's energy storage. It is related not only to the battery's charge, but also to the battery's voltage. The energy storage state SOE can be obtained by calculating the ratio of the battery's remaining energy at the current moment to the battery's rated energy.

[0065] For electrochemical state parameters, the diffusion coefficient of lithium ions in the electrode material was measured using the constant current intermittent titration technique (GITT). D s By controlling the charge and discharge current, recording the change of electrode potential over time, and combining the relevant formula to calculate D s The calculation process of this formula is as follows: ,in I is a constant current, is the constant current pulse duration, n is the number of electrons transferred per mole of active material by lithium ions,F is the Faraday constant, A is the electrode / electrolyte interface area, C 0 is the initial lithium ion concentration.

[0066] For the transport of lithium ions in the electrolyte, based on the concentrated solution theory and the Nernst-Planck equation, an electrochemical workstation was used to measure the reaction current density of the battery under different working conditions. The impedance characteristics of the electrolyte were analyzed through the electrochemical impedance spectroscopy (EIS) experiment, and the diffusion coefficient of lithium ions in the electrolyte was calculated. D e ,in D e The calculation process is specifically as follows: ,in is the standard deviation of the AC impedance.

[0067] The actual value of the electrochemical energy storage state at the target time, that is, the SOC, SOE, and the diffusion coefficient of lithium ions in the electrode material corresponding to the target time D s and the diffusion coefficient of lithium ions in the electrolyte D e When or after the target time is reached, the same method as above can be used to obtain it.

[0068] S107: Based on the current charge and discharge state parameters and electrochemical state parameters, and the battery state dual algorithm fusion model, predict and obtain the electrochemical energy storage state prediction value corresponding to the target time.

[0069] In this step, the charge and discharge state parameters at the current moment are input into the Kalman filter state model in the battery state dual algorithm fusion model to predict the SOC and SOE corresponding to the target moment, and the electrochemical state parameters are input into the P2D electrochemical model in the battery state dual algorithm fusion model to obtain and .

[0070] S108: Compare the predicted value of the electrochemical energy storage state with the actual value of the electrochemical energy storage state, and determine whether to execute a warning action based on the comparison result.

[0071] In this step, according to and The concentrations of lithium ions in the electrode and electrolyte corresponding to the target time are predicted respectively, thereby obtaining the predicted target time corresponding to the target time. c s and c e , and the SOC and SOE corresponding to the target moment obtained by the previous prediction are respectively the same as the actual SOC, SOE, cs 、 c e Compare the corresponding data and determine whether to execute the warning action based on the comparison result.

[0072] Optionally, you can choose the SOC, SOE, c s 、 c e Whether to execute a warning action is determined based on any one or more comparison results corresponding to the four parameters. For example, whether to execute a warning action is determined based on the comparison result corresponding to the SOC.

[0073] Optionally, a specific implementation of S108 is as follows:

[0074] S1081. Obtain the difference between the predicted value of the electrochemical energy storage state and the actual value of the electrochemical energy storage state.

[0075] Specifically, the difference between the energy storage state prediction value and the energy storage state actual value, as well as the difference between the electrochemical state prediction value and the electrochemical state actual value, is obtained, that is, the predicted SOC, SOE, c s 、 c e The actual SOC, SOE, c s 、 c e Compare the corresponding values ​​and obtain the corresponding four difference values.

[0076] S1802: When the difference is greater than or equal to the corresponding warning preset value, execute the warning action.

[0077] Specifically, the warning preset value can be, for example, 5% of the actual value of the electrochemical energy storage state. When the difference is greater than 5% of the actual value of the electrochemical energy storage state, it indicates that the battery state of the target battery is poor, and an alarm message is issued so that the staff can inspect and maintain the battery in time.

[0078] This embodiment determines an equivalent circuit model corresponding to the target battery based on the target battery, determines a state equation and an observation equation based on the equivalent circuit model, determines a Kalman filter state model corresponding to the target battery based on the state equation and observation equation, determines the diffusion equation of electrons in the electrodes and the transport equation of electrons in the electrolyte in the target battery based on the electrochemical principle of the target battery, determines a P2D electrochemical model corresponding to the target battery based on the diffusion equation of electrons in the electrodes and the transport equation of electrons in the electrolyte, and obtains a battery state dual-algorithm fusion model corresponding to the target battery based on the Kalman filter state model and the P2D electrochemical model. The battery state dual-algorithm fusion model is used to predict a predicted electrochemical energy storage state value at a target time based on the charge-discharge state parameters and electrochemical state parameters of the target battery at a first time. The predicted electrochemical energy storage state value at the target time is compared with the actual electrochemical energy storage state value at the target time, and an early warning of the electrochemical energy storage state of the target battery is issued based on the comparison result. This achieves the combination of the real-time performance of the Kalman filter and the high precision of the P2D model, improves the accuracy of battery state estimation, and can better adapt to the changes and complexity of batteries in different operating environments, thereby responding to rapid changes in the battery in a timely manner.

[0079] Optionally, the specific implementation of “determining the state equation and the observation equation according to the equivalent circuit model corresponding to the target battery” in S101 is:

[0080] S201 : Obtain the charge and discharge state parameters, process noise, and terminal voltage of the equivalent circuit model at the current moment.

[0081] In this step, the method of obtaining the charge and discharge state parameters can be referred to above and will not be repeated here.

[0082] For the process noise at the current moment w k , due to the state change error caused by uncertain factors such as the chemical reaction inside the battery and environmental factors, we can establish a probability distribution model of process noise by statistically analyzing historical data to construct a probability distribution model with a mean of 0 and a covariance of Q Gaussian distribution, that is w k ~N(0,Q), which means collecting enough historical error data of the state changes of charging and discharging parameters ,therefore ,in x i For the Sample values, Based on the i -1 sample pair i The predicted value between samples, is the sample mean, thus simulating the current moment k Process noise w k .

[0083] The terminal voltage is measured using a high-precision voltage sensor.

[0084] S202 : Determine the input vector corresponding to the current moment according to the charge and discharge voltage, charge and discharge current, and charge and discharge resistance at the current moment, and determine the state vector corresponding to the current moment according to the SOC and SOE corresponding to the current moment.

[0085] In this step, the current moment k The corresponding voltage is V k , the current is I k , the resistance is R int,k , arrange these three data in vector form and perform transposition operation to get the current moment k The input vector u k =[ V k , I k , R int,k ] T .

[0086] According to the current time SOC k by SOE k , get the state vector x k =[ SOC k , SOE k ] T .

[0087] S203 : Determine a state equation corresponding to the target battery according to the input vector, state vector, and process noise corresponding to the current moment.

[0088] In this step, based on process noise , the input vector and the state vector Construct the state equation corresponding to the energy storage battery ,in It is the target moment The state vector of is the state transition matrix, is the input matrix.

[0089] S204: Determine the observation equation corresponding to the target battery according to the terminal voltage.

[0090] In this step, the terminal voltage corresponding to the current moment k is used as the observation vector of the current moment k , construct the observation equation corresponding to the energy storage battery , where the observation matrix Describes the state vector With the observation vector The relationship between the observation noise The error between the observed value and the true value is caused by measurement error, sensor accuracy and other factors. Similarly, by statistically analyzing historical measurement data, a probability distribution model of observation noise can be established to construct a probability distribution model with a mean of 0 and a covariance of Gaussian distribution, that is , thus simulating Observation noise at time .

[0091] Optionally, after S204, the method further includes:

[0092] S205. Obtain a state prediction equation based on the state equation, and obtain a state update equation based on the observation equation.

[0093] In this step, the state prediction equation is based on the state equation Corresponding calculations are performed to calculate the prior estimate of the state vector at the next moment based on the corresponding state vector and input vector, and at the same time, the covariance matrix of the state prediction is calculated based on the state transfer matrix and the covariance matrix of the process noise. The state prediction equation in the Kalman filter state model is specifically:

[0094] Status prediction: ;

[0095]

[0096] ;

[0097] ;

[0098] Covariance prediction:

[0099] ;

[0100] ;

[0101] in, is based on The state of the moment The predicted value of the state at the moment, yes The state vector at time t, is the time step, i.e. the target moment and the current moment The target time difference between is the rated state capacity of the battery, yes The battery current at the moment, is the rated energy of the battery, yes The battery voltage at the moment, yes Battery power at the moment, is the fitting coefficient between SOE change and current in the state vector, is the fitting coefficient between SOE change and voltage in the state vector, is the fitting coefficient between SOE change and internal resistance in the state vector, yes The input vector at time t, yes The process noise at the moment, is the covariance matrix of the predicted state, yes The covariance matrix corresponding to the state vector at the moment, yes The variance of SOC in the state vector at each moment, yes The variance of SOE in the state vector at the moment, is the transpose of the state transition matrix, yes The covariance matrix of the moment process noise, is the mathematical expectation.

[0102] Among them, the state transfer matrix Describes the state vector from Time has come The evolution relationship of the moment, without considering the influence of input, for SOC and SOE, their changes mainly depend on their own dynamic characteristics. SOC and SOE are related to the corresponding current. In one time step Inside, , , Among them (the charging current >0, discharge current <0, is the rated capacity of the battery, is the rated energy of the battery), the corresponding state transfer matrix can be constructed as .

[0103] Input Matrix Describes the impact of the input vector on the state vector, where SOC is mainly affected by the current. , has no direct relationship with voltage and internal resistance. SOE is affected by the combined effects of current, voltage and internal resistance. It is assumed that the change of SOE has a linear relationship with current, voltage and internal resistance. , 、 and The fitting coefficients are determined by experiments or battery modeling. These coefficients describe the relationship between SOE changes and current, voltage and internal resistance and are obtained by least squares fitting. Specifically, it is assumed that Group experimental data , we can construct the matrix equation: ,in: , , , and through the least squares method, the optimal solution is: ,in , , thereby constructing the corresponding input matrix , and according to the state transfer matrix and Covariance matrix of moment-to-moment process noise , calculate the covariance matrix of the state prediction in , ,

[0104] The state update equation is based on the observation equation ,use The observation vector at the moment and the corresponding Kalman gain are used to estimate the prior state prediction Perform an estimated calculation and get The posterior estimate of the state vector at time At the same time, the covariance matrix of the state update is calculated according to the observation matrix, Kalman gain and the covariance matrix of the state prediction, so the state update equation corresponding to the Kalman filter state model is specifically:

[0105] Status Update: ;

[0106] ;

[0107] Covariance update: ;

[0108] in, It is the target moment The estimated value of the state, yes The Kalman gain at time t, is the observation vector and predicted value The linear fitting parameters between the SOC within, is the observation vector and predicted value Linear fitting parameters between inner SOEs, yes The covariance matrix of the estimated state at each moment, is the identity matrix.

[0109] The observation matrix Describes the influence of the state vector on the observation vector. Since the battery terminal voltage is related to SOC and SOE, it is assumed that there is a linear relationship ,in and is a coefficient related to battery characteristics, which can be obtained by fitting experimental data. Assuming that Group experimental data , construct the matrix equation: ,in , Z , , and the optimal solution is also , thus obtaining the corresponding observation matrix ,in, Kalman gain at time The calculation process is specifically as follows: ,in, is the transpose of the observation matrix, It is the covariance matrix of the observation noise. By continuously repeating the state prediction and state update steps from time 0 to time k, the corresponding SOC and SOE state update values ​​of the battery can be obtained according to the corresponding battery current, voltage and internal resistance, and the Kalman filter state model of the battery can be obtained.

[0110] S206: Obtain a Kalman filter state model according to the state prediction equation and the state update equation.

[0111] In this step, the Kalman filter state model is obtained according to the above-mentioned state prediction equation and state update equation.

[0112] In this embodiment, the construction of an equivalent circuit model is crucial. It provides a basis for monitoring, predicting and controlling the battery state. The equivalent circuit model simplifies the complex electrochemical process of the battery and uses components such as resistors and capacitors to represent the internal resistance, charging and discharging process of the battery. When constructing the model, it is first necessary to consider the key electrical characteristics of the battery, such as the open circuit voltage, internal resistance, and capacity, and use these characteristics to derive the battery's state equation and observation equation. The state equation describes the dynamic changes of the battery state, such as the battery's voltage, temperature, and charge state; and the observation equation is based on the relationship between the measured value and the state, and is usually expressed by information such as voltage and current obtained by the sensor. Based on these equations, the Kalman filter state model is obtained to model the noise and error of the battery system, continuously optimize the battery state estimation, and improve the accuracy and reliability of the prediction. This not only can estimate the current state of the battery in real time, but also can cope with the interference of measurement noise and uncertainty.

[0113] Optionally, based on the above embodiment, in order to improve the accuracy of the battery status dual algorithm fusion model, the parameters in the battery status dual algorithm fusion model may be modified. Specifically, the following modifications may be made:

[0114] S301. Control multiple test batteries from the same batch as the target battery to perform charge and discharge tests, and record the test time corresponding to each charge and discharge test, as well as charge and discharge state parameters and electrochemical state parameters. The electrochemical state parameters also include: open circuit voltage and AC impedance.

[0115] In this step, in order to improve the accuracy of the correction, multiple batteries belonging to the same batch as the target battery are used as test batteries. The multiple test batteries are divided into 10 groups, each containing 20 test batteries. The charge and discharge tests are carried out in a constant temperature and humidity environmental chamber with a temperature of 25°C and a humidity of 40%. The test equipment uses the Xinwei BTS-8000 battery testing system, and the charge and discharge cycle is carried out at a rate of 1C for 800 cycles. At the beginning of each test, the built-in time recording module of the environmental chamber is used to synchronize with the time of the battery testing system to accurately record the test time, including the start time and the end time.

[0116] During each charge and discharge process, the test battery's charge and discharge state parameters, including voltage, current, and internal resistance, were collected and recorded in real time. Simultaneously, electrochemical state parameters, including the battery's open-circuit voltage and AC impedance, were measured using an electrochemical workstation (model CHI660E). For example, during the 100th charge and discharge test, test battery 1 had a charge cutoff voltage of 4.2V, a discharge cutoff voltage of 2.7V, a charging current of 1A, and an internal resistance of 50mΩ. The electrochemical workstation measured its open-circuit voltage of 3.8V and an AC impedance of 0.2Ω at a frequency of 100Hz. The charge and discharge state parameters and electrochemical state parameters of all test batteries corresponding to each test time were stored in a fixed data format (e.g., CSV file) on a 1TB server hard drive. The file name contained information such as the test group number, battery number, and test number. Ultimately, the battery charge and discharge state parameters and electrochemical state parameters corresponding to each test time were obtained.

[0117] S302 : Obtaining a battery capacity deviation under any two charge and discharge tests corresponding to each test battery according to the charge and discharge state parameters of any two charge and discharge tests corresponding to each test battery.

[0118] In this step, the specific implementation process is as follows:

[0119] S3020. Obtain the time difference between any two test times.

[0120] Specifically, when the first test is started, the start time t1-start is accurately recorded through the time recording module built into the environmental chamber and the time synchronization function of the battery tester. After completing a complete 1C rate charge and discharge cycle, the end time t1-end is recorded. The duration of the first test is T1=t1-end-t1-start. The second test is carried out after an interval of 30 days. The start time t2-start and the end time t2-end are also recorded. The duration of the second test is T2=t2-end-t2-start. The time difference ΔT between any two test times is (t2-start-t1-end). For example, if the first test starts at 10:00:00 on October 1, 2024 and ends at 15:00:00 on October 1, 2024, and the second test starts at 10:00:00 on October 31, 2024, then the time difference ΔT=30×24×3600=2592000 seconds. The time difference data is stored in the test time record table in the database.

[0121] S3021. Based on the time difference between any two charge and discharge tests, and according to the charge and discharge state parameters of the corresponding any two charge and discharge tests, obtain the charge and discharge state change amplitude under the corresponding any two charge and discharge tests, so as to obtain the charge and discharge state change rate according to the charge and discharge state change amplitude.

[0122] Specifically, the database retrieves the state of charge (SOC) data of each battery during the two tests. Taking battery 1 as an example, at the end of the first test, =100%, at the end of the second trial =93%, the change range of its charge and discharge state =100%−93%=7%, according to the formula Calculate the rate of change of charge and discharge state, if = 2592000 seconds, then =7% / 2592000×3600×24×365≈8.76% / year. Similarly, calculate the rate of change of the state of energy (SOE). Assuming that the SOE of battery 1 is 95% in the first test and 88% in the second test, the change range is 7%. =7% / 2592000×3600×24×365≈8.76% / year. Combining the SOC and SOE change rates, we can get the comprehensive charge and discharge state change rate of battery 1. ≈8.76% / year. According to this method, the charge and discharge state change rates of the remaining 9 batteries are calculated in turn, and the results are stored in the battery state change rate table in the database. Finally, the charge and discharge state change rates between different batteries at any two test times are obtained.

[0123] S3022. Based on the time difference between any two charge and discharge tests and the charge and discharge state parameters of the any two charge and discharge tests, obtain the charge and discharge current efficiency of the any two charge and discharge tests, so as to obtain the energy conversion efficiency of the any two charge and discharge tests according to the charge and discharge current efficiency.

[0124] Specifically, taking battery 2 as an example, during the first test charging process, the total input charge , the total output charge during the discharge process , then the charge and discharge current efficiency of the first test is , assuming =3000mAh, =2850mAh, =2850 / 3000=95%, charging input energy , discharge output energy , energy conversion efficiency ,like =10.5Wh, =9.5Wh, =9.5 / 10.5≈90.48%. Repeat the above calculation process for the second test and get and , combining the current efficiency and energy conversion efficiency of the two tests, calculate the average energy conversion efficiency of battery 2 .

[0125] The same calculation is performed on other test batteries, and the energy conversion efficiency data of all test batteries are entered into the energy conversion efficiency table of the database, and finally the energy conversion efficiency of the test battery under any two test times is obtained.

[0126] S3023. Obtain the battery capacity deviation between different test batteries under any two charge and discharge tests corresponding to the test battery according to the charge and discharge state change rate and the energy conversion efficiency.

[0127] Specifically, the battery capacity deviation is calculated based on the charge and discharge state change rate and energy conversion efficiency between different test batteries at any two test times, and the charge and discharge state change rate of battery 1 is read from the database. =8.76% / year and average energy conversion efficiency =91%, the charge and discharge state change rate of battery 2 =10% / year and average energy conversion efficiency =88%, taking battery 1 as the benchmark, the capacity deviation coefficient of battery 2 relative to battery 1 , substituting the data into = ≈1.18, assuming the nominal capacity of battery 1 is =2600mAh, then the capacity deviation of battery 2 relative to battery 1 is calculated according to this method. The capacity deviation between the remaining batteries and battery 1 is calculated, and the number of each battery and the capacity deviation relative to the reference battery are recorded. =468mAh. According to this method, the capacity deviations between the remaining batteries and battery 1 are calculated, and the information such as the number of each battery and the capacity deviation relative to the benchmark battery are organized into a table and stored in the battery capacity deviation table in the database. These data can be used to judge the performance differences between batteries and provide key battery status evaluation indicators for electrochemical energy storage early warning based on Kalman filtering, so as to timely detect batteries with abnormal capacity and take targeted maintenance measures.

[0128] Optionally, the specific implementation of S3023 is as follows:

[0129] S30231. Evaluate the state attenuation of the test battery according to the charge and discharge state change rate to obtain the charge and discharge state attenuation coefficient.

[0130] Specifically, the battery state attenuation is evaluated based on the rate of change of charge and discharge states between different test batteries at any two test times. The charge and discharge tests are carried out in a constant temperature and humidity environment box (temperature 25°C, humidity 40%). The first test time is t1. The test battery is subjected to a complete charge and discharge cycle at a rate of 1C using a Xinwei battery tester. The state of charge (SOC) and state of energy (SOE) of each test battery during the charge and discharge process are monitored and recorded in real time.

[0131] The second test time is t2 (t2-t1=30 days, simulating the state after long-term use). Under the same environmental conditions, the charge and discharge cycle is carried out at a 1C rate and the SOC and SOE data are recorded. The charge and discharge state change rate of each test battery in the two test times is calculated.

[0132] Taking SOC as an example, the SOC change rate of battery i is The calculation formula is For example, if the SOC of battery 1 is 100% at the end of charging at t1 and 95% at the end of charging at t2, then its SOC change rate is (100-95) / 30≈0.167% / day. After calculating the SOC and SOE change rates of all batteries, the charge and discharge state attenuation coefficients of other batteries j are calculated based on battery 1. For example, if the SOC change rate of battery 2 is 0.2% / day, then its charge and discharge state attenuation coefficient relative to battery 1 is 0.2 / 0.167≈1.2. These attenuation coefficients are recorded in the database, and the charge and discharge state attenuation coefficients between different test batteries at any two test times are finally obtained.

[0133] S30232. Evaluate the energy loss of the test battery based on the energy conversion efficiency to obtain the charge and discharge energy loss.

[0134] Specifically, the energy conversion efficiency between different test batteries under any two test times is used to calculate the energy loss of the test battery, by recording the charge and discharge voltage, current and time data of each test battery in real time. Taking test battery i as an example, during the charging process, according to the formula Calculate the input energy, where During charging The voltage at the moment, for The current at the moment, is the total charging time; during the discharge process, according to the formula Calculate the output energy, is the total discharge time, the corresponding energy conversion efficiency For example, at t1, the charging input energy of test battery 3 is 100Wh, the discharging output energy is 90Wh, and the energy conversion efficiency is 90 / 100=90%; at t2, the charging input energy is 105Wh, the discharging output energy is 88Wh, and the energy conversion efficiency is 88 / 105≈83.8%. Calculate the difference in energy conversion efficiency between the two test times. , for example, the difference of battery 3 is 90%−83.8%=6.2%, and according to the formula Calculate the battery charge and discharge energy loss. If the charging input energy of battery 3 at t2 is 105Wh, then its energy loss is 105×6.2%=6.51Wh. Use this method to calculate the energy loss of all test batteries at two test times and record it in the corresponding table of the database. Finally, the battery charge and discharge energy loss between different test batteries at any two test times is obtained.

[0135] S30233. Calculate the battery capacity deviation of the charge and discharge energy loss according to the charge and discharge state attenuation coefficient to obtain the battery capacity deviation under any two charge and discharge tests corresponding to the test battery.

[0136] Specifically, the battery capacity deviation is calculated based on the charge and discharge state attenuation coefficient between different test batteries at any two test times, and the battery charge and discharge energy loss between different test batteries at any two test times is read from the database.

[0137] Taking test battery 1 as a reference, the battery capacity deviation of other test batteries j can be For example, if the charge-discharge attenuation coefficient of test battery 4 is 1.3 (relative to battery 1) and the energy loss is 8Wh, then its battery capacity deviation is 8 × 1.3 = 10.4Wh. By traversing all battery data, the battery capacity deviation between different test batteries is calculated. The results are organized into a table containing fields such as battery number, two test times, charge-discharge attenuation coefficient, energy loss, and battery capacity deviation, and stored in a database. This battery capacity deviation data can be used to analyze performance differences between batteries and provide a basis for battery status assessment for electrochemical energy storage early warning based on Kalman filtering, allowing for the timely identification of abnormally performing batteries and the implementation of appropriate maintenance measures.

[0138] S303, obtaining the electrochemical state parameter deviation of the test battery under any two charge-discharge tests according to the electrochemical state parameters of each test battery under any two charge-discharge tests;

[0139] In this step, the electrochemical state parameters of the test battery, such as open circuit voltage, AC impedance and other data, are obtained. Taking the test battery 3 and the test battery 4 at the 200th and 400th tests as examples, the open circuit voltage of the test battery 3 at the 200th test is =3.85V, 400th test =3.8V; open circuit voltage of test battery 4 at the 200th test =3.9V, 400th test =3.82V.

[0140] Calculate the error value of the open circuit voltage for test battery 3 and test battery 4, =|3.8−3.82−(3.85−3.9)|=0.03V. Similarly, the error values ​​of other electrochemical state parameters such as AC impedance under different batteries and different test times are calculated. The electrochemical state parameter error values ​​of all test batteries are sorted into a table, including fields such as battery number, test number1, open circuit voltage error value, AC impedance error value, etc., and stored in the electrochemical state parameter error table of the database to provide a basis for subsequent model correction.

[0141] Correspondingly, before S107, it also includes:

[0142] S1071. Determine two charge and discharge tests with a target time difference between the target time and the current time, and determine a battery capacity deviation between the two charge and discharge tests as a target battery capacity deviation, and an electrochemical state parameter deviation as a target electrochemical state parameter deviation.

[0143] In this step, based on the target time difference, the battery capacity deviation when the time difference between the two charge and discharge tests is the target time difference is determined. If there are multiple battery capacity deviations when the time difference between the two charge and discharge tests is the target time difference, the average value of the multiple battery capacity deviations is taken as the target battery capacity deviation.

[0144] Similarly, the target electrochemical state parameter deviation is determined through comparative calculation.

[0145] S1072. Correct the Kalman filter state model in the battery state dual algorithm fusion model according to the target battery capacity deviation to obtain a corrected Kalman filter state model; correct the P2D electrochemical model in the battery state dual algorithm fusion model according to the target electrochemical state parameter deviation to obtain a corrected P2D electrochemical model.

[0146] In this step, for the Kalman filter state model, the target battery capacity deviation value is substituted into the formula to update the Kalman gain. Assuming that the original Kalman gain , according to the target battery capacity deviation , through a specific formula , obtain the modified Kalman gain, and thus obtain the modified Kalman filter state model.

[0147] For the P2D electrochemical model, the concentration and reaction current density of the target battery electrode are corrected according to the target electrochemical state parameter deviation to obtain the corrected electrode concentration and corrected reaction current density:

[0148] Specifically, taking the open circuit voltage error as an example, the open circuit voltage is closely related to the lithium ion concentration distribution inside the battery. According to the Nernst equation, the open circuit voltage is related to the chemical potential of lithium ions in the electrode material, and the lithium ion concentration directly affects the chemical potential. When there is an open circuit voltage error When the battery thermodynamic model is established, the change in lithium ion concentration can be deduced and The relationship between, assuming that through a large number of experiments and theoretical analysis, the relationship is obtained ,in It is a coefficient obtained by experimental fitting. For example, if the open circuit voltage error of a battery is =0.03V, , then the change in lithium ion concentration is = 0.003 , and then update the concentration of lithium ions in the electrode to .

[0149] For the update of reaction current density, parameters such as AC impedance error can reflect the changes in the kinetic process of electrochemical reactions inside the battery. The charge transfer resistance in the AC impedance spectrum is related to the reaction current density. ,in, , , represents the transfer coefficient, dimensionless and usually takes 0.5, Indicates the number of electrons transferred per mole of active material by lithium ions. Under small AC perturbations (such as AC impedance experiments), the relationship between overpotential and reaction current density can be linearized, thereby analyzing the AC impedance error. , the reaction current density change can be established and relationship, such as , The coefficient is also determined by experiment. Assuming that the AC impedance error of a battery is = 0.05Ω, =10A / Ω, then the change in reaction current density =0.5A, the updated reaction current density is .

[0150] S1073. Obtain a revised P2D electrochemical model based on the updated lithium ion concentration in the electrode and the updated reaction current density.

[0151] In this step, the Kalman gain, electron concentration reaction current density in the battery state dual algorithm fusion model are updated to the corrected Kalman gain, corrected electron concentration and corrected reaction current density, and then the predicted value of the electrochemical energy storage state at the target moment is predicted.

[0152] For example, in the state update equation, by After obtaining the Kalman gain, Obtain the modified Kalman gain, then the Kalman filter state model in the battery state dual algorithm fusion model is adjusted according to the modified Kalman gain. The state vector at the moment is updated to make the obtained state vector more accurate.

[0153] Accordingly, the P2D electrochemical model in the battery state dual algorithm fusion model is based on the updated lithium ion concentration and reaction current density , combining the lithium ion diffusion equation and the transport equation to calculate the predicted value of the lithium ion diffusion coefficient in the electrode and electrolyte and .

[0154] Accordingly, one implementation of S107 is:

[0155] S1074. Based on the current charge and discharge state parameters and electrochemical state parameters, and the revised battery state dual algorithm fusion model, predict the electrochemical energy storage state prediction value corresponding to the target time.

[0156] In this step, the current charge and discharge state parameters and electrochemical state parameters are input into the revised battery state dual algorithm fusion model to measure or predict the electrochemical energy storage state corresponding to the target time.

[0157] In this embodiment, the Kalman gain, the electron concentration and the reaction current density are corrected, and then the battery status dual-algorithm fusion model is updated according to the corrected Kalman gain, the corrected electron concentration and the corrected reaction current density, so that the prediction accuracy of the battery status dual-algorithm fusion model is higher, thereby improving the accuracy of the battery status warning.

[0158] Figure 2 This is a schematic diagram of the structure of the system for establishing a battery status warning model in the electrochemical energy storage provided in the embodiment of the present application. Figure 2As shown, the system for establishing a battery status early warning model in electrochemical energy storage includes: a first model determination module 210, a second model determination module 220, and a model fusion module 230. Optionally, the system for establishing a battery status early warning model in electrochemical energy storage also includes: a prediction module 240 and / or a correction module 250.

[0159] A first model determination module 210 is configured to determine an equivalent circuit model corresponding to the target battery based on the target battery, and to determine a state equation and an observation equation based on the equivalent circuit model corresponding to the target battery; and further configured to determine a Kalman filter state model corresponding to the target battery based on the state equation and the observation equation;

[0160] A second model determination module 220 is configured to determine, based on the electrochemical principle of the target battery, the diffusion equation of electrons in the electrode and the transport equation in the electrolyte of the target battery; and further configured to determine, based on the diffusion equation of electrons in the electrode and the transport equation in the electrolyte, the P2D electrochemical model corresponding to the target battery;

[0161] A model fusion module 230 is configured to obtain a dual-algorithm fusion model of a battery state corresponding to a target battery based on a Kalman filter state model and a P2D electrochemical model. The dual-algorithm fusion model of the battery state is configured to predict a predicted electrochemical energy storage state value corresponding to a target moment based on charge and discharge state parameters and electrochemical state parameters corresponding to the target battery at a first moment. The predicted electrochemical energy storage state value corresponding to the target moment is compared with an actual electrochemical energy storage state value corresponding to the target moment, so as to provide an early warning of the electrochemical energy storage state of the target battery based on the comparison result.

[0162] The charge and discharge state parameters include: charge and discharge voltage, charge and discharge current, charge and discharge resistance, remaining capacity SOC, battery remaining energy state SOE; the electrochemical state parameters include: diffusion coefficient of electrons in electrodes, diffusion coefficient of electrons in electrolytes, radial coordinates of electron particles, first-order and second-order partial derivatives of electron concentration in electrodes with respect to radial coordinates, first-order and second-order partial derivatives of electron concentration in electrolyte with respect to radial coordinates, and reaction current density.

[0163] Optionally, the first model determination module 210 determines the state equation and the observation equation according to the equivalent circuit model corresponding to the target battery, specifically for:

[0164] Obtain the charge and discharge state parameters, process noise, and terminal voltage of the equivalent circuit model at the current moment;

[0165] According to the current charge and discharge voltage, charge and discharge current, and charge and discharge resistance, the input vector corresponding to the current moment is determined, and according to the current SOC and SOE corresponding to the current moment, the state vector corresponding to the current moment is determined;

[0166] Determine the state equation corresponding to the target battery based on the input vector, state vector, and process noise corresponding to the current moment;

[0167] According to the terminal voltage, the observation equation corresponding to the target battery is determined.

[0168] Optionally, the first model determination module 210 determines the Kalman filter state model corresponding to the target battery according to the state equation and the observation equation, specifically for:

[0169] The state prediction equation is obtained according to the state equation, and the state update equation is obtained according to the observation equation;

[0170] According to the state prediction equation and the state update equation, the Kalman filter state model is obtained.

[0171] Optionally, the prediction module 240 is configured to:

[0172] Obtain the charge and discharge state parameters and electrochemical state parameters of the target battery at the current moment, as well as the actual value of the electrochemical energy storage state at the target moment;

[0173] After obtaining the battery state dual-algorithm fusion model corresponding to the target battery based on the Kalman filter state model and the P2D electrochemical model, it also includes:

[0174] Based on the current charge and discharge state parameters and electrochemical state parameters, as well as the battery state dual algorithm fusion model, the electrochemical energy storage state prediction value corresponding to the target time is predicted;

[0175] The predicted value of the electrochemical energy storage state is compared with the actual value of the electrochemical energy storage state, and based on the comparison result, it is determined whether to perform the early warning action.

[0176] Optionally, the correction module 250 is configured to:

[0177] Control multiple test batteries from the same batch as the target battery to perform charge and discharge tests, and record the test time corresponding to each charge and discharge test, as well as charge and discharge state parameters and electrochemical state parameters. The electrochemical state parameters also include: open circuit voltage and AC impedance;

[0178] According to the charge and discharge state parameters of each test battery corresponding to any two charge and discharge tests, the battery capacity deviation under any two charge and discharge tests corresponding to the test battery is obtained;

[0179] According to the electrochemical state parameters of any two charge-discharge tests corresponding to each test battery, the electrochemical state parameter deviation under any two charge-discharge tests corresponding to the test battery is obtained;

[0180] And before predicting the electrochemical energy storage state prediction value corresponding to the target time based on the current charge and discharge state parameters and electrochemical state parameters and the battery state dual algorithm fusion model, it is used to:

[0181] According to the target time difference between the target time and the current time, the battery capacity deviation when the time difference between the two charge and discharge tests is the target time difference is obtained, which is recorded as the target battery capacity deviation, and the electrochemical state parameter deviation, which is recorded as the target electrochemical state parameter deviation;

[0182] The Kalman filter state model in the battery state dual algorithm fusion model is corrected according to the target battery capacity deviation to obtain a corrected Kalman filter state model. The P2D electrochemical model in the battery state dual algorithm fusion model is corrected according to the target electrochemical state parameter deviation to obtain a corrected P2D electrochemical model.

[0183] According to the revised Kalman filter state model and the revised P2D electrochemical model, a revised battery state dual algorithm fusion model is obtained.

[0184] Optionally, the correction module 250 obtains the battery capacity deviation under any two charge and discharge tests corresponding to the test battery according to the charge and discharge state parameters of any two charge and discharge tests corresponding to each test battery, specifically for:

[0185] Get the time difference between any two test times;

[0186] Based on the time difference between any two charge-discharge tests, and according to the charge-discharge state parameters of the corresponding any two charge-discharge tests, the charge-discharge state change amplitude under the corresponding any two charge-discharge tests is obtained, so as to obtain the charge-discharge state change rate according to the charge-discharge state change amplitude;

[0187] Based on the time difference between any two charge and discharge tests, and according to the charge and discharge state parameters of the any two charge and discharge tests, the charge and discharge current efficiency of the any two charge and discharge tests is obtained, so as to obtain the energy conversion efficiency of the any two charge and discharge tests according to the charge and discharge current efficiency;

[0188] According to the charge and discharge state change rate and energy conversion efficiency, the battery capacity deviation between different test batteries under any two charge and discharge tests corresponding to the test battery is obtained.

[0189] Optionally, the correction module 250 obtains the battery capacity deviation between different test batteries under any two charge and discharge tests corresponding to the test battery according to the charge and discharge state change rate and the energy conversion efficiency, specifically for:

[0190] The state attenuation of the test battery is evaluated according to the charge and discharge state change rate to obtain the charge and discharge state attenuation coefficient;

[0191] The energy loss of the test battery is evaluated based on the energy conversion efficiency to obtain the charge and discharge energy loss;

[0192] The battery capacity deviation is calculated based on the charge and discharge energy loss according to the charge and discharge state attenuation coefficient, and the battery capacity deviation under any two charge and discharge tests corresponding to the test battery is obtained.

[0193] Optionally, the prediction module 240 compares the predicted electrochemical energy storage state value with the actual electrochemical energy storage state value, and determines whether to perform a warning action based on the comparison result, including:

[0194] Obtaining the difference between the predicted value of the electrochemical energy storage state and the actual value of the electrochemical energy storage state;

[0195] When the difference is greater than or equal to the pre-set warning value, the warning action is executed.

[0196] The embodiment of the present application provides a system for establishing a battery status early warning model in electrochemical energy storage. The specific implementation process can be found in the above-mentioned method embodiment. Its implementation principles and technical effects are similar and will not be repeated in this embodiment.

[0197] Figure 3 This is a schematic diagram of the structure of an electronic device provided in one embodiment of the present application. Figure 3 As shown, the electronic device includes: a processor 310 and a memory 320.

[0198] The memory 320 stores computer-executable instructions.

[0199] The processor 310 executes the computer-executable instructions stored in the memory 320 , so that the processor 310 performs the method described in any one of the above embodiments.

[0200] The specific implementation process of the electronic device provided in the embodiment of the present application can be found in the above-mentioned method embodiment. Its implementation principle and technical effects are similar, and will not be repeated here in this embodiment.

[0201] In the above Figure 3In the illustrated embodiment, it should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the present invention may be directly implemented by a hardware processor or implemented by a combination of hardware and software modules in the processor.

[0202] The memory may include a high-speed RAM memory, and may also include a non-volatile storage NVM, such as at least one disk storage.

[0203] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be classified into address buses, data buses, and control buses. For ease of illustration, the buses in the drawings of this application are not limited to just one bus or just one type of bus.

[0204] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions. When a processor executes the computer-executable instructions, the method shown in the above method embodiment is implemented.

[0205] The computer-readable storage medium mentioned above can be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The computer-readable storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0206] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist in a device as discrete components.

[0207] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some or all of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for establishing a battery status early warning model in electrochemical energy storage, characterized in that: include: Determining an equivalent circuit model corresponding to the target battery according to the target battery, and determining a state equation and an observation equation according to the equivalent circuit model corresponding to the target battery; Determining a Kalman filter state model corresponding to the target battery according to the state equation and the observation equation; Determining the diffusion equation of electrons in the electrode and the transport equation of electrons in the electrolyte in the target battery according to the electrochemical principle of the target battery; Determining a P2D electrochemical model corresponding to the target battery according to the electron diffusion equation in the electrode and the electron transport equation in the electrolyte; According to the Kalman filter state model and the P2D electrochemical model, a dual-algorithm fusion model of the battery state corresponding to the target battery is obtained, the dual-algorithm fusion model of the battery state is used to predict the electrochemical energy storage state prediction value corresponding to the target moment according to the charge and discharge state parameters and electrochemical state parameters corresponding to the target battery at the first moment, the electrochemical energy storage state prediction value includes: an energy storage state prediction value and an electrochemical state prediction value, the energy storage state prediction value corresponding to the target moment is used to compare with the energy storage state actual value in the electrochemical energy storage state actual value corresponding to the target moment, and the electrochemical state prediction value is used to compare with the electrochemical state actual value in the electrochemical energy storage state actual value, so as to provide an early warning of the electrochemical energy storage state of the target battery according to any comparison result; The charge and discharge state parameters include: charge and discharge voltage, charge and discharge current, charge and discharge resistance, remaining capacity SOC, battery remaining energy state SOE; the electrochemical state parameters include: the diffusion coefficient of the electron in the electrode, the diffusion coefficient of the electron in the electrolyte, the radial coordinates of the electron particles, the first-order partial derivative and the second-order partial derivative of the electron concentration in the electrode with respect to the radial coordinate, the first-order partial derivative and the second-order partial derivative of the electron concentration in the electrolyte with respect to the radial coordinate, and the reaction current density; Also includes: Obtaining the charge and discharge state parameters and electrochemical state parameters of the target battery at the current moment, as well as the actual value of the energy storage state and the actual value of the electrochemical state; After obtaining the battery state dual-algorithm fusion model corresponding to the target battery according to the Kalman filter state model and the P2D electrochemical model, the method further includes: According to the charge and discharge state parameters and electrochemical state parameters at the current moment, and the battery state dual algorithm fusion model, predict and obtain the energy storage state prediction value and the electrochemical state prediction value; Comparing the energy storage state prediction value and the electrochemical state prediction value with the energy storage state actual value and the electrochemical state actual value respectively, and determining whether to execute a warning action according to any comparison result; and controlling a plurality of test batteries from the same batch as the target battery to perform charge and discharge tests, and recording the test time corresponding to each charge and discharge test, as well as charge and discharge state parameters and electrochemical state parameters, wherein the electrochemical state parameters further include: open circuit voltage and AC impedance; Obtaining the battery capacity deviation under any two charge and discharge tests corresponding to the test battery according to the charge and discharge state parameters of any two charge and discharge tests corresponding to each test battery; Obtaining, according to the electrochemical state parameters of any two charge-discharge tests corresponding to each test battery, a deviation of the electrochemical state parameters under any two charge-discharge tests corresponding to the test battery; Before predicting the electrochemical energy storage state prediction value corresponding to the target time based on the charge and discharge state parameters and electrochemical state parameters at the current time and the battery state dual algorithm fusion model, the method further includes: Determining, based on a target time difference between the target moment and the current moment, two charge-discharge tests with a time difference equal to the target time difference, and determining a battery capacity deviation between the two charge-discharge tests as a target battery capacity deviation, and determining the electrochemical state parameter deviation as a target electrochemical state parameter deviation; Correcting the Kalman filter state model in the battery state dual-algorithm fusion model according to the target battery capacity deviation to obtain a corrected Kalman filter state model, and correcting the P2D electrochemical model in the battery state dual-algorithm fusion model according to the target electrochemical state parameter deviation to obtain a corrected P2D electrochemical model; A revised battery state dual-algorithm fusion model is obtained according to the revised Kalman filter state model and the revised P2D electrochemical model.

2. The method according to claim 1, characterized in that The determining of the state equation and the observation equation according to the equivalent circuit model corresponding to the target battery includes: Obtaining the charge and discharge state parameters, process noise, and terminal voltage of the equivalent circuit model at the current moment; Determine the input vector corresponding to the current moment according to the charge and discharge voltage, charge and discharge current, and charge and discharge resistance at the current moment, and determine the state vector corresponding to the current moment according to the SOC and SOE corresponding to the current moment; Determining a state equation corresponding to the target battery according to the input vector, the state vector, and the process noise corresponding to the current moment; An observation equation corresponding to the target battery is determined according to the terminal voltage.

3. The method according to claim 2, characterized in that Determining the Kalman filter state model corresponding to the target battery according to the state equation and the observation equation includes: Obtain a state prediction equation according to the state equation, and obtain a state update equation according to the observation equation; The Kalman filter state model is obtained according to the state prediction equation and the state update equation.

4. The method according to claim 1, wherein Obtaining the battery capacity deviation under any two charge and discharge tests corresponding to the test battery according to the charge and discharge state parameters of any two charge and discharge tests corresponding to each test battery includes: Get the time difference between any two test times; Based on the time difference between any two charge-discharge tests, and according to the charge-discharge state parameters of the corresponding any two charge-discharge tests, the charge-discharge state change amplitude under the corresponding any two charge-discharge tests is obtained, so as to obtain the charge-discharge state change rate according to the charge-discharge state change amplitude; Based on the time difference between any two charge and discharge tests, and according to the charge and discharge state parameters of the any two charge and discharge tests, the charge and discharge current efficiency of the any two charge and discharge tests is obtained, so as to obtain the energy conversion efficiency of the any two charge and discharge tests according to the charge and discharge current efficiency; According to the charge-discharge state change rate and the energy conversion efficiency, a battery capacity deviation between different test batteries under any two charge-discharge tests corresponding to the test battery is obtained.

5. The method according to claim 4, characterized in that Obtaining, based on the charge-discharge state change rate and the energy conversion efficiency, a battery capacity deviation between different test batteries under any two charge-discharge tests corresponding to the test battery, includes: Evaluate the state attenuation of the test battery according to the charge and discharge state change rate to obtain a charge and discharge state attenuation coefficient; Evaluating the energy loss of the test battery according to the energy conversion efficiency to obtain the charge and discharge energy loss; The battery capacity deviation of the charge and discharge energy loss is calculated according to the charge and discharge state attenuation coefficient to obtain the battery capacity deviation under any two charge and discharge tests corresponding to the test battery.

6. The method according to claim 1, characterized in that The step of comparing the energy storage state prediction value and the electrochemical state prediction value with the energy storage state actual value and the electrochemical state actual value, and determining whether to execute a warning action based on any comparison result, includes: Obtaining a difference between the energy storage state prediction value and the energy storage state actual value, and a difference between the electrochemical state prediction value and the electrochemical state actual value; When the difference is greater than or equal to the corresponding warning preset value, a warning action is executed.

7. A system for establishing a battery status early warning model in electrochemical energy storage, characterized in that: include: a first model determination module, configured to determine an equivalent circuit model corresponding to a target battery according to the target battery, and to determine a state equation and an observation equation according to the equivalent circuit model corresponding to the target battery; and further configured to determine a Kalman filter state model corresponding to the target battery according to the state equation and the observation equation; a second model determination module, configured to determine, based on the electrochemical principle of the target battery, a diffusion equation for electrons in the electrode and a transport equation for electrons in the electrolyte in the target battery; and further configured to determine, based on the diffusion equation for electrons in the electrode and the transport equation for electrons in the electrolyte, a P2D electrochemical model corresponding to the target battery; a model fusion module for obtaining a dual-algorithm fusion model of the battery state corresponding to the target battery based on the Kalman filter state model and the P2D electrochemical model, wherein the dual-algorithm fusion model of the battery state is used to predict a predicted electrochemical energy storage state value corresponding to the target moment based on the charge and discharge state parameters and electrochemical state parameters corresponding to the target battery at the first moment, and the predicted electrochemical energy storage state value corresponding to the target moment is used to compare with the actual electrochemical energy storage state value corresponding to the target moment, so as to provide an early warning of the electrochemical energy storage state of the target battery based on the comparison result; The charge and discharge state parameters include: charge and discharge voltage, charge and discharge current, charge and discharge resistance, remaining capacity SOC, battery remaining energy state SOE; the electrochemical state parameters include: the diffusion coefficient of the electron in the electrode, the diffusion coefficient of the electron in the electrolyte, the radial coordinates of the electron particles, the first-order partial derivative and the second-order partial derivative of the electron concentration in the electrode with respect to the radial coordinate, the first-order partial derivative and the second-order partial derivative of the electron concentration in the electrolyte with respect to the radial coordinate, and the reaction current density; The prediction module is used to obtain the charge and discharge state parameters and electrochemical state parameters of the target battery at the current moment, as well as the actual value of the electrochemical energy storage state at the target moment; and after the model fusion module obtains the battery state dual-algorithm fusion model corresponding to the target battery based on the Kalman filter state model and the P2D electrochemical model, it is also used to: predict the electrochemical energy storage state prediction value corresponding to the target moment based on the charge and discharge state parameters and electrochemical state parameters at the current moment, and the battery state dual-algorithm fusion model; Comparing the predicted value of the electrochemical energy storage state with the actual value of the electrochemical energy storage state, and determining whether to execute a warning action based on the comparison result; A correction module is used to control multiple test batteries from the same batch as the target battery to perform charge and discharge tests, and record the test time corresponding to each charge and discharge test, as well as charge and discharge state parameters and electrochemical state parameters. The electrochemical state parameters also include: open circuit voltage and AC impedance; According to the charge and discharge state parameters of each test battery corresponding to any two charge and discharge tests, the battery capacity deviation under any two charge and discharge tests corresponding to the test battery is obtained; According to the electrochemical state parameters of any two charge-discharge tests corresponding to each test battery, the electrochemical state parameter deviation under any two charge-discharge tests corresponding to the test battery is obtained; Before the prediction module predicts the electrochemical energy storage state prediction value corresponding to the target time based on the current charge and discharge state parameters and electrochemical state parameters and the battery state dual algorithm fusion model, it is used to: According to the target time difference between the target time and the current time, the battery capacity deviation when the time difference between the two charge and discharge tests is the target time difference is obtained, which is recorded as the target battery capacity deviation, and the electrochemical state parameter deviation, which is recorded as the target electrochemical state parameter deviation; The Kalman filter state model in the battery state dual algorithm fusion model is corrected according to the target battery capacity deviation to obtain a corrected Kalman filter state model. The P2D electrochemical model in the battery state dual algorithm fusion model is corrected according to the target electrochemical state parameter deviation to obtain a corrected P2D electrochemical model. According to the revised Kalman filter state model and the revised P2D electrochemical model, a revised battery state dual algorithm fusion model is obtained.

8. An electronic device, characterized in that: include: processor and memory; Memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 1 to 6.

Images