A method for estimating Soc of lithium battery

Abstract

The present invention provides a lithium battery Soc estimation method in the field of lithium battery Soc estimation technology, including: step S10, the lithium battery is equivalent to a first-order Thevenin circuit, and the output voltage and output current of the lithium battery are collected in real time; step S20, a first state space equation group is constructed based on the first-order Thevenin circuit, the output voltage and the output current; step S30, the first-order Thevenin circuit is identified online by the deviation compensation least squares method, the circuit parameters of the first-order Thevenin circuit are obtained, and the first state space equation group is updated based on the circuit parameters; step S40, the first state space equation is discretized by Euler transformation to obtain a second state space equation group; step S50, the second state space equation group is continuously updated by the joint improved adaptive attenuated extended Kalman filter algorithm; step S60, the lithium battery Soc is estimated based on the second state space equation group. The advantages of the present invention are: the accuracy and anti-interference performance of Soc estimation are greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of lithium battery Soc estimation, and in particular to a lithium battery Soc estimation method. Background Art

[0002] In today's era of advocating green economy and sustainable development, new energy vehicles are considered to be an important direction for future development, and power batteries play a key role in new energy vehicles. They can not only store energy, but also convert electrical energy into chemical energy. At present, power batteries mainly use lead-acid batteries, lead-carbon batteries and lithium batteries.

[0003] As lithium battery technology is still immature, it has shortcomings such as short driving range on a single charge, poor safety performance and short service life. In order to ensure the safe and efficient operation of power batteries during use, lithium batteries need to be managed and controlled through a battery management system BMS. The basis of management and control is the estimation of the lithium battery Soc. By transmitting the estimated Soc to the vehicle control unit, it can prevent the power battery of new energy vehicles from overcharging and over-discharging during operation, thereby improving the reliability and safety of the power battery.

[0004] Traditionally, there are the following methods for estimating the Soc of lithium batteries:

[0005] 1. The ampere-hour integration method will produce cumulative errors. As the battery works for a long time, the error will increase, and it is overly dependent on the initial Soc value.

[0006] 2. Kalman Filter (KF), which is a recursive filtering algorithm that uses state-space equations to describe the system. It is a recursive linear minimum variance estimation method. Its disadvantage is that it is limited to linear systems, and lithium batteries have strong nonlinear characteristics. Therefore, the Kalman filter algorithm cannot be directly used to estimate the Soc of lithium batteries.

[0007] 3. Extended Kalman Filter (EKF) algorithm uses Taylor series to linearize non-systems, ignores higher-order terms and retains first-order terms to estimate Soc. However, when the system noise is uncertain, it is easy to cause filter divergence. The EKF algorithm is highly dependent on the model. The nonlinear function of the observation equation ignores higher-order terms when linearizing, which is prone to linearization errors. Moreover, the BMS will be accompanied by noise signals during the current acquisition process, resulting in poor accuracy of Soc estimation.

[0008] 4. Neural network algorithm: It does not require the establishment of an accurate circuit model and has the advantages of high nonlinearity and strong self-learning ability. However, its training process is too complicated and is not suitable for real embedded systems. It is also overly dependent on sample data.

[0009] 5. Open circuit voltage method. This method has certain reliability. However, this method is under ideal experimental conditions. It obtains OCV values ​​under different Socs by standing still for a long time, and then estimates Soc using the table lookup method. As the lithium battery is used for longer and longer, the accuracy of its Soc estimation will gradually decrease due to the influence of battery aging factors. Even if there is a functional relationship under experimental conditions, this functional relationship will also be affected by external environmental factors, such as temperature.

[0010] Therefore, how to provide a lithium battery Soc estimation method to improve the accuracy of Soc estimation and anti-interference performance has become a technical problem that needs to be solved urgently. Summary of the invention

[0011] The technical problem to be solved by the present invention is to provide a lithium battery Soc estimation method to improve the accuracy of Soc estimation and anti-interference performance.

[0012] The present invention is implemented as follows: a lithium battery Soc estimation method comprises the following steps:

[0013] Step S10, the lithium battery is equivalent to a first-order Thevenin circuit, and the output voltage and output current of the lithium battery are collected in real time;

[0014] Step S20, constructing a first state space equation group based on the first-order Thevenin circuit, the output voltage and the output current;

[0015] Step S30, performing online identification on the first-order Thevenin circuit by using a deviation-compensated least squares method to obtain circuit parameters of the first-order Thevenin circuit, and updating the first state-space equation group based on the circuit parameters;

[0016] Step S40, performing Euler transformation discretization on the first state space equation to obtain a second state space equation group;

[0017] Step S50, continuously updating the second state space equations by combining an improved adaptive attenuated extended Kalman filter algorithm;

[0018] Step S60: estimating the lithium battery Soc based on the second state-space equations.

[0019] Further, in step S10, the first-order Thevenin circuit includes a resistor R0, a resistor R1, a capacitor C1 and an equivalent battery cell; after the resistor R1 and the capacitor C1 are connected in parallel, one end is connected to the resistor R0; the resistor R0 is connected to the positive electrode of the equivalent battery cell;

[0020] The output voltage is U L , the output current is I L .

[0021] Furthermore, in step S20, the construction process of the first state space equation group is as follows:

[0022] According to Kirchhoff's voltage law, we get the following equations:

[0023]

[0024] The calculation formula for the state of charge of a lithium battery is as follows:

[0025]

[0026] Combining equation (1) and taking the derivative of equation (2), we get the first state space expression:

[0027]

[0028] Among them, U1 represents the polarization voltage across capacitor C1; dU1 represents the differential of U1; dt represents the differential of charge and discharge time; Soc represents the remaining capacity of the equivalent battery cell; Q represents the actual capacity of the lithium battery; Indicates the derivation of U1; It represents the derivative of Soc; Soc(0) represents the initial value of the state of charge of the lithium battery; Soc(t) represents the state of charge of the lithium battery at time t; η represents the charge and discharge efficiency of the lithium battery, which is approximately 1; Uoc(Soc) represents the open circuit voltage value; Uoc represents the voltage of the equivalent battery cell.

[0029] Furthermore, the step S30 specifically includes:

[0030] Step S31, online identification of the first-order Thevenin circuit is performed by using the deviation-compensated least squares method to obtain identification parameters θ1, θ2, θ3:

[0031]

[0032]

[0033]

[0034] Wherein, T' represents the sampling time; τ represents the time constant, and τ=R1C1;

[0035] Step S32: Calculate the circuit parameters including the values ​​of R0, R1 and C1 based on θ1, θ2, θ3 and τ:

[0036]

[0037]

[0038]

[0039] Further, in step S40, the formula of the second state space equation group is:

[0040]

[0041] Wherein, Δt represents the sampling time of Euler transformation; k represents the charging and discharging k moment after Euler transformation.

[0042] Furthermore, the step S60 is specifically as follows:

[0043] Based on the second state-space equation group, the Soc value Soc(k+1) at the k+1 moment is output to estimate the Soc of the lithium battery at the next moment.

[0044] The advantages of the present invention are:

[0045] By using the first-order Thevenin circuit to equate the lithium battery, the static and dynamic characteristics of the lithium battery can be more accurately reflected. By using the deviation-compensated least squares method to perform online identification on the first-order Thevenin circuit, the problem of unbiased identification model caused by data saturation, uncertainty noise and other problems in the conventional least squares method is overcome. By using the joint improved adaptive attenuated extended Kalman filter algorithm to update the second state-space equations, compared with the traditional ampere-hour integration method and the extended Kalman filter algorithm, it has better anti-interference performance, and ultimately greatly improves the accuracy of Soc estimation and anti-interference performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The present invention will be further described below in conjunction with embodiments with reference to the accompanying drawings.

[0047] Figure 1 It is a flow chart of a lithium battery Soc estimation method of the present invention.

[0048] Figure 2 1 is a circuit diagram of a first-order Thevenin circuit of the present invention.

[0049] Figure 3 It is a fitting curve diagram of Soc and OCV of the present invention.

[0050] Figure 4 It is a schematic diagram of Soc estimation comparison of the present invention.

[0051] Figure 5 It is a schematic diagram of the comparison of Soc estimation errors of the present invention. DETAILED DESCRIPTION

[0052] The technical solution in the embodiment of the present application has the following overall idea: based on the first-order Thevenin circuit, the deviation-compensated least squares method and the joint improved adaptive attenuated extended Kalman filter algorithm are combined to improve the anti-noise performance, so as to improve the accuracy of Soc estimation.

[0053] Please refer to Figures 1 to 5 As shown, a preferred embodiment of a lithium battery Soc estimation method of the present invention comprises the following steps:

[0054] Step S10, the lithium battery is equivalent to a first-order Thevenin circuit, and the output voltage and output current of the lithium battery are collected in real time;

[0055] Step S20, constructing a first state space equation group based on the first-order Thevenin circuit, the output voltage and the output current;

[0056] Step S30, performing online identification on the first-order Thevenin circuit by using a deviation-compensated least squares method to obtain circuit parameters of the first-order Thevenin circuit, and updating the first state-space equation group based on the circuit parameters;

[0057] Step S40, performing Euler transformation discretization on the first state space equation to obtain a second state space equation group;

[0058] Step S50, continuously updating the second state space equations by combining an improved adaptive attenuated extended Kalman filter algorithm;

[0059] Step S60: estimating the lithium battery Soc based on the second state-space equations.

[0060] In step S10, the first-order Thevenin circuit includes a resistor R0, a resistor R1, a capacitor C1 and an equivalent battery cell; after the resistor R1 and the capacitor C1 are connected in parallel, one end is connected to the resistor R0; the resistor R0 is connected to the positive electrode of the equivalent battery cell;

[0061] The output voltage is U L , the output current is I L .

[0062] The key to estimating Soc (state of charge) through algorithms is to establish a suitable equivalent model for lithium batteries. Common equivalent models include black box models, electrochemical models, neural network models, equivalent circuit models, etc. The black box model treats the lithium battery as a black box. It does not need to understand the internal material parameters and chemical reaction process of the battery, but can only reflect the simple correspondence of experimental data; the electrochemical model describes the changes in parameters such as battery terminal voltage and electromotive force through the chemical reaction inside the battery, which can well reflect the chemical reaction inside the battery, but this further complicates the model, and the establishment of the model will be too complicated and the amount of calculation is too large; neural network modeling requires a large amount of sample data for training, and the conditions are difficult to achieve; the equivalent circuit model describes the static and dynamic characteristics of the battery through components such as resistors R and capacitors C, which can effectively reflect the internal mechanism of the battery, and has a simple structure and is easy to calculate. Since the first-order Thevenin circuit can more accurately reflect the static and dynamic characteristics of the battery, the first-order Thevenin circuit is used to perform the equivalent of the lithium battery.

[0063] In step S20, the construction process of the first state space equation group is as follows:

[0064] According to Kirchhoff's voltage law, we get the following equations:

[0065]

[0066] The calculation formula for the state of charge of a lithium battery is as follows:

[0067]

[0068] Combining equation (1) and taking the derivative of equation (2), we get the first state space expression:

[0069]

[0070] Among them, U1 represents the polarization voltage across capacitor C1; dU1 represents the differential of U1; dt represents the differential of charge and discharge time; Soc represents the remaining capacity of the equivalent battery cell; Q represents the actual capacity of the lithium battery; Indicates the derivation of U1; It represents the derivative of Soc; Soc(0) represents the initial value of the state of charge of the lithium battery; Soc(t) represents the state of charge of the lithium battery at time t; η represents the charge and discharge efficiency of the lithium battery, which is approximately 1; Uoc(Soc) represents the open circuit voltage value, which is calculated based on a certain functional relationship between Uoc and Soc; Uoc represents the voltage of the equivalent battery cell.

[0071] The step S30 specifically includes:

[0072] Step S31, online identification of the first-order Thevenin circuit is performed by using the deviation-compensated least squares method to obtain identification parameters θ1, θ2, θ3:

[0073]

[0074]

[0075]

[0076] Wherein, T' represents the sampling time; τ represents the time constant, and τ=R1C1;

[0077] Step S32: Calculate the circuit parameters including the values ​​of R0, R1 and C1 based on θ1, θ2, θ3 and τ:

[0078]

[0079]

[0080]

[0081] The premise of establishing an accurate model is to identify the unknown parameters in the model. Common identification methods include offline identification and online identification. Offline identification includes pulse discharge method and least squares method. The pulse discharge method can be used to perform HPPC experiments according to the FreedomCAR Battery Experiment Manual to achieve parameter identification of the battery model; the least squares method (LS) will have data saturation problems as the amount of data in the identification system increases. As the battery continues to work, the battery model parameters change, especially for batteries that have been used for a long time. This will cause the time-varying parameters to not be tracked well. At this time, the parameters of offline identification will cause large errors, so online identification algorithms are usually selected. Online identification algorithms include recursive least squares (RLS), weighted recursive adaptive least squares, recursive least squares with forgetting factor, and bias-compensated recursive least squares (BCRLS). When a lithium battery is working, the collection of current and voltage is often accompanied by uncertain noise signals, which will cause deviations in the identified results. The conventional least squares method will cause the identified model to no longer be unbiased due to problems such as data saturation and uncertain noise. Therefore, the present invention selects the deviation-compensated least squares method for real-time online identification of the battery model (first-order Thevenin circuit).

[0082] In step S31, the detailed process of identifying the identification parameters θ1, θ2, and θ3 is as follows:

[0083] According to Kirchhoff's voltage law, in the S domain (complex frequency domain) U L and I L The following relationship exists:

[0084]

[0085] Wherein, τ represents the time constant, and τ=R1C1;

[0086] Based on the above formula, the transfer function G(S) can be derived:

[0087]

[0088] Wherein, the transfer function G(S) represents the ratio of the Laplace transform of the linear system response (i.e., output) quantity to the Laplace transform of the excitation (i.e., input) quantity under zero initial conditions;

[0089] Perform a bilinear transformation on the transfer function G(S) and set s=2(1+z -1 ) / T'(1-z -1 ), and the discretized transfer function G(z -1 ):

[0090]

[0091] Where T' represents the sampling time; θ1, θ2, θ3 represent the parameters to be identified; k represents the current time, i.e., the time index;

[0092] The discretized transfer function G(z -1 ) can be further deduced:

[0093] U L (k)-Uoc(k)=θ1(U L (k)-Uoc(k-1))+θ2I L (k)+θ3I L (k-1);

[0094] Since the time T' is a very small interval, Uoc(k) can be regarded as a constant value within a sampling interval, and the above formula can be simplified to:

[0095] U L (k)=θ1U L (k-1)+θ2I L (k)+θ3I L (k-1)+(1-θ1)Uoc(k-1);

[0096] Define the parameter to be identified as θ4 = (1-θ1)Uoc(k-1), and then simplify the above formula to:

[0097] U L (k)=θ1U L (k-1)+θ2I L(k)+θ3I L (k-1)+θ4;

[0098] definition As the parameter matrix to be identified, the first-order Thevenin circuit is identified online using the following formula of the least square method of deviation compensation to obtain the identification parameters θ1, θ2, and θ3:

[0099]

[0100]

[0101]

[0102]

[0103]

[0104]

[0105]

[0106]

[0107] in, Indicates U L (k-1), I L (k) I L The result of transposing (k-1) and 1; e(k) represents the error at time k; K represents the gain matrix; represents the recursive least squares method to estimate the parameters; J(k) represents the J value corresponding to the error criterion function at time k; λ represents the forgetting factor; σ 2 (k) represents the estimated noise variance at time k; D is the correlation matrix; P(k) represents the updated covariance matrix at time k; represents the estimated parameters after recursive least squares bias compensation.

[0108] In step S40, the formula of the second state space equation group is:

[0109]

[0110] Wherein, Δt represents the sampling time of Euler transformation; k represents the charging and discharging k moment after Euler transformation.

[0111] The step S50 is specifically as follows:

[0112] The U1(k) and Uoc(k) in the second state space equations are continuously updated by the joint improved adaptive decay extended Kalman filter algorithm. The calculation process of the joint improved adaptive decay extended Kalman filter algorithm is as follows:

[0113] The EKF state estimation time update, error covariance time update, Kalman gain matrix calculation, error estimation measurement update, state estimation measurement update, and error covariance measurement update equations are as follows:

[0114]

[0115] For FEKF, the error covariance time update P k|k-1 =A k-1 P k-1|k-1 A T k-1 +Q k-1 Change to P k|k-1 =A k-1 αP k-1|k-1 A T k-1 +Q k-1 , where α is the forgetting factor.

[0116] represents the posterior state estimate at time k; Represents the prior state estimate at time k-1; represents the prior state estimate at time k; Q k-1 represents the k-1 moment process noise w before improvement k Variance of K k represents the Kalman gain matrix; e k represents the error estimate at time k; e T k represents the transpose of the error estimate at time k; K T k represents the transpose of the Kalman gain matrix at time k; y k Represents the output value at time k; represents the measurement function; R k-1 represents the measurement noise v at time k-1 before improvement k The variance of .

[0117] Based on FEKF, the process noise w is redefined k ~(q k ,Q k )(i.e. w k The mean value of k , the variance is Q k ) and measurement noise ν k ~(r k ,Rk )(i.e. v k The mean value is r k , the variance is R k ), the two are independent of each other. The noise mean is no longer 0 as before, but is defined as q k and r k , and the state time estimate is updated Change to Error Estimate Update Updated to

[0118] One problem with the traditional Sage-Husa adaptive filter is that the filtering result may diverge and the accumulated error becomes larger and larger. and

[0119] R k =(1-d k-1 )R k-1 +d k-1 (e k e T k -H k P k|k-1 H T ) are respectively redefined as Q k ' and R k ':

[0120]

[0121] Then the improved adaptive noise covariance is updated as follows:

[0122]

[0123] Among them, q k represents the process noise w at time k k The mean value of Q k ′ represents the improved process noise w at time k k The variance of k represents the measurement noise v at time k k The mean value of R k ′ represents the improved measurement noise v at time k k The variance of d represents the variable factor; d k represents the variable factor value of d updated at time k; b represents the forgetting factor, and 0<b<1.

[0124] The step S60 is specifically as follows:

[0125] Based on the second state-space equation group, the Soc value Soc(k+1) at the k+1 moment is output to estimate the Soc of the lithium battery at the next moment.

[0126] Experimental verification:

[0127] The estimated Soc and OCV (open circuit voltage) are fitted by cftool in the matlab toolbox, such as Figure 3 shown.

[0128] OCV is the voltage value obtained by leaving the lithium battery at different Soc for a long time. The present invention obtains the voltage value of the lithium battery by leaving the lithium battery at Soc of 100%, 90%, 80% ... 0% for a long time, and then fits it through the cftool toolbox in matlab, and fits it 6 times to obtain:

[0129] Uoc=10.06soc 6 -20.53soc 5 +7.103soc 4 +11soc 3 -10.48soc 2 +3.397soc+2.813;

[0130] Among them U oc Indicates the voltage value of OCV.

[0131] Taking the test data of lithium battery laboratory as the object, the verification results of using EKF, FEKF and improved AFEKF (the present invention) to perform online Soc estimation of lithium battery during charging and discharging are as follows Figure 4 , Figure 5 shown.

[0132] from Figure 4 , Figure 5 It can be seen that the maximum absolute error of EKF estimating Soc is 12.6%, and the maximum absolute error of FEKF estimating Soc is 4.2%. Although the improved AFEKF algorithm has a small jump at the beginning, it will converge quickly, and the final absolute error is around 2.36%. Compared with the other two algorithms, the anti-interference performance of the present invention has obvious advantages in the case of noise interference, and the accuracy of Soc estimation is also greatly improved.

[0133] In summary, the advantages of the present invention are:

[0134] By using the first-order Thevenin circuit to equate the lithium battery, the static and dynamic characteristics of the lithium battery can be more accurately reflected. By using the deviation-compensated least squares method to perform online identification on the first-order Thevenin circuit, the problem of unbiased identification model caused by data saturation, uncertainty noise and other problems in the conventional least squares method is overcome. By using the joint improved adaptive attenuated extended Kalman filter algorithm to update the second state-space equations, compared with the traditional ampere-hour integration method and the extended Kalman filter algorithm, it has better anti-interference performance, and ultimately greatly improves the accuracy of Soc estimation and anti-interference performance.

[0135] Although the specific implementation modes of the present invention are described above, those skilled in the art should understand that the specific implementation modes described are only illustrative and are not intended to limit the scope of the present invention. Equivalent modifications and changes made by those skilled in the art in accordance with the spirit of the present invention should be included in the scope of protection of the claims of the present invention.

Claims

1. A lithium battery Soc estimation method, characterized in that: The steps include: Step S10, the lithium battery is equivalent to a first-order Thevenin circuit, and the output voltage and output current of the lithium battery are collected in real time; Step S20, constructing a first state space equation group based on the first-order Thevenin circuit, the output voltage and the output current; Step S30, performing online identification on the first-order Thevenin circuit by using a deviation-compensated least squares method to obtain circuit parameters of the first-order Thevenin circuit, and updating the first state-space equation group based on the circuit parameters; Step S40, performing Euler transformation discretization on the first state space equation to obtain a second state space equation group; Step S50, continuously updating the second state space equations by combining an improved adaptive attenuated extended Kalman filter algorithm; Step S60, estimating the Soc of the lithium battery based on the second state-space equation group; In step S50, the calculation process of the joint improved adaptive attenuated extended Kalman filter algorithm is as follows: The EKF state estimation time update, error covariance time update, Kalman gain matrix calculation, error estimation measurement update, state estimation measurement update, and error covariance measurement update equations are as follows: For FEKF, the error covariance time update P k|k-1 =A k-1 P k-1|k-1 A T k-1 +Q k-1 Change to P k|k-1 =A k-1 αP k-1|k- 1A T k-1 +Q k-1 , where α is the forgetting factor; represents the posterior state estimate at time k; Represents the prior state estimate at time k-1; represents the prior state estimate at time k; Q k-1 represents the k-1 moment process noise w before improvement k Variance of K k represents the Kalman gain matrix; e k represents the error estimate at time k; e T k represents the transpose of the error estimate at time k; K T k represents the transpose of the Kalman gain matrix at time k; y k Represents the output value at time k; represents the measurement function; R k-1 represents the measurement noise v at time k-1 before improvement k The variance of Based on FEKF, the process noise w is redefined k ~(q k ,Q k ) and measurement noise ν k ~(r k ,R k ), the two are independent of each other; the noise means at this time are defined as q k and r k , and the state time estimate is updated Change to Error Estimate Update Updated to Re- and R k =(1-d k-1 )R k-1 +d k-1 (e k e T k -H k P k|k-1 H T ) are respectively redefined as Q k ' and R k ': The improved adaptive noise covariance is updated as follows: Among them, q k represents the process noise w at time k k The mean value of Q k ' represents the improved k-time process noise w k The variance of k represents the measurement noise v at time k k The mean value of R k ' represents the improved measurement noise v at time k k The variance of d represents the variable factor; d k represents the variable factor value of d updated at time k; b represents the forgetting factor, and 0<b<1.

2. A lithium battery Soc estimation method as claimed in claim 1, characterized in that: In step S10, the first-order Thevenin circuit includes a resistor R0, a resistor R1, a capacitor C1 and an equivalent battery cell; after the resistor R1 and the capacitor C1 are connected in parallel, one end is connected to the resistor R0; the resistor R0 is connected to the positive electrode of the equivalent battery cell; The output voltage is U L , the output current is I L .

3. A lithium battery Soc estimation method as claimed in claim 2, characterized in that: In step S20, the construction process of the first state space equation group is as follows: According to Kirchhoff's voltage law, we get the following equations: The calculation formula for the state of charge of a lithium battery is as follows: Combining equation (1) and taking the derivative of equation (2), we get the first state space equation group: Among them, U1 represents the polarization voltage across capacitor C1; dU1 represents the differential of U1; dt represents the differential of charge and discharge time; Soc represents the remaining capacity of the equivalent battery cell; Q represents the actual capacity of the lithium battery; Indicates the derivation of U1; It represents the derivative of Soc; Soc(0) represents the initial value of the state of charge of the lithium battery; Soc(t) represents the state of charge of the lithium battery at time t; η represents the charge and discharge efficiency of the lithium battery, which is approximately 1; Uoc(Soc) represents the open circuit voltage value; Uoc represents the voltage of the equivalent battery cell.

4. A lithium battery Soc estimation method as claimed in claim 2, characterized in that: The step S30 specifically includes: Step S31, online identification of the first-order Thevenin circuit is performed by using the deviation-compensated least squares method to obtain identification parameters θ1, θ2, θ3: Wherein, T' represents the sampling time; τ represents the time constant, and τ=R1C1; Step S32: Calculate the circuit parameters including the values ​​of R0, R1 and C1 based on θ1, θ2, θ3 and τ:

5. A lithium battery Soc estimation method as claimed in claim 2, characterized in that: In step S40, the formula of the second state space equation group is: Wherein, Δt represents the sampling time of Euler transformation; k represents the charging and discharging k moment after Euler transformation.

6. A lithium battery Soc estimation method as claimed in claim 5, characterized in that: The step S60 is specifically as follows: Based on the second state-space equation group, the Soc value Soc(k+1) at the k+1 moment is output to estimate the Soc of the lithium battery at the next moment.

Images