A lithium battery soc and soh joint estimation method

Abstract

The application discloses a lithium battery SOC and SOH joint estimation method, and belongs to the field of lithium ion battery state estimation, and comprises the following steps: according to the relationship between the front and rear charging capacity ratio and the capacity loss rate of different end voltage positions of the lithium battery, a nonlinear correlation formula of SOC and SOH is established; based on the second-order RC equivalent circuit model of the lithium battery, the overall model parameter table is identified and created offline; the extended Kalman filtering algorithm is used to identify SOC and substitute into the nonlinear correlation formula to obtain SOH; the overall model parameter table is updated according to the SOH calculation result, and the estimation cycle is realized. The nonlinear correlation formula of SOC and SOH suitable for a wide end voltage range is established, the close coupling relationship between the two is fully considered, the accurate mapping of SOC and SOH at different end voltage positions is realized within one time step, a joint estimation framework more suitable for actual application scenarios is established, and the reliability and accuracy of the lithium battery state joint estimation are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of lithium-ion battery state estimation, in particular to a lithium battery SOC and SOH joint estimation method. BACKGROUND

[0002] Lithium-ion batteries are widely used in communication devices, electric vehicles, energy storage power stations and many other fields due to their high energy density and small size. In actual use, the state of charge (SOC) or the state of health (SOH) of the lithium battery cannot be directly measured by the sensor. These two states are related to the complex electrochemical process inside the battery, so they have a close coupling relationship. Decoupling SOC and SOH and improving the accuracy of joint estimation are the main technical difficulties at present.

[0003] The patent document with publication number CN115825746A discloses a method and device for estimating the state of a lithium battery. The method includes determining a battery model corresponding to the lithium battery to be estimated. The battery model includes an EECM electrochemical equivalent circuit model combined with an EMM chemical mechanism model simulating the lithium ion solid phase diffusion process of the positive and negative electrodes of the lithium battery. The EECM model uses the potential difference on the surface of the lithium ion particles of the positive and negative electrodes of the lithium battery to represent the OCV open circuit voltage of the lithium battery. State equations describing the state changes of the state variables related to the battery model are obtained, and measurement equations corresponding to the battery model are created based on the external characteristic equations of the battery model. Based on the EFK algorithm, the state equations and the measurement equations are iteratively calculated to jointly estimate the SOC and SOH of the lithium battery.

[0004] However, the model established based on the electrochemical mechanism in this invention is very complex, the joint estimation calculation is large, and online estimation is difficult, which has obvious shortcomings.

[0005] The patent document with publication number CN110554324A discloses a SOC and SOH joint estimation method. The method includes: obtaining a training sample set by performing a cyclic charge-discharge experiment on a lithium battery, wherein each training sample in the training sample set includes SOC and environmental temperature; training a SOH estimation neural network based on the training sample set; based on the current SOH of the lithium battery to be measured, estimating the SOC of the lithium battery every first preset time, and when the total time reaches the second preset time, estimating the SOH based on the SOC and environmental temperature using the neural network, to realize the joint estimation of SOC and SOH. However, this invention uses a machine learning joint estimation method, which requires a large number of samples and consumes a lot of manpower and resources.

[0006] In addition, the SOC and SOH coupling relationship of the traditional filtering method is mostly based on the transformation of ampere-hour integral formula, however, the inconsistency of time scale makes the SOH estimation extremely susceptible to measurement noise and the selected estimation window size. In addition, the existing double-filter joint estimation method increases the calculation cost and significantly reduces the calculation stability. Therefore, a new SOC and SOH coupling method is needed to solve the problem of insufficient accuracy of SOC and SOH joint estimation in the prior art. SUMMARY

[0007] In view of the problem that the SOC and SOH are closely coupled in the prior art and it is difficult to ensure the estimation accuracy, the present application provides a lithium battery SOC and SOH joint estimation method, by analyzing the relationship between the ratio of the front and rear charging capacities of the lithium battery at different terminal voltage positions and the capacity loss rate, a nonlinear correlation formula of SOC and SOH suitable for a wide terminal voltage range is established, which significantly reduces the real-time coupling of the SOC and SOH joint estimation process, and completes the SOH prediction within one time step.

[0008] A lithium battery SOC and SOH joint estimation method, comprising:

[0009] Step 1: According to the charging data at different aging stages, the ratio of the front and rear charging capacities of the lithium battery at different terminal voltage positions at different aging stages is obtained, 1-SOH is defined as the capacity loss rate, and a linear relationship between the ratio of the front and rear charging capacities and the capacity loss rate is established by data fitting;

[0010] Step 2: Define the SOC value at the selected terminal voltage as SOC t , according to the linear relationship between SOC t and step 1, the nonlinear correlation formula of SOC t and SOH is obtained by conversion;

[0011] Step 3: Establish a second-order RC equivalent circuit model of the lithium battery, solve the state space equation and measurement equation of the second-order RC equivalent circuit model, and offline identify the model parameters of the second-order RC equivalent circuit model at different aging stages, the relationship between open circuit voltage OCV and SOC OCV-SOC, and establish a total model parameter table;

[0012] Step 4: Identify SOC t by the extended Kalman filtering algorithm, and substitute it into the nonlinear correlation formula obtained in step 2, and calculate SOH within one time step;

[0013] Step 5: According to the SOH calculation result obtained in step 4, the current maximum available capacity is updated, and the overall model parameter table described in step 3 is inquired to update the model parameters of the second-order RC equivalent circuit model and the OCV-SOC, and the next SOC estimation is continued to realize the estimation cycle.

[0014] The present application establishes a nonlinear correlation formula of SOC and SOH suitable for different terminal voltage positions according to the relationship between the ratio of the front and rear charging capacities of the lithium battery at different terminal voltage positions and the capacity loss rate; a second-order RC equivalent circuit model of the lithium battery is established, and a state space equation and a measurement equation are established accordingly; an extended Kalman filtering algorithm is used to identify the SOC at the selected terminal voltage position t , the calculation result of SOH is obtained by substituting the derived nonlinear correlation formula, and the accuracy of SOC estimation in the battery aging process is improved in turn, so as to realize accurate and efficient joint estimation of the battery state in a wide terminal voltage range.

[0015] Further, the charging data of different aging stages refers to the corresponding data provided by the public data set, such as the CALCE battery public data set and the Aachen University battery public data set.

[0016] Further, step 1 is specifically:

[0017] The ratio of the front and rear charging capacities of different terminal voltage positions is Ratio:

[0018]

[0019] Wherein, C b represents the charging capacity from SOC 0 to the selected terminal voltage position in a complete charging process, C a represents the charging capacity from the selected terminal voltage position to full charge in a complete charging process.

[0020] Capacity loss rate:

[0021] Capacity loss=1-SOH (2)

[0022] According to the charging data of different aging stages, the Pearson correlation coefficient is used to analyze the correlation of Ratio and Capacity loss, and the negative linear relationship between Ratio and Capacity loss is obtained:

[0023] Ratio=a·Capacity loss+b (4)

[0024] Wherein a and b are coefficients, which are obtained by least square method.

[0025] Further, step 2 is specifically:

[0026] The SOC value at the selected terminal voltage is defined as SOC t :

[0027]

[0028] Simultaneous equations (4), (5) are obtained SOC t and the expression of Capacity loss:

[0029]

[0030] Converting equation (6) can obtain the SOC corresponding to different terminal voltage positions t and the nonlinear correlation of SOH:

[0031]

[0032] Further, in step 3, a second-order RC equivalent circuit model of the lithium battery is established, including a voltage source U OCV , an ohmic resistance R0, a polarization resistance R s and a capacitor C s in parallel to form an RC network, the terminal voltage is U s , the polarization resistance R p and the capacitor C p in parallel to form an RC network, the terminal voltage is U p , and the measured terminal voltage of the entire circuit is represented by U t .

[0033] Further, in step 3, the state space equation and the measurement equation of the second-order RC equivalent circuit model are solved, SOC and the terminal voltages U s , U p of the two RC networks are selected as state variables to solve the state space equation, and the measured terminal voltage U t of the entire circuit is selected as the measurement variable to establish the measurement equation.

[0034] Further, in step 3, the model parameters of the second-order RC equivalent circuit model in different aging stages are identified, including R0, R s , R p , C s , C p , the ohmic resistance R0 is identified by the instantaneous voltage change after constant current discharge and power-off according to the pulse experiment in different aging stages; the polarization resistance R s , R p and the polarization capacitor C s , C p are identified by the voltage relaxation process after the instantaneous voltage change.

[0035] Further, in step 3, the relationship OCV-SOC between open circuit voltage OCV and SOC is identified offline, the open circuit voltage is the voltage source U in the second-order RC equivalent circuit model OCV The open circuit voltage at different SOC intervals is obtained through low-current open circuit voltage experiments at different aging stages, and the OCV-SOC is obtained by data fitting.

[0036] Further, in step 3, the total model parameter table is obtained, which contains the current maximum available capacity C n , R s , R p , C s , C p , OCV-SOC corresponding to R n .

[0037] The second-order RC equivalent circuit model established for the lithium battery has the characteristics of low complexity and small amount of calculation.

[0038] Further, in step 4, the SOH is calculated in one time step, before the extended Kalman filter algorithm is used, the state space equation and the measurement equation obtained in step 3 need to be converted into a standard form, and the measured terminal voltage U t and the current corresponding to the terminal voltage U t are used as the input of the extended Kalman filter algorithm, and the estimation process is: state parameter initialization, time update, measurement update, to obtain the SOC of the selected terminal voltage position t , and substitute it into the nonlinear correlation in step 2, to calculate the SOH in one time step.

[0039] The SOC is estimated by using the extended Kalman filter algorithm, which has the characteristics of high precision, adaptive battery aging state, small calculation resource demand, and rapid convergence.

[0040] Further, in step 5, the current maximum available capacity is updated, which means that the current maximum available capacity C n in the state space equation is updated according to the SOH estimation result, and the mapping relationship between the current maximum available capacity C n and the SOH is:

[0041] C n = SOH·C n(0) (23)

[0042] Wherein, C n(0) is the rated capacity of the battery.

[0043] Further, in step 5, the equivalent circuit model parameters are updated, which means that the equivalent circuit model parameters are updated according to the calculated C nQuery the overall model parameter table described in step 3, update R0, R s 、R p 、C s 、C p 、OCV-SOC.

[0044] Compared with the prior art, the beneficial effects obtained by the present application are:

[0045] (1) By establishing a nonlinear correlation formula of SOC and SOH suitable for a wide range of end voltages, the close coupling relationship between SOC and SOH is fully considered, and accurate mapping of SOC and SOH at different end voltage positions is realized. The present application establishes a more practical application scenario of SOC and SOH joint estimation framework, and improves the reliability and accuracy of lithium battery state joint estimation;

[0046] (2) The SOH estimation process is simplified to be completed within one time step, and accurate mapping from SOC to SOH is realized at different end voltage positions. Compared with existing machine learning methods, the applicable voltage range is wider, and compared with model-based methods, the calculation cost is lower and the calculation stability is better. BRIEF DESCRIPTION OF DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0048] Figure 1 The flowchart of the lithium battery SOC and SOH joint estimation method of the present application;

[0049] Figure 2 The specific flowchart of the lithium battery SOC and SOH joint estimation method shown in the embodiment of the present application;

[0050] Figure 3 The distribution graph of the ratio and capacity loss of the No. 37 battery in the CS2 battery public data set selected in the present embodiment at different end voltage positions;

[0051] Figure 4 The distribution graph of the ratio and capacity loss of the No. 22, 25, 26, 27, 28 battery in the battery public data set of Aachen University selected in the present embodiment at different end voltage positions;

[0052] Figure 5Pearson correlation coefficient distribution of Ratio and Capacity loss for batteries at different terminal voltage positions in the RWTH Aachen University Battery Public Dataset and the CALCE Battery Public Dataset selected in this embodiment.

[0053] Figure 6 This is a distribution diagram of the ratio and capacity loss at the 4.2V terminal voltage position for batteries No. 22, 25, 27, and 28 in the publicly available battery dataset from RWTH Aachen University selected in this embodiment.

[0054] Figure 7 This is a schematic diagram showing the spatial distribution of the original data of SOC and SOH in the publicly available dataset, based on the nonlinear correlation equation derived in this invention.

[0055] Figure 8 This is a diagram of the second-order RC equivalent circuit model of the lithium battery established in this embodiment;

[0056] Figure 9 To illustrate this embodiment, the extended Kalman filter algorithm is used to determine the State of Charge (SOC) at the 4.2V terminal voltage of battery number 26 in the RWTH Aachen University battery public dataset. t The predicted values ​​are shown in the graph.

[0057] Figure 10 This is a graph showing the SOH prediction results for battery number 26 in the RWTH Aachen University battery public dataset, provided by the nonlinear correlation between SOC and SOH in this embodiment.

[0058] Figure 11 The image shows the SOC estimation results of battery No. 26 in the RWTH Aachen University battery public dataset provided in this embodiment during the constant current charging process at the end of the cycle aging test.

[0059] Figure 12 The image shows the SOC estimation results of battery No. 26 from the RWTH Aachen University battery public dataset provided in this embodiment during the pulse test process at the end of the cycle aging test. Detailed Implementation

[0060] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0061] like Figure 1 As shown in Figure 2, a joint estimation method for SOC and SOH of lithium batteries includes the following steps:

[0062] Step 1: According to the charging data of different aging stages, the ratio of the front and rear charging capacities of the lithium battery at different terminal voltage positions under different aging stages is obtained, 1-SOH is defined as the capacity loss rate, and a linear relationship between the ratio of the front and rear charging capacities and the capacity loss rate is established through data fitting;

[0063] The ratio of the front and rear charging capacities of the lithium battery at different terminal voltage positions:

[0064]

[0065] wherein, C b represents the charging capacity from SOC 0 to the selected terminal voltage position in a complete charging process, C a represents the charging capacity after the selected terminal voltage until full charging in a complete charging process.

[0066] Capacity loss:

[0067] Capacity loss=1-SOH (2)

[0068] In this embodiment, the Ratio and Capacity loss of the CS2 No. 37 battery in the CALCE battery public data set and the No. 22, 25, 26, 27, 28 batteries in the Aachen University battery public data set at different selected terminal voltage positions are analyzed, and are plotted in Figure 3 and Figure 4 respectively, and the voltage interval is 0.02 V.

[0069] The Pearson correlation coefficient p (X,Y) between the Ratio and the Capacity loss is calculated:

[0070]

[0071] wherein, E(XY) represents the joint expectation of data X and data Y, E(X) represents the expectation of data X, and E(Y) represents the expectation of data Y.

[0072] The results of the Pearson correlation coefficient at different terminal voltages are plotted in Figure 5 . It can be found that the Pearson correlation coefficients of the Ratio and the Capacity loss of the two data sets at different terminal voltages are almost less than-0.9, and most of them are less than-0.94, which indicates that the Ratio and the Capacity loss maintain a very high negative linear correlation at almost all selected terminal voltages, and therefore, a linear relationship between the Ratio and the Capacity loss is established:

[0073] Ratio = a Capacity loss + b (4)

[0074] wherein the coefficients a and b are obtained by least square method;

[0075] The Ratio and Capacity loss data of the No. 22, 25, 27 and 28 batteries in the Aachen University battery public data set at the 4.2V terminal voltage position are selected as examples and plotted in Figure 6 .

[0076] Step 2: define the SOC value at the selected terminal voltage as SOC t , according to the linear relationship between SOC t and step 1, the nonlinear correlation formula between SOC t and SOH is obtained by conversion;

[0077] Define the SOC value at the selected terminal voltage as SOC t :

[0078]

[0079] The formula (5) and the formula (4) in step 1 are combined to obtain the relationship formula between SOC t and Capacity loss, which is expressed as follows:

[0080]

[0081] The formula (6) is converted:

[0082]

[0083] The (2) (7) are combined to obtain the nonlinear correlation formula between the SOC t and SOH:

[0084]

[0085] The SOH calculated according to the formula (8) and the original data are shown in Figure 7 , it can be seen that the nonlinear correlation formula derived is highly consistent with the actual data, and the root mean square error of the SOH solved according to the formula (8) is only 0.01017.

[0086] Step 3: establish a second-order RC equivalent circuit model of the lithium battery, solve the state space equation and measurement equation of the second-order RC equivalent circuit model, and offline identify the model parameters of the second-order RC equivalent circuit model at different aging stages, the relationship between open circuit voltage OCV and SOC OCV-SOC, and establish a general model parameter table;

[0087] An example of the second-order RC equivalent circuit model is shown inFigure 8 The second-order RC equivalent circuit model shown, containing voltage source U OCV , ohmic internal resistance R0, polarization resistance R s and capacitor C s in parallel form an RC network, and the terminal voltage is U s , polarization resistance R p and capacitor C p in parallel form an RC network, and the terminal voltage is U p , and the terminal voltage of the entire circuit is represented by U t ;

[0088] The differential equation set describing the second-order RC equivalent circuit model can be expressed as follows:

[0089]

[0090] where I represents the current; and the formula (9) is discretized to obtain:

[0091]

[0092] where k represents the current time step, k-1 is the previous time step, τ s =R s ·C s , τ p =R p ·C p are the time constants of the two RC networks, and T represents the sampling interval;

[0093] According to the ampere-hour integration method, the definition of SOC is:

[0094]

[0095] where t0 represents the initial time, t represents the current time, I(t) is the current at the current time, I(t) greater than 0 indicates that the battery is discharging, I(t) less than 0 indicates that the battery is charging, η is the coulomb efficiency, and Cn is the current maximum available capacity of the current battery;

[0096] In combination with formula (11), SOC and terminal voltage U s , U p are selected as state variables, and the state space equation is solved:

[0097]

[0098] The terminal voltage U t is selected as the measurement variable, and the measurement equation is expressed as:

[0099] U t,k =U OCV,k -U p,k -Us,k -I k ·R 0,k (13)

[0100] According to the pulse test data of different aging stages of battery No. 22 provided by the battery public data set of Aachen University of Technology, different C n corresponding to different SOC values Battery parameters:

[0101] The ohmic internal resistance R0 is identified by the instantaneous voltage change after constant current discharge and power-off, which is expressed as follows:

[0102]

[0103] ΔU represents the voltage change value at the moment of power-off, and I is the current size of the constant current discharge process before power-off;

[0104] The polarization resistance and capacitance are identified by the voltage relaxation process after the instantaneous voltage change, and the relaxation voltage U r (t) is expressed as:

[0105]

[0106] In the formula, t p is the power-on time before relaxation;

[0107] The parameter values can be obtained by applying the least square method, and the objective function is expressed as:

[0108]

[0109] In the formula, U r,real (k) represents the true relaxation voltage at the current time step, and the parameters that minimize the objective function J are the equivalent circuit model parameters at this time;

[0110] According to the low-current open-circuit voltage experimental data of different aging stages of battery No. 22 provided by the battery public data set of Aachen University of Technology, the different C n corresponding to OCV-SOC is obtained by data fitting.

[0111] According to the obtained R0, R s , R p , C s , C p , OCV-SOC, and the corresponding current maximum available capacity C n , the overall model parameter table is established.

[0112] Step 4: Identify SOC t by the extended Kalman filter algorithm, and substitute it into the nonlinear correlation formula obtained in step 2 to calculate SOH within a time step;

[0113] Convert the formula (12) (13) to the canonical form as follows:

[0114]

[0115] where, represents the lithium battery state space matrix, y k = U t,k represents the measurement matrix, A k-1 , B k-1 , C k , D k represents the coefficient, u k-1 represents the system input at time k-1, ω k-1 and v k are the process noise and measurement noise, respectively;

[0116] Apply extended Kalman filter to predict SOC t , the steps are as follows:

[0117] 1) State parameter initialization

[0118] Initialize state vector x0, process noise covariance Q, measurement noise covariance R and initial error covariance P0

[0119] 2) Time update

[0120] State estimation:

[0121]

[0122] where, represents the optimal estimation of system state at time k-1, represents the prediction of system state at time k according to and system input u k-1 at time k-1;

[0123] Error covariance prediction:

[0124]

[0125] where, P k-1 represents the corresponding error covariance matrix, P (k|k-1) represents the error covariance matrix of k-1 , A k-1 represents the Jacobian matrix of formula (18);

[0126] 3) Measurement update

[0127] Calculate the Kalman gain matrix:

[0128]

[0129] where K k represents the Kalman gain matrix, C k represents the Jacobian matrix of the output equation;

[0130] State update:

[0131]

[0132] where, represents the optimal estimation of the system state at time k, represents the output equation;

[0133] Error covariance update:

[0134] P k = (1-K k C k )P (k|k-1) (22)

[0135] where P k represents the corresponding error covariance matrix, and the SOC t is obtained by substituting it into equation (8) to calculate the SOH value within one time step.

[0136] Step 5: According to the SOH calculation result obtained in step 4, update the current maximum available capacity, and query the overall model parameter table described in step 3 to update the model parameters of the second-order RC equivalent circuit model and the OCV-SOC, continue the next SOC estimation, and realize the estimation cycle.

[0137] According to the SOH calculation result obtained in step 4, update the current maximum available capacity C n , where the relationship between SOH and C n can be expressed as:

[0138] C n = SOH·C n(0) (23)

[0139] C n(0) is the rated capacity of the battery;

[0140] Query the overall model parameter table described in step 3 to update the equivalent circuit model parameters,

[0141] In this embodiment, the model parameters of the equivalent circuit model are updated every N1 period; the relationship between the open-circuit voltage and the SOC is updated every N2 period, and N1 and N2 are different time intervals selected;

[0142] Thereafter, the extended Kalman filter algorithm is used to continue the next SOC estimation process.

[0143] Experimental results

[0144] like Figure 9 As shown, the SOC of battery #26 in the RWTH Aachen University Battery Public Dataset is calculated using extended Kalman filtering. t The results closely match the actual values. Testnumber indicates the number of tests, with each test spaced 30 aging cycles apart. SOC t The mean absolute error and root mean square error are only 0.006215186 and 0.007334851, respectively.

[0145] like Figure 10 The figure shows the prediction results of SOC for battery No. 26 in the RWTH Aachen University battery public dataset. The method described here calculates SOH in one time step while ensuring the accuracy of the results. The mean absolute error and root mean square error of SOH are only 1.5021% and 1.8148%, respectively.

[0146] like Figure 11 The figure shows the SOC estimation results of the constant current charging process of battery No. 26 in the RWTH Aachen University battery public dataset at the end of the aging experiment (test number = 23). The results show that the initial error can converge quickly and maintains high accuracy throughout the process, with the mean absolute error and root mean square error being only 0.6087% and 0.7719%, respectively.

[0147] like Figure 12 The figure shows the SOC estimation results of the pulse test process at the end of the aging experiment (test number = 23) of battery No. 26 in the RWTH Aachen University battery public dataset. Compared with the traditional extended Kalman filter method, the joint estimation method provided by this invention has significantly reduced the error, with the mean absolute error and root mean square error being 2.0082% and 2.1637%, respectively.

[0148] The results show that the joint estimation method has high reliability and accuracy.

Claims

1.A method for combined SOC and SOH estimation of a lithium battery, characterized in that, The method comprises the following steps: Step 1: According to the charging data of different aging stages, the front and rear charging capacity ratio of the lithium battery at different voltage positions in different aging stages is obtained, and the front and rear charging capacity ratio is defined as The capacity loss rate is defined, and a linear relationship between the front and rear charging capacity ratio and the capacity loss rate is established through data fitting. The front-to-rear charging capacity ratio at the different terminal voltage positions Ratio: (1) wherein, represents the charge capacity from SOC of 0 to the selected terminal voltage position in a complete charging process, represents the charge capacity after the selected terminal voltage until full charge in a complete charging process; rate of capacity loss : (2) According to the charging data of different aging stages, the Pearson correlation coefficient is used to analyze the correlation of and , and the negative linear relationship between and is obtained. (4) wherein and are coefficients, obtained by least square calculation; Step 2: Define the SOC value at the selected terminal voltage as , according to the linear relationship described in Step 1, obtain a nonlinear correlation formula with SOH through conversion; The : (5) Combining formula (5) with formula (4) in step 1 gives with the relationship ​ (6) Converting formula (6): (7) (2) (7) to obtain said Nonlinear correlation with SOH: (8)； Step 3: A second-order RC equivalent circuit model of the lithium battery is established, state space equations and measurement equations of the second-order RC equivalent circuit model are solved, model parameters of the second-order RC equivalent circuit model in different aging stages, a relationship between open-circuit voltage OCV and SOC OCV-SOC are identified offline, and a general model parameter table is established; Step 4: Identify by extended Kalman filter algorithm and substitute into the nonlinear correlation formula obtained in step 2, and calculate the SOH in a time step. Step 5: According to the SOH calculation result obtained in step 4, the current maximum available capacity is updated, the general model parameter table in step 3 is inquired, the model parameters of the second-order RC equivalent circuit model and the OCV-SOC are updated, the next SOC estimation is continued, and the estimation cycle is realized. 2.The lithium battery SOC and SOH joint estimation method according to claim 1, characterized in that, The charging data of different aging stages refer to the corresponding data provided by the public data set. 3.The lithium battery SOC and SOH joint estimation method according to claim 1, characterized in that, In step 3, the second-order RC equivalent circuit model includes a voltage source , an ohmic internal resistance , a polarization internal resistance , and a capacitor in parallel, forming an RC network, and the terminal voltage is , the polarization internal resistance , and the capacitor in parallel, forming an RC network, and the terminal voltage is , and the entire circuit measured terminal voltage is represented by . 4.The lithium battery SOC and SOH joint estimation method according to claim 3, characterized in that, In step 3, the model parameters of the second-order RC equivalent circuit model in different aging stages are identified offline, including 、 、 、 、 , the pulse experiment is adopted, and the instantaneous voltage change after constant current discharge and power-off is used to identify the ohmic internal resistance ; the voltage relaxation process after the instantaneous voltage change is used to identify the polarization internal resistance 、 and the polarization capacitance 、 . 5.The lithium battery SOC and SOH joint estimation method according to claim 3, characterized in that, The open circuit voltage described in step 3 is the voltage source in the second order RC equivalent circuit model OCV-SOC is obtained by low current open circuit voltage experiment. 6.The lithium battery SOC and SOH joint estimation method according to claim 3, characterized in that, In step 3, the table of overall model parameters contains the current maximum available capacity and its corresponding , , , , , OCV-SOC. 7.The lithium battery SOC and SOH joint estimation method according to claim 1, characterized in that, In Step 5, the updating of the current maximum available capacity refers to updating the current maximum available capacity in the state space equation according to the SOH estimation result The current maximum available capacity in the above equation is updated according to the SOH estimation result The mapping relationship with the SOH is as follows: (23) wherein, C is the battery rated capacity. 8.The lithium battery SOC and SOH joint estimation method according to claim 6, characterized in that, In step 5, the model parameters of the updated second order RC equivalent circuit model and OCV-SOC are referred to as the model parameters of the updated second order RC equivalent circuit model and OCV-SOC calculated in step 4. The table of overall model parameters referred to in step 3 is queried, and the model parameters of the second order RC equivalent circuit model and OCV-SOC are updated 、 、 、 、 and OCV-SOC, and the next SOC estimation is continued with the updated model parameters of the equivalent circuit model and OCV-SOC.

Images