Energy storage battery SOC (State of Charge) and SOH (State of Health) joint estimation method based on self-adaptive double-extended Kalman filtering

Abstract

The invention discloses an energy storage battery SOC and SOH joint estimation method based on adaptive double extended Kalman filtering, and the method comprises the steps: constructing a second-order RC equivalent circuit model of an energy storage battery, and building a state-space equation of a battery state and a battery parameter; measuring the open-circuit voltage of the energy storage battery under different SOCs through a pulse discharge experiment, and fitting a polynomial function relationship between the SOC and the open-circuit voltage by using a least square method; establishing a battery capacity model and an internal resistance model for energy storage battery SOH estimation; the measurement noise variance is dynamically adjusted based on an improved Sage-Husa adaptive algorithm, and joint estimation of the SOC and the SOH of the energy storage battery is realized through an iteration mode. According to the method, a relatively large state estimation error caused by fixed measurement noise is avoided, self-adaptive adjustment of the measurement noise is realized, and the reliability and the accuracy of SOC and SOH joint estimation of the energy storage battery are improved.

Description

Technical Field

[0001] This invention relates to the field of battery state estimation technology, and in particular to a joint estimation method for the SOC and SOH of an energy storage battery based on adaptive dual extended Kalman filtering. Background Technology

[0002] Energy storage batteries can be integrated with various new energy distributed generation systems or used independently at the load end to perform functions such as peak shaving and valley filling, peak and frequency regulation, and power quality optimization. They are a crucial component of active distribution networks and microgrids. To ensure the safe and stable operation of energy storage batteries, real-time monitoring of their operating status is paramount. Accurately estimating the state of charge (SOC) and state of health (SOH) of energy storage batteries is fundamental to estimating other battery states. This not only helps prevent overcharging or over-discharging of energy storage batteries but also effectively extends battery life, ensuring the safety and stability of the energy storage system. Therefore, accurately understanding the battery's SOC and SOH is of profound significance for improving the operating efficiency of energy storage batteries and extending their service life.

[0003] Currently, methods for estimating the State of Charge (SOC) of energy storage batteries include empirical estimation methods, ampere-hour integration methods, open-circuit voltage methods, and Extended Kalman Filter (EKF) methods; methods for estimating State of Harm (SOH) include feature-based estimation methods and data-driven estimation methods. However, although existing research has proposed improvement strategies for both SOC and SOH estimation methods based on their respective characteristics, in practical applications, SOC and SOH exhibit a strong coupling relationship and mutual constraints. On the one hand, SOH decay leads to changes in key parameters such as battery capacity and internal resistance, directly affecting the accuracy of the SOC estimation model. If only the traditional EKF algorithm with fixed parameters is used to estimate SOC, the model mismatch error will increase significantly as the battery ages. On the other hand, dynamic fluctuations in SOC affect battery polarization characteristics and terminal voltage response, making it difficult for SOH estimation methods based on static features or offline data to capture the true aging state. For example, under dynamic operating conditions, if the impact of SOC on the battery's internal state is not considered in real time, SOH estimation based on the voltage curve may be biased. Furthermore, the non-stationarity of measurement noise affects both the SOC and SOH estimation processes simultaneously, making it difficult to guarantee the robustness of the joint system through single-dimensional noise adaptation.

[0004] Therefore, methods that optimize SOC or SOH in isolation have inherent limitations. There is an urgent need to construct a joint estimation framework that can update battery SOC and SOH online synchronously. Through a dynamic mutual correction mechanism between parameters and state, the accuracy and reliability of energy storage system state assessment can be fundamentally improved. Summary of the Invention

[0005] The purpose of this invention is to provide a joint estimation method for the SOC and SOH of an energy storage battery based on an adaptive dual extended Kalman filter, which solves the problem of dynamic changes in measurement noise characteristics in the prior art. At the same time, by using the dual extended Kalman filter algorithm to jointly estimate the SOC and SOH of the energy storage battery, the accuracy is improved compared with the single estimation of the SOC and SOH of the energy storage battery.

[0006] The above-mentioned objective of the present invention is achieved through the following technical solution:

[0007] A joint estimation method for SOC and SOH of energy storage batteries based on adaptive dual extended Kalman filtering is implemented according to the following steps:

[0008] (1) Construct a second-order RC equivalent circuit model containing polarization effect, and establish a discrete state-space equation with SOC as the state variable and terminal voltage as the observation quantity.

[0009] (2) The open-circuit voltage of the energy storage battery under different SOC was measured by pulse discharge experiment, and the polynomial function relationship between SOC and open-circuit voltage was fitted by least squares method to obtain the mapping relationship between SOC and open-circuit voltage.

[0010] (3) Establish a battery capacity model and an internal resistance model to estimate the SOH of the energy storage battery, providing a basic model for the joint estimation in step (4). The battery capacity model is used to estimate the SOH of the energy storage battery based on the change in the maximum available capacity of the battery, and the internal resistance model is used to estimate the SOH of the energy storage battery based on the change in the ohmic internal resistance of the battery. The estimation results of the battery capacity model and the internal resistance model are combined by taking the average value to obtain the final estimation result of the SOH of the energy storage battery.

[0011] (4) Based on the state-space equation obtained in step (1), the mapping relationship between SOC and open-circuit voltage obtained in step (2), and the battery capacity model and internal resistance model obtained in step (3), the SOC and SOH of the energy storage battery are jointly estimated using a dual extended Kalman filter. During the estimation process, the improved Sage-Husa adaptive algorithm is used to dynamically adjust the measurement noise variance to improve the filtering accuracy.

[0012] Furthermore, in step (1), the second-order RC equivalent circuit model of the energy storage battery consists of three parts connected in series by the open-circuit voltage source of the battery and then connected to the main circuit. The three parts are the ohmic internal resistance R0, the battery concentration difference polarization RC parallel circuit and the battery electrochemical polarization RC parallel circuit.

[0013] Furthermore, in step (1), a second-order RC equivalent circuit model incorporating polarization effects is constructed, and a discrete state-space equation is established with SOC as the state variable and terminal voltage as the observation, including:

[0014] Based on the constructed second-order RC equivalent circuit model of the energy storage battery, the state-space equations for the operating state and battery parameters of the energy storage battery are constructed as follows:

[0015]

[0016]

[0017] Where k represents the value of each variable in the k-th step, This is called the sampling time, which is the time from k to k+1, and is a constant value. x represents the battery state, θ represents the battery parameters, I represents the battery's operating current, R1 and C1 are the battery's electrochemical polarization resistance and polarization capacitance, respectively, R2 and C2 are the battery's concentration gradient polarization resistance and polarization capacitance, respectively, and q represents the battery's capacity. and Let $\mathbf{k}$ be the battery state at step $k$ and $\mathbf{k}$ be the parameter-independent process noise, and let their covariance matrices be $\mathbf{k}$ respectively. and , is used to represent model error;

[0018] Based on the constructed second-order RC equivalent circuit model of the energy storage battery, the terminal voltage equation of the energy storage battery is constructed as follows:

[0019]

[0020] Among them, U k U is the terminal voltage of the energy storage battery, OCV is the open-circuit voltage of the energy storage battery, and U is the open-circuit voltage of the energy storage battery. 1,k U 2,k These represent the voltages in the battery concentration difference polarization RC parallel circuit and the battery electrochemical polarization RC parallel circuit, respectively. R0 is the ohmic internal resistance of the battery. k This is the measurement noise at step k, and its covariance matrix is ​​R. k , is used to represent measurement error.

[0021] Furthermore, the mapping relationship between the fitted SOC and open-circuit voltage in step (2) is expressed in polynomial form as follows:

[0022]

[0023] The state variables and formulas are as follows:

[0024]

[0025] Where K0-K6 are the polynomial fitting coefficients of the energy storage battery OCV and SOC, and Q is the remaining capacity of the energy storage battery.

[0026] Furthermore, in step (3), the SOH of the energy storage battery is estimated based on the battery capacity model and the internal resistance model. The expression for estimating the SOH of the energy storage battery using the battery capacity model is shown below:

[0027]

[0028] Among them, SOH q Let SOH,q be the energy storage battery estimated using a battery capacity model. EOL q represents the capacity value at the end of the energy storage battery's lifespan. BOL Let be the capacity value of the new battery; the expression for estimating the SOH of the energy storage battery using the internal resistance model is shown below:

[0029]

[0030] Among them, SOH R For the energy storage battery SOH estimated by the battery capacity model, R 0END R is the ohmic internal resistance at the end of the energy storage battery's life. 0NEW The ohmic internal resistance of the new battery;

[0031] The two methods described above are effectively combined by averaging to estimate the SOH of the battery, as shown in the following expression:

[0032] .

[0033] Furthermore, step 4 specifically includes:

[0034] Step 4.1: First, initialize the state and model parameters, as shown in the following equation:

[0035]

[0036] Step 4.2: Determine whether the parameter state update step size has been reached. If the parameter state update step size has been reached, proceed to step 4.3; otherwise, jump to step 4.5.

[0037] Step 4.3: The sampling time of the parameter filter is t, the step size of the parameter filter is i∈{1, 2, ..., ∞}, and the time update equation of the parameter filter is:

[0038]

[0039] Step 4.4: Measurement update of the parameter filter estimates the SOH of the energy storage battery. The measurement update equation is as follows:

[0040]

[0041] Step 4.5: The sampling time of the state filter is T, the step size of the state filter is k∈{1, 2, ..., ∞}, and the time update equation of the state filter is:

[0042]

[0043] Step 4.6: The measurement update of the state filter realizes the posterior estimation of the SOC of the energy storage battery. The measurement update equation is:

[0044]

[0045] in:

[0046]

[0047]

[0048]

[0049] in, , , , The initial value for the algorithm. This was obtained from experimental data on energy storage batteries. This is the initial state of the battery. and These represent the initial system parameter covariance matrices for the parametric filter and the state filter, respectively; K is the Kalman gain, Y is the measured terminal voltage of the energy storage battery, the superscript "-" indicates the prior value of the variable in the algorithm; the top-level label "^" indicates the estimated value of the variable in the algorithm;

[0050] Step 4.7: Dynamically adjust the measurement noise variance using the improved Sage-Husa adaptive algorithm, and then jump to step 4.2.

[0051] Furthermore, step 4.7 improves the Sage-Husa adaptive algorithm by dynamically adjusting the measurement noise variance, as shown in the following expression:

[0052]

[0053] Among them, R min For allowed R k The minimum value of R max For allowed R k The maximum value, α is an adjustment coefficient used to control the update speed, e kThis relates to the terminal voltage prediction error. The beneficial effects of this invention are as follows: Compared to a single SOC and SOH estimation system, the adaptive dual extended Kalman filter joint estimation framework constructed in this invention exhibits significant advantages in several aspects. First, through real-time interactive correction of SOC and SOH, the long-term cumulative error problem caused by the solidification of model parameters in traditional single estimation is solved, thereby improving the accuracy of SOC and SOH estimation for energy storage batteries. Second, this invention fuses the SOH estimation results of the battery capacity model and internal resistance model using the mean method, effectively reducing the sensitivity bias of a single model to specific operating conditions or aging stages, and improving the robustness and comprehensiveness of health status assessment. Finally, this method uses an improved Sage-Husa algorithm to adaptively adjust the measurement noise variance, enabling the system to better adapt to noise interference under different operating conditions, thus providing more accurate and reliable SOC and SOH estimation results in practical application scenarios with high uncertainty. Attached Figure Description

[0054] Figure 1 This is the second-order RC equivalent circuit model of the energy storage battery constructed in this invention.

[0055] Figure 2 The flowchart shows the algorithm for estimating the SOC and SOH of an energy storage battery using the adaptive dual extended Kalman filter algorithm proposed in this invention.

[0056] Figure 3 This is a topology diagram of the test system of the present invention.

[0057] Figure 4 This is a graph showing the SOC estimation results of the present invention.

[0058] Figure 5 The graph shows the SOH estimation results of this invention. Detailed Implementation

[0059] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0060] The present invention proposes a joint estimation method for the State of Charge (SOC) and State of Harm (SOH) of an energy storage battery based on adaptive dual extended Kalman filtering, which is implemented according to the following steps:

[0061] Step 1: Construct a second-order RC equivalent circuit model of the energy storage battery and establish discrete state-space equations.

[0062] Numerous studies have demonstrated the excellent performance of equivalent circuit models in estimating State of Charge (SOC) and State of Harshness (SOH). These models accurately describe the physical and chemical changes within the battery while maintaining low computational cost, thus meeting the accuracy and efficiency requirements of energy storage batteries. The established second-order RC equivalent circuit model for the energy storage battery is shown below. Figure 1 As shown. In Figure 1 In this context, I represents the battery's operating current, with its positive direction being... Figure 1 In the diagram, R0 is the ohmic internal resistance of the battery, R1 and C1 are the electrochemical polarization resistance and polarization capacitance of the battery, respectively, U1 is the voltage of the parallel section, R2 and C2 are the concentration difference polarization resistance and polarization capacitance of the battery, respectively, U2 is the voltage of the parallel section, U is the terminal voltage of the energy storage battery, and OCV is the open circuit voltage of the energy storage battery.

[0063] based on Figure 1 The state-space model of the energy storage battery is constructed as follows:

[0064] (1)

[0065] (2)

[0066] Where q is the capacity of the energy storage battery.

[0067] The State of Charge (SOC) of an energy storage battery refers to the ratio of the currently stored electrical energy in the battery or energy storage system to its maximum capacity. Since the Open Value Capacity (OCV) of an energy storage battery is a function of SOC, this method uses a polynomial function to fit the relationship between OCV and SOC, as shown in the following equation:

[0068] (3)

[0069] Among them, K i (i = 0, 1, 2, 3, 4, 5, 6) are the polynomial fitting coefficients.

[0070] Equation (1) is discretized using a difference method. The discretization criterion and the state-space equation of the established dual extended Kalman filter are shown below:

[0071] (4)

[0072] (5)

[0073] (6)

[0074] (7)

[0075] (8)

[0076] Where k represents the value of each variable in the k-th step, This is called the sampling time, which is the time from k to k+1, and is a constant value. x represents the battery state, and θ represents the battery parameters. and Let $\mathbf{k}$ be the battery state at step $k$ and $\mathbf{k}$ be the parameter-independent process noise, and let their covariance matrices be $\mathbf{k}$ respectively. and U is used to represent the model error. k U is the terminal voltage of the energy storage battery, OCV is the open-circuit voltage of the energy storage battery, and U is the open-circuit voltage of the energy storage battery. 1,k U 2,k The voltages, v, are the voltages of the battery concentration difference polarization RC parallel circuit and the battery electrochemical polarization RC parallel circuit, respectively. k This is the measurement noise at step k, and its covariance matrix is ​​R. k , is used to represent the measurement error, and Q is the remaining power of the energy storage battery.

[0077] Step 2: Measure the open-circuit voltage of the energy storage battery under different SOCs through pulse discharge experiments, and use the least squares method to fit the polynomial function relationship between SOC and open-circuit voltage to obtain the mapping relationship between SOC and open-circuit voltage OCV.

[0078] Step 3: Establish battery capacity model and internal resistance model to estimate the SOH of the energy storage battery. The battery capacity model is based on the change of the battery's maximum usable capacity, and the internal resistance model is based on the change of the battery's ohmic internal resistance. The estimation results of the two models are combined by taking the average to obtain the final estimate of the SOH of the energy storage battery.

[0079] State of Health (SOH) of an energy storage battery is an indicator of the battery's health. It reflects the battery's overall health and remaining lifespan, and is typically determined by comparing it with the battery's initial performance (such as capacity and internal resistance). Therefore, a battery capacity model and an internal resistance model are established to estimate the SOH of an energy storage battery. The battery capacity model estimates the SOH based on the change in the battery's maximum usable capacity, while the internal resistance model estimates the SOH based on the change in the battery's ohmic internal resistance.

[0080] The expression for estimating the SOH of an energy storage battery using a battery capacity model is shown below:

[0081] (9)

[0082] Among them, SOH q Let SOH,q be the energy storage battery estimated using a battery capacity model. EOL q represents the capacity value at the end of the energy storage battery's lifespan. BOL This is the capacity value of the new battery.

[0083] The expression for estimating the SOH of an energy storage battery using the internal resistance model is shown below:

[0084] (10)

[0085] Among them, SOHR For the energy storage battery SOH estimated by the battery capacity model, R 0END R is the ohmic internal resistance at the end of the energy storage battery's life. 0NEW The ohmic internal resistance of the new battery.

[0086] The two methods described above are effectively combined by averaging to estimate the SOH of the battery, as shown in the following expression:

[0087] (11)

[0088] Step 4: Use dual extended Kalman filter (DEKF) to jointly estimate the SOC and SOH of the energy storage battery. During the estimation process, the measurement noise variance is dynamically adjusted based on the improved Sage-Husa adaptive algorithm to improve the filtering accuracy.

[0089] This invention uses DEKF to estimate battery state and parameters. Two types of EKF are used as parameter filters and state filters, respectively.

[0090] Step 4.1: First, initialize the state and model parameters, as shown in the following equation:

[0091] (12)

[0092] Step 4.2: Determine whether the parameter state update step size has been reached. If the parameter state update step size has been reached, proceed to step 4.3; otherwise, jump to step 4.5.

[0093] Step 4.3: The sampling time of the parameter filter is t, the step size of the parameter filter is i∈{1, 2, ..., ∞}, and the time update equation of the parameter filter is:

[0094] (13)

[0095] Step 4.4: The measurement update of the parameter filter enables the estimation of the SOH of the energy storage battery. The measurement update equation is as follows:

[0096] (14)

[0097] Step 4.5: The sampling time of the state filter is T, the step size of the state filter is k∈{1, 2, ..., ∞}, and the time update equation of the state filter is:

[0098] (15)

[0099] Step 4.6: The measurement update of the state filter enables a posterior estimation of the SOC of the energy storage battery. The measurement update equation is as follows:

[0100] (16)

[0101] in:

[0102] (17)

[0103] (18)

[0104] (19)

[0105] In equations (12)-(19), , , , The initial value for the algorithm. This can be obtained from experimental data on energy storage batteries. This is the initial state of the battery. and These represent the initial system parameter covariance matrices for the parametric filter and the state filter, respectively. K is the Kalman gain, Y is the measured terminal voltage of the energy storage battery, and the superscript "-" indicates the prior value of the variable in the algorithm. The top-level label "^" indicates the estimated value of the variable in the algorithm.

[0106] Step 4.7: Dynamically adjust the measurement noise variance using the Sage-Husa adaptive algorithm, and simultaneously jump to step 4.2. To limit R... k To address the over-updating issue, the Sage-Husa adaptive algorithm is improved, as shown in the following expression:

[0107] (20)

[0108] (twenty one)

[0109] Among them, R min For allowed R k The minimum value of R max For allowed R k The maximum value, α is an adjustment coefficient used to control the update speed, e k This represents the terminal voltage prediction error.

[0110] The recursive process of the entire algorithm is as follows: Figure 2 As shown.

[0111] To verify the accuracy of the joint estimation of SOC and SOH in this invention, Figure 3The energy storage system using a lithium nickel cobalt manganese oxide battery as an example is used for verification. The battery energy storage system consists of three battery packs connected in series and parallel. Each battery pack consists of four individual cells connected in series. Each individual cell is analyzed using a second-order RC equivalent circuit model. The individual cells are in a constant current discharge state during operation. The joint online estimation of SOC and SOH is performed only on cell number 1. The estimated results of SOC and SOH for the energy storage battery are as follows: Figure 4 and Figure 5 As shown, the adaptive double extended Kalman filter algorithm has higher accuracy in estimating SOC, and can also achieve accurate estimation of SOH.

[0112] The features of this invention are:

[0113] In step (1), the second-order RC equivalent circuit model of the energy storage battery consists of three parts connected in series from the battery's open-circuit voltage source and then connected to the main circuit. The three parts are the ohmic internal resistance R0, the battery concentration difference polarization RC parallel circuit, and the battery electrochemical polarization RC parallel circuit. The state-space equation corresponding to this equivalent circuit is constructed by differentiating the derivative forms to establish the discrete state-space equation.

[0114] In step (2), the open-circuit voltage of the energy storage battery under different SOCs is measured by pulse discharge experiment, and it is assumed that the relationship between the energy storage battery OCV and SOC can be fitted by a polynomial function relationship. The polynomial coefficients are calculated by the least squares method.

[0115] The energy storage battery SOH estimation model described in step (3) is based on the energy storage battery SOH estimated by the battery capacity model and the internal resistance model. The average value of the SOH estimation results of the two models is taken as the final estimation result of the energy storage SOH, thereby realizing the effective combination of the two estimation methods.

[0116] In step (4), two extended Kalman filters with different time scales are constructed to estimate the operating state and parameters of the energy storage battery. Specifically, the SOC estimation is included in the energy storage battery operating state estimation, and the SOH estimation is included in the energy storage battery operating parameter estimation. During the estimation process, an improved Sage-Husa adaptive algorithm is used to dynamically adjust the measurement noise variance while avoiding over-adjustment of the variance, thus improving the filtering accuracy.

[0117] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A joint estimation method for SOC and SOH of an energy storage battery based on adaptive dual extended Kalman filtering, characterized in that, Includes the following steps: (1) Construct a second-order RC equivalent circuit model containing polarization effect, and establish a discrete state-space equation with SOC as the state variable and terminal voltage as the observation quantity. (2) The open-circuit voltage of the energy storage battery under different SOC was measured by pulse discharge experiment, and the polynomial function relationship between SOC and open-circuit voltage was fitted by least squares method to obtain the mapping relationship between SOC and open-circuit voltage. (3) Establish a battery capacity model and an internal resistance model to estimate the SOH of the energy storage battery, providing a basic model for the joint estimation in step (4). The battery capacity model is used to estimate the SOH of the energy storage battery based on the change in the maximum available capacity of the battery, and the internal resistance model is used to estimate the SOH of the energy storage battery based on the change in the ohmic internal resistance of the battery. The estimation results of the battery capacity model and the internal resistance model are combined by taking the average value to obtain the final estimation result of the SOH of the energy storage battery. (4) Based on the state-space equation obtained in step (1), the mapping relationship between SOC and open-circuit voltage obtained in step (2), and the battery capacity model and internal resistance model obtained in step (3), the SOC and SOH of the energy storage battery are jointly estimated using a dual extended Kalman filter. During the estimation process, the improved Sage-Husa adaptive algorithm is used to dynamically adjust the measurement noise variance to improve the filtering accuracy.

2. The method for joint estimation of SOC and SOH of energy storage battery based on adaptive dual extended Kalman filtering according to claim 1, characterized in that: In step (1), the second-order RC equivalent circuit model of the energy storage battery consists of three parts connected in series by the open-circuit voltage source of the battery and then connected to the main circuit. The three parts are the ohmic internal resistance R0, the battery concentration difference polarization RC parallel circuit and the battery electrochemical polarization RC parallel circuit.

3. The method for joint estimation of SOC and SOH of energy storage battery based on adaptive dual extended Kalman filtering according to claim 1, characterized in that: In step (1), a second-order RC equivalent circuit model incorporating polarization effects is constructed, and a discrete state-space equation is established with SOC as the state variable and terminal voltage as the observation, including: Based on the constructed second-order RC equivalent circuit model of the energy storage battery, the state-space equations for the operating state and battery parameters of the energy storage battery are constructed as follows: ； ； Where k represents the value of each variable in the k-th step, This is called the sampling time, which is the time from k to k+1, and is a constant value. x represents the battery state, θ represents the battery parameters, I represents the battery's operating current, R1 and C1 are the battery's electrochemical polarization resistance and polarization capacitance, respectively, R2 and C2 are the battery's concentration gradient polarization resistance and polarization capacitance, respectively, and q represents the battery's capacity. and Let $\mathbf{k}$ be the battery state at step $k$ and $\mathbf{k}$ be the parameter-independent process noise, and let their covariance matrices be $\mathbf{k}$ respectively. and , is used to represent model error; Based on the constructed second-order RC equivalent circuit model of the energy storage battery, the terminal voltage equation of the energy storage battery is constructed as follows: ； Among them, U k U is the terminal voltage of the energy storage battery, OCV is the open-circuit voltage of the energy storage battery, and U is the open-circuit voltage of the energy storage battery. 1,k U 2,k These represent the voltages in the battery concentration difference polarization RC parallel circuit and the battery electrochemical polarization RC parallel circuit, respectively. R0 is the ohmic internal resistance of the battery. k This is the measurement noise at step k, and its covariance matrix is ​​R. k , is used to represent measurement error.

4. The method for joint estimation of SOC and SOH of energy storage battery based on adaptive dual extended Kalman filtering according to claim 3, characterized in that: In step (2), the mapping relationship between the fitted SOC and the open-circuit voltage is expressed in polynomial form as follows: ； The state variables and formulas are as follows: ； Where K0-K6 are the polynomial fitting coefficients of the energy storage battery OCV and SOC, and Q is the remaining capacity of the energy storage battery.

5. The method for joint estimation of SOC and SOH of energy storage battery based on adaptive dual extended Kalman filtering according to claim 1, characterized in that: In step (3), the SOH of the energy storage battery is estimated based on the battery capacity model and the internal resistance model. The expression for estimating the SOH of the energy storage battery using the battery capacity model is shown below: ； Among them, SOH q Let SOH,q be the energy storage battery estimated using a battery capacity model. EOL q represents the capacity value at the end of the energy storage battery's lifespan. BOL Let be the capacity value of the new battery; the expression for estimating the SOH of the energy storage battery using the internal resistance model is shown below: ； Among them, SOH R For the energy storage battery SOH estimated by the battery capacity model, R 0END R is the ohmic internal resistance at the end of the energy storage battery's life. 0NEW The ohmic internal resistance of the new battery; The two methods described above are effectively combined by averaging to estimate the SOH of the battery, as shown in the following expression: 。 6. The method for joint estimation of SOC and SOH of energy storage battery based on adaptive dual extended Kalman filtering according to claim 4, characterized in that: Step 4 specifically includes: Step 4.1: First, initialize the state and model parameters, as shown in the following equation: ； Step 4.2: Determine whether the parameter state update step size has been reached. If the parameter state update step size has been reached, proceed to step 4.3; otherwise, jump to step 4.

5. Step 4.3: The sampling time of the parameter filter is t, the step size of the parameter filter is i∈{1, 2, ..., ∞}, and the time update equation of the parameter filter is: ； Step 4.4: Measurement update of the parameter filter estimates the SOH of the energy storage battery. The measurement update equation is as follows: ； Step 4.5: The sampling time of the state filter is T, the step size of the state filter is k∈{1, 2, ..., ∞}, and the time update equation of the state filter is: ； Step 4.6: The measurement update of the state filter realizes the posterior estimation of the SOC of the energy storage battery. The measurement update equation is: ； in: ； ； ； in, , , , The initial value for the algorithm. This was obtained from experimental data on energy storage batteries. This is the initial state of the battery. and These represent the initial system parameter covariance matrices for the parametric filter and the state filter, respectively; K is the Kalman gain, Y is the measured terminal voltage of the energy storage battery, the superscript "-" indicates the prior value of the variable in the algorithm; the top-level label "^" indicates the estimated value of the variable in the algorithm; Step 4.7: Dynamically adjust the measurement noise variance using the improved Sage-Husa adaptive algorithm, and then jump to step 4.

2.

7. The method for joint estimation of SOC and SOH of energy storage battery based on adaptive dual extended Kalman filtering according to claim 6, characterized in that: Step 4.7 improves the Sage-Husa adaptive algorithm by dynamically adjusting the measurement noise variance, as shown in the following expression: ； Among them, R min For allowed R k The minimum value of R max For allowed R k The maximum value, α is an adjustment coefficient used to control the update speed, e k This represents the terminal voltage prediction error.

Images