Joint estimation method and system for SOC and SOH of all cells in energy storage system

Abstract

A method and system for jointly estimating the SOC and SOH of all cells in an energy storage system includes: measuring the terminal voltage of each battery, using the cell with the largest terminal voltage and the cell with the smallest terminal voltage as characteristic cells; establishing a circuit model for the characteristic cell; jointly estimating the SOC and SOH of the characteristic cell based on the circuit model of the characteristic cell; fusion-estimating the SOC and SOH of each cell based on the circuit model of each cell; and using the SOC and SOH estimates of the characteristic cell obtained by the joint estimation to correct the SOC and SOH estimates of each cell obtained by the fusion estimation, as the result of the joint estimation of the SOC and SOH of all cells in the energy storage system. The present invention achieves accurate evaluation of the SOC and SOH of each cell, which serves as a key indicator for implementing active balancing technology.

Description

Technical Field

[0001] The present invention belongs to the technical field of state estimation of large energy storage batteries, and specifically relates to a method and system for jointly estimating the SOC and SOH of all battery cells in an energy storage system. Background Art

[0002] With the rapid development of renewable energy, energy storage systems are playing an increasingly important role in balancing power supply and demand, improving grid reliability, and optimizing energy utilization. In energy storage systems, battery state assessment, particularly accurate estimation of state of charge (SOC) and state of health (SOH), is critical to ensuring efficient and stable system operation. However, existing SOC and SOH estimation methods still have some shortcomings.

[0003] In the existing technology, most SOC and SOH estimation methods are mainly calculated for characteristic cells, focusing only on the status of a single battery and ignoring the overall characteristics of the system. The limitation of this method is that the performance status of a single battery may vary significantly due to manufacturing differences, usage environment and working conditions, resulting in a large deviation between its actual performance and the estimated value. Especially in large-scale energy storage systems, due to the parallel and series characteristics of the battery pack, the state change of a single battery will directly affect the performance of the entire system. Therefore, the method of estimating SOC and SOH for characteristic cells alone fails to capture the dynamic changes of the entire system, which often leads to the inability to reflect the actual operating status of the energy storage system in real time and accurately.

[0004] With the development of new power systems, energy storage battery systems with large energy storage capacity and power output capabilities, which are integrated from a large number of energy storage battery cells, are becoming increasingly widespread. Large-scale energy storage battery systems with capacities exceeding megawatt-hours and charge and discharge powers exceeding megawatts primarily address the problem of jointly estimating the SOC and SOH of series-connected battery clusters. Large-scale energy storage battery systems have a large number of cells connected in series. Traditional methods of jointly estimating the SOC and SOH of each cell would consume significant hardware and computing resources. To effectively apply active balancing technology, a method for jointly estimating the SOC and SOH of all cells in the entire energy storage system is crucial. Summary of the Invention

[0005] To address the deficiencies in the prior art, the present invention provides a method and system for jointly estimating the SOC and SOH of all battery cells in an energy storage system, which accurately evaluates the SOC and SOH of each battery cell as key indicators for implementing active balancing technology. By comprehensively evaluating the status of all battery cells, the differences between individual battery cells are identified, thereby optimizing the energy distribution of the battery pack and ensuring the balanced operation of each battery unit during the charging and discharging process. This not only improves the overall utilization efficiency of the battery, but also extends the battery life and reduces the safety hazards caused by individual cell imbalance.

[0006] The present invention adopts the following technical solutions.

[0007] The present invention proposes a method for jointly estimating the SOC and SOH of all cells in an energy storage system, comprising:

[0008] Measure the terminal voltage of each battery, and use the cell with the largest terminal voltage and the cell with the smallest terminal voltage as characteristic cells; establish a circuit model for the characteristic cells;

[0009] Based on the circuit model of the characteristic cell, the SOC and SOH of the characteristic cell are jointly estimated;

[0010] Based on the circuit model of each battery cell, the SOC and SOH of each battery cell are estimated in a fusion manner;

[0011] The SOC estimated value and SOH estimated value of the characteristic battery cell obtained by the joint estimation are used to correct the SOC estimated value and SOH estimated value of each battery cell obtained by the fusion estimation, and the result is the joint estimation result of the SOC and SOH of all battery cells in the energy storage system.

[0012] The circuit model of the characteristic battery cell is a second-order RC circuit model, which includes: the first RC link and the second RC link.

[0013] A hybrid Kalman filter is constructed using an adaptive extended Kalman filter and an adaptive unscented Kalman filter. Based on the circuit model of the characteristic cell, the SOC and SOH of the characteristic cell are jointly estimated, including:

[0014] Based on the circuit model of the characteristic cell, the adaptive extended Kalman filter is used to estimate the SOC of the characteristic cell. The estimated SOC value of the characteristic cell at the estimation moment is the first input in the SOH estimation of the characteristic cell based on the adaptive unscented Kalman filter. Estimating the observed values of the measurement equation at the moment of measurement;

[0015] The SOH of the characteristic cell is estimated based on the adaptive unscented Kalman filter. The posterior estimated value of the characteristic cell capacity at the estimation time is the first input in the SOC estimation of the characteristic cell based on the adaptive extended Kalman filter. The input values of the state-space equations at the estimation time.

[0016] Based on the circuit model of the characteristic cell, the adaptive extended Kalman filter is used to estimate the SOC of the characteristic cell, including:

[0017] Based on the second-order RC circuit model of the characteristic battery cell, the state space equation and measurement equation for SOC estimation are constructed;

[0018] After initialization, the estimated values of the measurement noise covariance matrix and the estimated values of the noise covariance are improved;

[0019] A priori estimation of the state vector and error covariance is performed based on the estimated value of the improved noise covariance;

[0020] Determine the Kalman gain matrix based on the estimated value of the improved measurement noise covariance matrix, and use the Kalman gain matrix to perform a posteriori estimation of the state vector and error covariance based on the state space equation and measurement equation for SOC estimation;

[0021] From The state vector estimation at the estimation time is obtained Estimation time characteristic battery cell Estimation results.

[0022] The state space equation and measurement equation for SOC estimation are as follows:

[0023]

[0024] Where, For the The state vector at the estimation time, For the System inputs for estimation time, is the process noise of the state space equation, For the The terminal voltage measurement value of the battery cell at the estimation time, is the measurement noise of the measurement equation, For the The state matrix at the estimation time, For the The input matrix at the estimation moment, For the The observation matrix at the estimation moment, For the Direct transfer matrix at the estimation moment;

[0025] Among them, the characteristic battery is used in the Estimated time value , the voltage of the first RC link , the voltage of the second RC link Construction The state vector at the estimation time satisfies , with the first Estimated current of characteristic cells at the time of evaluation As system input, satisfy .

[0026] The matrices in the state space equation and the measurement equation satisfy the following relationship:

[0027]

[0028]

[0029]

[0030]

[0031] Where, is the equation iteration time step, = , and are the resistance and capacitance of the first RC link, = , and are the resistance and capacitance of the second RC link respectively, For the Estimated capacity value of the characteristic cell at the estimation time, is the charge and discharge efficiency of the characteristic battery cell, For the Estimation of the open circuit voltage of the characteristic cell at the time of evaluation, is the internal resistance of the characteristic cell, and is the polarization internal resistance of the characteristic cell.

[0032] The measurement noise covariance matrix The estimated value and noise covariance of The estimated value of is improved to satisfy the following relationship:

[0033]

[0034] Where, is the forgetting factor, and its value range is , 、 All are Estimation of the intermediate parameters constructed at the moment, For the The estimated value of the measurement noise covariance matrix at the estimation moment, For the The estimated value of the noise covariance at the estimation time, For the The Kalman gain matrix at the estimation moment, For the The estimated value of the state vector at the estimation time.

[0035] The SOH estimation of characteristic cells is performed based on the adaptive unscented Kalman filter, including:

[0036] Based on the ampere-hour integration algorithm, using the Estimation time characteristic battery cell Estimation results, establish state space equations and measurement equations for SOH estimation;

[0037] After initialization, the capacity of the characteristic cell is estimated a priori, the error covariance of the output estimate is calculated, and the error covariance of the capacity a priori estimate of the characteristic cell and the output estimate is calculated. Estimated time and Estimated time The estimated value of the difference between the estimated results is used as the output estimate;

[0038] The Kalman gain matrix is determined using the error covariance of the output estimation and the prior estimation of the characteristic cell capacity and the error covariance of the output estimation. The Kalman gain matrix is then used to perform a posteriori estimation of the characteristic cell capacity and covariance based on the state space equation and measurement equation for SOH estimation.

[0039] Get the first After estimating the a posteriori estimate of the capacity of the characteristic cell at the time of estimation, the SOH estimate is calculated using the following relationship:

[0040]

[0041] Where, For the Estimation time characteristic battery cell Estimation results, For the The posterior estimate of the capacity of the characteristic cell at the estimation time, is the rated capacity of the characteristic battery cell.

[0042] The state space equation and measurement equation for SOH estimation are as follows:

[0043]

[0044] Where, For the Estimating the capacity of the characteristic battery cell at the time of evaluation, is the system noise, For the Estimated time and Estimated time The difference between the estimated results, For the Estimation time characteristic battery cell Estimation results, is the charge and discharge efficiency of the characteristic battery cell, For the Estimating the current of the characteristic cell at the moment of evaluation, is the equation iteration time step, To measure noise.

[0045] Based on the circuit model of each cell, the SOC and SOH of each cell are estimated, including:

[0046] Identify the internal resistance of each battery cell based on the cell circuit model;

[0047] The internal resistance of each battery cell obtained by identification is used to correct the SOC of each battery cell calculated by combining the ampere-hour integration and the open circuit voltage.

[0048] The SOC difference of each battery cell is used to estimate the SOH of each battery cell based on the open circuit voltage characteristics of the battery cell.

[0049] The estimated capacity value determined based on the estimated SOH value of each battery cell is used to obtain the estimated SOC value of each battery cell using the ampere-hour integration method.

[0050] The calculation formula for the estimated value of SOH of each battery cell is as follows:

[0051]

[0052]

[0053] Where, For the Estimated time Estimated capacity of each battery cell, For the Estimated time The accumulated charge capacity of each battery cell, For the Estimated time The SOC difference of each battery cell, For the Estimated time Estimated SOH value of each battery cell, is the rated capacity of the battery cell;

[0054]

[0055]

[0056] Where, is the first Estimated time The SOC of each cell is obtained by identifying the Internal resistance of a cell , calculate the The open circuit voltage value of each battery cell is obtained, and the SOC value corresponding to the open circuit voltage is obtained by the open circuit voltage lookup table method. .

[0057] The estimated SOC value of each battery cell satisfies the following relationship:

[0058]

[0059] Where, For the Estimated time The estimated SOC value of each battery cell, is the charge and discharge efficiency of the battery cell, For the Estimated current of all cells at the moment, For the Estimated time Estimated capacity of each battery cell, is the time step for the calculation iteration.

[0060] Taking the SOC estimated value of the characteristic cell obtained by the joint estimation as a constraint, the SOC estimated value of each cell obtained by the fusion estimation is corrected using the following relationship:

[0061]

[0062] Where, For the Estimated time Correction value of the estimated SOC value of each battery cell, For the The estimated SOC value of the characteristic cell corresponding to the maximum terminal voltage at the estimation moment, For the The estimated SOC value of the characteristic cell corresponding to the minimum terminal voltage at the estimation moment, For the Estimated time Estimated SOC value of each battery cell.

[0063] Taking the SOH estimated value of the characteristic cell obtained by the joint estimation as a constraint, the SOH estimated value of each cell obtained by the fusion estimation is corrected by the following relationship:

[0064] when hour,

[0065]

[0066] when hour:

[0067]

[0068] Where, For the Estimated time Correction value of the estimated SOH value of each battery cell, For the The minimum value of the estimated SOH value of each battery cell at the estimation time, For the The maximum value of the estimated SOH value of each cell at the estimation time, For the Estimated time Estimated SOH value of each battery cell, For the The estimated SOH value of the characteristic cell corresponding to the maximum terminal voltage at the estimation moment, For the The estimated SOH value of the characteristic cell corresponding to the minimum terminal voltage at the estimation moment.

[0069] The present invention also proposes a system for jointly estimating the SOC and SOH of all cells in an energy storage system, comprising:

[0070] The SOC and SOH joint estimation module of the characteristic battery cell is used to measure the terminal voltage of each battery, and use the battery cell with the largest terminal voltage and the battery cell with the smallest terminal voltage as the characteristic battery cell; establish a circuit model of the characteristic battery cell; and based on the circuit model of the characteristic battery cell, jointly estimate the SOC and SOH of the characteristic battery cell;

[0071] The SOC and SOH fusion estimation module of each battery cell is used to fusion estimate the SOC and SOH of each battery cell based on the circuit model of each battery cell;

[0072] The SOC and SOH joint estimation module for all battery cells is used to correct the SOC and SOH estimated values of each battery cell obtained by fusion estimation based on the SOC estimated values and SOH estimated values of the characteristic battery cells obtained by the joint estimation, and use them as the SOC and SOH joint estimation results of all battery cells in the energy storage system.

[0073] The present invention also provides a terminal, comprising a processor and a storage medium; the storage medium is used to store instructions; and the processor is used to operate according to the instructions to execute steps of the method.

[0074] The present invention also relates to a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method when the program is executed by a processor.

[0075] The present invention offers at least one beneficial effect compared to existing technologies: its method ensures accurate system evaluation and balancing under dynamic operating conditions, improving the management of energy storage systems and enhancing their safety and reliability. By comprehensively evaluating the battery pack's SOC and SOH, the algorithm achieves effective proactive balancing, optimizes charge and discharge strategies, and reduces operational risks, thereby promoting the wider application of energy storage systems in renewable energy utilization and promoting the sustainable development of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 This is a flow chart of a method for jointly estimating the SOC and SOH of all cells in an energy storage system proposed by the present invention;

[0077] Figure 2 is a second-order RC circuit model of a characteristic cell in an embodiment of the present invention;

[0078] Figure 3 Schematic diagram of joint estimation of SOC and SOH of a characteristic battery cell in an embodiment of the present invention;

[0079] Figure 4 It is a simplified circuit model of each battery cell in the embodiment of the present invention;

[0080] Figure 5 Schematic diagram of SOC estimation of each battery cell in an embodiment of the present invention;

[0081] Figure 6 Schematic diagram of SOH estimation of each battery cell in an embodiment of the present invention. DETAILED DESCRIPTION

[0082] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. The embodiments described in this application are only part of the embodiments of the present invention, not all of them. Based on the spirit of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0083] Aiming at the engineering application problem of joint estimation of SOC and SOH of all cells in large-scale energy storage battery systems, the present invention proposes a joint estimation method of SOC and SOH of all cells in energy storage systems that can comprehensively estimate the accuracy and save computing resources, such as Figure 1 Shown, including:

[0084] Step 1: Measure the terminal voltage of each battery cell, and use the cell with the largest terminal voltage and the cell with the smallest terminal voltage as characteristic cells; and establish a circuit model for the characteristic cells.

[0085] The present invention first selects and constructs a high-precision circuit model of a characteristic cell, and then applies a high-precision algorithm to realize the joint estimation of the SOC and SOH of the characteristic cell.

[0086] Specifically, step 1 includes:

[0087] Step 1.1, measuring the terminal voltage of each battery cell, and taking the cell with the largest terminal voltage and the cell with the smallest terminal voltage as the characteristic cell;

[0088] In the embodiment, selecting the monomers (battery cells) with the largest and smallest terminal voltages in the energy storage system as the characteristic cells is a non-restrictive and preferred choice. Those skilled in the art can select monomers with different parameters such as capacity and internal resistance as characteristic cells.

[0089] Step 1.2, establish a second-order RC circuit model of the characteristic cell;

[0090] Figure 2 is a second-order RC circuit model of the characteristic cell in the embodiment of the present invention; the characteristic cell high-precision circuit model used in the present invention is a second-order RC circuit model that facilitates online identification and solution of model parameters; wherein, is the open circuit voltage of the characteristic cell circuit model, is the equivalent internal resistance of the characteristic cell circuit model, and corresponds to the ohmic polarization of the battery. The two RC links correspond to the hysteresis effect of electrochemical polarization. The parameters of the first RC link are and , the parameters of the second RC link are and , is the terminal voltage of the battery cell, is the current of the battery cell, the charging direction is positive and the discharging direction is negative;

[0091] Step 1.3, based on the second-order RC circuit model of the characteristic cell, determine the relationship between the terminal voltage and current of the battery cell;

[0092] The complex frequency domain mathematical relationship between terminal voltage and current obtained from Kirchhoff's voltage law is as follows:

[0093]

[0094] Where, is the terminal voltage of the battery cell in the complex frequency domain, is the open circuit voltage of the battery cell in the complex frequency domain, is the current of the battery cell in the complex frequency domain, is the equivalent internal resistance of the battery cell, and are the resistance and capacitance of the first RC link, and are the resistance and capacitance of the second RC link respectively, For the operator.

[0095] The high-precision second-order RC circuit model for batteries includes key parameters such as the battery's open-circuit voltage, ohmic internal resistance, and second-order RC circuit. By performing detailed electrochemical characterization tests on characteristic cells and fitting exponential polynomial data, a dynamic model that conforms to actual operating conditions is established. This model can reflect changes in the battery state in real time and provide accurate data support for subsequent combined SOC and SOH estimation.

[0096] Step 2: Based on the circuit model of the characteristic cell, jointly estimate the SOC and SOH of the characteristic cell;

[0097] Due to the coupling effect between SOC and SOH, estimating SOC or SOH alone cannot guarantee the accurate and true state value of the battery. Therefore, the present invention constructs a hybrid Kalman filter to achieve a joint estimation of SOC and SOH of the characteristic battery cell, including:

[0098] At the current estimation moment, the SOH estimation result of the characteristic cell at the previous estimation moment is used to estimate the SOC of the characteristic cell using the Adaptive Extended Kalman Filter (AEKF). The SOC estimation result of the characteristic cell is used to estimate the SOH of the characteristic cell using the Adaptive Unscented Kalman Filter (AUKF). Through multiple iterations, mutual feedback correction of the SOC and SOH estimation results of the characteristic cell is achieved to obtain the estimation results of the SOC and SOH of the characteristic cell.

[0099] Specifically, step 2 includes:

[0100] Step 2.1: Based on the circuit model of the characteristic cell, use AEKF to estimate the SOC of the characteristic cell, as follows:

[0101] Step 2.1.1: Based on the second-order RC circuit model of the characteristic cell, construct the state space equation and measurement equation for SOC estimation as follows:

[0102]

[0103] Where, For the The state vector at the estimation time, For the System inputs for estimation time, is the process noise of the state space equation, For the The terminal voltage measurement value of the battery cell at the estimation time, is the measurement noise of the measurement equation, For the The state matrix at the estimation time, For the The input matrix at the estimation moment, For the The observation matrix at the estimation moment, For the Direct transfer matrix at the estimation moment;

[0104] Among them, the characteristic battery is used in the Estimated time value , the voltage of the first RC link , the voltage of the second RC link Construction The state vector at the estimation time satisfies , with the first Estimated current of characteristic cells at the time of evaluation As system input, satisfy ; Then, the matrices in the state space equation and the measurement equation satisfy the following relationship:

[0105]

[0106]

[0107]

[0108]

[0109] Where, is the equation iteration time step, = , and are the resistance and capacitance of the first RC link, = , and are the resistance and capacitance of the second RC link respectively, For the Estimated capacity value of the characteristic cell at the estimation time, is the charge and discharge efficiency of the characteristic battery cell, For the Estimation of the open circuit voltage of the characteristic cell at the time of evaluation, is the internal resistance of the characteristic cell, and is the polarization internal resistance of the characteristic cell;

[0110] Among them, the open circuit voltage test data is used to fit the first The open circuit voltage of the characteristic cell at the time of estimation is as follows:

[0111]

[0112] Where, is the fitting parameter of SOC, is the exponential upper limit of the relationship fitting, For the SOC at the time of estimation Power, is the fitting parameter of the current, 、 、 is the first set of joint fitting parameters of SOC, 、 、 is the second set of joint fitting parameters of SOC;

[0113] Step 2.1.2, perform initialization settings according to the following relationship:

[0114]

[0115] Where, is the initial value of the state vector estimate, is the state vector exist = the initial value at time 0, is the mathematical expectation operation, is the initial value of the error covariance matrix estimate, represents the matrix transpose, is the initial value of the noise covariance matrix, is a positive definite initialization matrix, is the rated capacity of the characteristic battery cell.

[0116] Step 2.1.3, to avoid and Loss of positive definiteness in the calculation, after initialization, the measurement noise covariance matrix The estimated value and noise covariance of The estimated value of is improved to satisfy the following relationship:

[0117]

[0118] Where, is the forgetting factor, and its value range is .

[0119] Step 2.1.4, based on the improved estimate of the noise covariance, makes a priori estimates of the state vector and error covariance, including:

[0120] The time update of state estimation satisfies the following relationship:

[0121]

[0122] Where, For the Estimated state vector The prior estimate of For the Estimated state vector estimated value of;

[0123] The time update of the error covariance satisfies the following relationship:

[0124]

[0125] Where, For the Estimated time error covariance The prior estimate of For the Estimated noise covariance estimated value.

[0126] Step 2.1.5, based on the estimated value of the improved measurement noise covariance matrix, determine the Kalman gain matrix. Using the Kalman gain matrix, based on the state space equation and the measurement equation, perform a posteriori estimation of the state vector and the error covariance, including:

[0127] The Kalman gain matrix satisfies the following relationship:

[0128]

[0129] Where, For the The Kalman gain matrix at the estimation moment, For the Estimation time measurement noise covariance matrix estimated value of;

[0130] The measurement update of state estimation satisfies the following relationship:

[0131]

[0132] Where, Characteristic cells are Estimated time Estimation results, is the voltage estimation result of the first RC link, is the voltage estimation result of the second RC link;

[0133] The measurement update of the error covariance satisfies the following relationship:

[0134]

[0135] Where, is the identity matrix;

[0136] Step 2.1.6, from The characteristic cell is obtained from the state vector estimation at the estimation time. Estimated time Estimation results , and from Estimated time advances to Estimated time.

[0137] Step 2.2: Based on the SOC estimation results of the characteristic cell, use AUKF to estimate the SOH of the characteristic cell, as follows:

[0138] Step 2.2.1, based on the ampere-hour integration algorithm, using Estimation time characteristic battery cell Based on the estimation results, the state space equation and measurement equation for SOH estimation are established as follows:

[0139]

[0140] Where, For the Estimating the capacity of the characteristic battery cell at the time of evaluation, is the system noise, For the Estimated time and Estimated time The difference between the estimated results, is the charge and discharge efficiency of the characteristic battery cell, For the Estimating the current of the characteristic cell at the moment of evaluation, is the equation iteration time step, To measure noise;

[0141] Step 2.2.2: Initialize the settings according to the following relationship:

[0142]

[0143] Where, for = Estimated capacity value of the characteristic cell at time 0, for =The capacity of the characteristic cell at time 0, is the initial value of the error covariance estimate, is the initial value of the system noise covariance matrix estimate, To detect the initial value of the noise covariance matrix estimate, is the noise covariance matching window length.

[0144] Step 2.2.3, after initialization, make a priori estimate of the capacity of the characteristic cell and the error covariance between the priori estimate of the capacity of the characteristic cell and the output estimate, where Estimated time and Estimated time The estimated value of the difference between the estimated results is used as the output estimate; it is as follows:

[0145] 1) Select the sigma point for the estimated capacity of the characteristic cell as follows:

[0146]

[0147] Where, For the The estimated capacity value of the characteristic cell at the estimation time sigma points, For the The posterior estimate of the capacity of the characteristic cell at the estimation time, The characteristic cell capacity The unscented transformation value of the sigma point, For the The posterior estimate of the error covariance at the estimation moment, is the dimension of the state vector of the characteristic cell, Indicates taking the matrix No. OK, is the adjustment coefficient between the capacity estimation mean and error covariance of the characteristic cell, .

[0148] Using the sigma point symmetric sampling strategy, the corresponding weighting coefficient is:

[0149]

[0150] Where, For the The weighting coefficient corresponding to each sigma point;

[0151] 2) Prior estimation of the capacity of characteristic cells

[0152] The prior estimation of the capacity of the characteristic cell satisfies the following relationship:

[0153]

[0154] Where, For the A priori estimation of the capacity of the characteristic battery cell at the estimation time;

[0155] The error covariance of the capacity prior estimate satisfies the following relationship:

[0156]

[0157] Where, For the The error covariance of the prior estimation of the capacity of the characteristic battery cell at the estimation time, For the Estimated value of the system noise covariance matrix at the estimation time;

[0158] 3) Select the sigma point again for the estimated capacity of the characteristic cell, as follows:

[0159]

[0160] The weighting coefficients corresponding to each sigma point remain unchanged.

[0161] 4) The output estimation and output covariance estimation are obtained as follows:

[0162]

[0163]

[0164] Where, For the The output estimate at the estimation time is Estimated time and Estimated time The estimated value of the difference between the estimates, For the The output estimate at the estimation time is sigma points, is the error covariance of the output estimate, For the Estimation of the noise covariance matrix at the estimation moment;

[0165] Calculate the prior estimate of the capacity of the characteristic battery cell With output estimation The error covariance is as follows:

[0166]

[0167] Where, is the error covariance between the prior estimation of the capacity of the characteristic cell and the output estimation;

[0168] In step 2.2.4, the Kalman gain matrix is determined using the error covariance of the output estimate and the prior estimate of the characteristic cell capacity and the error covariance of the output estimate. The Kalman gain matrix is used to perform a posteriori estimation of the characteristic cell capacity and a posteriori estimation of the covariance based on the state space equation and measurement equation for SOH estimation, as follows:

[0169] Based on the Kalman filter principle, the posterior estimation of the characteristic cell capacity and the posterior estimation of the covariance are:

[0170]

[0171] Where, For the The error covariance of the posterior estimation of the capacity of the characteristic battery cell at the estimation time, For the Estimated state vector A priori estimate of ;

[0172] Adaptive covariance matching is:

[0173]

[0174] Step 2.2.5, obtain The posterior estimate of the capacity of the characteristic cell at the estimation time Then, the estimated SOH value is calculated by the following relationship:

[0175]

[0176] Where, is the rated capacity of the characteristic cell;

[0177] And from the Estimated time advances to Estimated time.

[0178] When the present invention adopts hybrid Kalman filtering to jointly estimate the SOC and SOH of the characteristic battery cell, the first SOC estimation value of characteristic cell at the estimation time , is the first input in the SOH estimation of characteristic cells based on AUKF The observed value of the measurement equation at the time of estimation; the SOH estimation of the characteristic cell based on AUKF is obtained Posterior estimate of the characteristic cell capacity at the estimation time , is the first input in the SOC estimation of characteristic cells based on the AEKF algorithm. The input values of the state-space equations at the estimation time.

[0179] like Figure 3 As shown in the figure, the AEKF algorithm and the AUKF algorithm fully interact in the joint estimation process of characteristic battery cells and SOC and SOH. By using AEKF to estimate SOC and jointly estimating SOH with AUKF, the estimation accuracy is significantly improved. In the battery system, the estimation of SOC and SOH has highly nonlinear characteristics. After the interaction between the two, the AUKF algorithm can provide a more accurate nonlinear state prediction for the AEKF algorithm, helping the AEKF algorithm to better handle linearization errors. Conversely, the AEKF algorithm can assist the AUKF algorithm in optimizing the measurement update steps, thereby significantly improving the overall accuracy of SOC and SOH estimation; and, the algorithm robustness is enhanced. When the system is subject to external interference and causes the estimation of one algorithm to deviate, the other algorithm can correct the deviation through information transmission during the interaction process. For example, if the AUKF algorithm's SOH estimation fluctuates greatly due to strong noise interference at a certain moment, the AEKF algorithm, based on its own relatively stable linear estimation characteristics, can pass relatively reliable SOC-related information to the AUKF algorithm to assist it in recalibrating the SOH estimation; moreover, the dynamic response capability is improved, and the interaction between the two algorithms can accelerate the tracking of these dynamic changes. During the battery operation, when the SOC changes rapidly, the AEKF algorithm can quickly capture this change trend and pass the information to the AUKF algorithm. Based on this, the AUKF algorithm can adjust its SOH estimation more quickly; in addition, it can also explore potential information associations. SOC and SOH are not completely independent variables. There is an inherent physical connection between them. The two algorithms can comprehensively analyze the battery's measurement data (such as voltage, current, etc.) from different angles.

[0180] In summary, the present invention constructs a high-precision second-order RC circuit model of the battery by selecting characteristic cells. A hybrid Kalman filter algorithm combining AEKF and AUKF is then used to perform high-precision joint estimation of the state of charge (SOC) and state of health (SOH) for the selected characteristic cells. The characteristic cells are selected based on their representativeness and reliability, ensuring that the constructed circuit model accurately reflects the electrochemical characteristics and dynamic behavior of the battery, and providing a foundation for state estimation of other cells in the entire energy storage system.

[0181] Step 3: Based on the circuit model of each battery cell, the SOC and SOH of each battery cell are estimated.

[0182] Specifically, step 3 includes:

[0183] Step 3.1, based on the cell circuit model, uses a method combining ampere-hour integration and open circuit voltage to estimate the SOC of each cell; including:

[0184] Step 3.1.1: Identify the internal resistance of each cell online based on the cell circuit model.

[0185] The present invention adopts Figure 4 The simple circuit model shown is used to estimate the SOC and SOH of all battery cells, reducing the algorithm complexity and the consumption of battery management system hardware computing power in engineering application scenarios.

[0186] Based on a simple circuit model of the battery cell, the internal resistance of each battery cell is identified in real time, satisfying the following relationship:

[0187]

[0188] Where, For the The internal resistance of a cell, For the The terminal voltage change of each cell, is the change of energy storage cluster current.

[0189] A simple battery circuit model is constructed for all cells, and the internal resistance of all cells is measured using online identification technology. This online identification process uses real-time data acquisition and analysis to accurately capture the changes in the internal resistance of the cells under different charge and discharge states, providing an important basis for SOC and SOH assessment, ensuring the real-time and effectiveness of the method.

[0190] Step 3.1.2, using the identified internal resistance of each cell, correct the SOC of each cell calculated by combining the ampere-hour integration and the open circuit voltage;

[0191] Specifically, the SOC of each cell is estimated by combining the ampere-hour integration and the open circuit voltage. Figure 5 Shown, including:

[0192] 1) Use the open circuit voltage lookup table method to obtain the initial SOC value of all cells .

[0193] 2) Calculate the SOC value of each battery cell using the ampere-hour integration method to meet the following relationship:

[0194]

[0195] Where, For the Estimated time The SOC of each battery cell, is the charge and discharge efficiency of the battery cell, For the Estimated current of all cells at the moment, For the Estimated time The capacity of a battery cell, is the time step for the calculation iteration.

[0196] 3) Using real-time recognition Internal resistance of a cell , calculate the The open circuit voltage value of each battery cell is obtained, and the SOC value corresponding to the open circuit voltage is obtained by the open circuit voltage lookup table method. , and then correct the SOC of each battery cell calculated by the ampere-hour integration method to satisfy the following relationship:

[0197]

[0198]

[0199] Where, For the Estimated time The SOC difference of each battery cell, For the Estimated time The estimated SOC value of each cell after correction, is the correction coefficient, and its value range is .

[0200] Step 3.2, estimate the SOH of each cell based on the open circuit voltage characteristics of the cell;

[0201] The SOH estimation principle of each cell used in the present invention is as follows: Figure 6 As shown, the charge and discharge amount of each battery cell is accumulated. SOC difference of each battery cell Corresponding open circuit voltage difference and accumulated charge , the calculation formula for the estimated value of SOH of each battery cell is as follows:

[0202]

[0203]

[0204] Where, For the Estimated time Estimated capacity of each battery cell, For the Estimated time The accumulated charge capacity of each battery cell, For the Estimated time Estimated SOH value of each battery cell, is the rated capacity of the battery cell.

[0205] By combining the open-circuit voltage characteristics of battery cells, a SOH assessment model is established. By periodically measuring the battery's open-circuit voltage, analyzing its relationship with SOC, and using this relationship to derive an estimated SOH value, this method effectively identifies battery aging and performance degradation by dynamically monitoring changes in the cell's open-circuit voltage, ensuring real-time SOH assessment.

[0206] In step 3.3, using the estimated capacity value determined based on the estimated SOH value of each battery cell, the estimated SOC value of each battery cell is obtained using the ampere-hour integration method, which satisfies the following relationship:

[0207]

[0208] Where, For the Estimated time The estimated SOC value of each battery cell, For the Estimated time Estimated capacity of each battery cell;

[0209] A method combining ampere-hour integration and open-circuit voltage is used to estimate the SOC of all cells. Specifically, this involves integrating the battery current to calculate the cumulative charge, and combining this with real-time open-circuit voltage measurement to build an SOC estimation model. This model can adaptively adjust to different operating conditions, improving the accuracy and stability of SOC estimation.

[0210] The fused estimation of SOC and SOH of each battery cell is achieved through the capacity estimation value of each battery cell at different estimation moments.

[0211] Step 4: Use the SOC estimation value and SOH estimation value of the characteristic cell obtained by the joint estimation to correct the SOC estimation value and SOH estimation value of each cell obtained by the fusion estimation.

[0212] The SOC and SOH of the characteristic cell are estimated using a more accurate battery model, AEKF and AUKF combined algorithm, achieving high-precision estimation of the SOC and SOH of the characteristic cell; while the SOC and SOH of all cells are obtained using a simpler method, using a relatively simple battery model and the ampere-hour integration method to estimate the SOC and SOH, which has lower accuracy; therefore, in step 4, the high-precision SOC and SOH estimation results of the characteristic cell are used to correct the SOC and SOH of all cells, thereby reducing the overall computational complexity while improving the estimation accuracy of all cells.

[0213] Taking the SOC estimated value of the characteristic cell obtained by the joint estimation as a constraint, the SOC estimated value of each cell obtained by the fusion estimation is corrected using the following relationship:

[0214]

[0215] Where, For the Estimated time Correction value of the estimated SOC value of each battery cell, For the The estimated SOC value of the characteristic cell corresponding to the maximum terminal voltage at the estimation moment, For the The estimated SOC value of the characteristic cell corresponding to the minimum terminal voltage at the estimation moment, For the Estimated time SOC estimation of each battery cell;

[0216] Taking the SOH estimated value of the characteristic cell obtained by the joint estimation as a constraint, the SOH estimated value of each cell obtained by the fusion estimation is corrected by the following relationship:

[0217] when hour,

[0218]

[0219] when hour:

[0220]

[0221] Where, For the Estimated time Correction value of the estimated SOH value of each battery cell, For the The minimum value of the estimated SOH value of each battery cell at the estimation time, For the The maximum value of the estimated SOH value of each cell at the estimation time, For the Estimated time Estimated SOH value of each battery cell, For the The estimated SOH value of the characteristic cell corresponding to the maximum terminal voltage at the estimation moment, For the The estimated SOH value of the characteristic cell corresponding to the minimum terminal voltage at the estimation moment.

[0222] As can be seen from the above relationship, the present invention uses the high-precision SOC and SOH estimates of the characteristic cell as a benchmark to correct the SOC and SOH estimates of all cells. By establishing a similarity relationship between the characteristic cell and other cells, the precise state information of the characteristic cell is effectively propagated to the entire battery pack, ensuring a more accurate state estimate for all cells in the pack.

[0223] This invention proposes a joint SOC and SOH estimation algorithm for characteristic cells that combines an adaptive extended Kalman filter with an adaptive unscented Kalman filter. This algorithm better adapts to the fast-changing SOC characteristics and nonlinear SOH characteristics of energy storage batteries. It also proposes a joint SOC and SOH estimation algorithm that combines a high-precision battery model with a simplified model, significantly reducing the algorithm's computational complexity. The algorithm uses the high-precision SOC and SOH estimates of characteristic cells to calibrate the SOC and SOH estimates of all batteries, improving the estimation accuracy of the SOC and SOH of all batteries. This method is applicable to various types of energy storage batteries, including lithium-ion batteries, sodium-ion batteries, and lead-acid batteries. By performing appropriate model adjustments and parameter optimization based on the characteristics of different battery types, this method can be widely applied to various energy storage systems. This method has low computational complexity and resource usage. Through efficient model simplification and algorithm optimization, it can complete the joint SOC and SOH estimation with fewer computational resources in practical applications, ensuring the effectiveness and practicality of the method in large-scale energy storage systems.

[0224] The present invention also proposes a system for jointly estimating the SOC and SOH of all cells in an energy storage system, comprising:

[0225] The SOC and SOH joint estimation module of the characteristic battery cell is used to measure the terminal voltage of each battery, and use the battery cell with the largest terminal voltage and the battery cell with the smallest terminal voltage as the characteristic battery cell; establish a circuit model of the characteristic battery cell; and based on the circuit model of the characteristic battery cell, jointly estimate the SOC and SOH of the characteristic battery cell;

[0226] The SOC and SOH fusion estimation module of each battery cell is used to fusion estimate the SOC and SOH of each battery cell based on the circuit model of each battery cell;

[0227] The SOC and SOH joint estimation module for all battery cells is used to correct the SOC and SOH estimated values of each battery cell obtained by fusion estimation based on the SOC estimated values and SOH estimated values of the characteristic battery cells obtained by the joint estimation, and use them as the SOC and SOH joint estimation results of all battery cells in the energy storage system.

[0228] In this embodiment, Long Short-Term Memory (LSTM) neural networks are used to implement the SOC and SOH estimates of characteristic cells obtained through joint estimation, and to correct the SOC and SOH estimates of each cell obtained through fusion estimation. This process can be combined with advanced monitoring and data analysis platforms, leveraging historical operating data to improve the estimation accuracy of the SOC and SOH of all batteries, thereby enabling intelligent monitoring and management of the entire energy storage battery system. On the monitoring and data analysis platform, a high-precision model-based joint SOC and SOH estimation algorithm for characteristic cells is deployed to fully utilize the platform's powerful hardware resources. The SOC and SOH joint estimation algorithm is redundantly deployed on the monitoring and data analysis platform and the energy storage battery management system, providing mutual backup and improving the operational reliability of the energy storage system. The SOC and SOH joint estimation results are combined with cloud computing and big data analysis technologies, and artificial intelligence algorithms are applied to compensate and correct the joint SOC and SOH estimation results for all cells.

[0229] The present invention uses an LSTM-based artificial intelligence algorithm to implement compensation correction of the platform's combined SOC and SOH estimation results for all battery cells. The steps are as follows:

[0230] (1) Obtaining algorithm dataset

[0231] In constructing the hybrid filtering algorithm, the parameters related to SOC and SOH in each algorithm cycle constitute the algorithm input data set.

[0232]

[0233] Where, For the The input dataset for the estimation moment, … For battery cell 1, ..., The terminal voltage, … For battery cell 1, ..., The current, … For battery cell 1, ..., The estimated state of charge, … For battery cell 1, ..., The estimated capacity of To estimate the time, is the total number of battery cells.

[0234] The difference between the estimated value and the reference value in each estimation cycle is calculated to construct the algorithm output data set.

[0235]

[0236] Where, For the The output dataset at the estimation time, … For battery cell 1, ..., The difference between the estimated SOC value and the reference value, … For battery cell 1, ..., The difference between the estimated SOH value and the reference value.

[0237] (2) Training the LTSM neural network

[0238] The algorithm input data set and the algorithm output data set are sent to the LTSM neural network model for training until the model output error is less than the preset value.

[0239] (3) Algorithm compensation

[0240] The LTSM neural network is deployed to compensate and correct the SOC and SOH estimates output by the hybrid filtering algorithm.

[0241]

[0242] Where, 、 Respectively Estimated time The prior and posterior estimates of the SOC of each battery cell, 、 Respectively Estimated time Prior and posterior estimates of the SOH of each battery cell.

[0243] The above is only part of the implementation principle of the present invention and does not limit the present invention in any form. Parallel prediction methods using other total current prediction methods and total voltage prediction methods all fall within the scope of protection of the technical solution of the present invention.

[0244] The present disclosure may be a system, method and / or computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0245] A computer-readable storage medium can be a tangible device that can hold and store instructions for use by an instruction execution device. A computer-readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punched card or raised structure in a groove on which instructions are stored, and any suitable combination thereof. As used herein, a computer-readable storage medium is not to be construed as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse passing through a fiber optic cable), or an electrical signal transmitted through an electrical wire.

[0246] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions to be stored in the computer-readable storage medium in each computing / processing device.

[0247] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages such as Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, the state information of the computer-readable program instructions is used to personalize an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), so that the electronic circuit can execute the computer-readable program instructions, thereby implementing various aspects of the present disclosure.

[0248] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A method for jointly estimating the SOC and SOH of all cells in an energy storage system, characterized in that: include: Measure the terminal voltage of each battery, and take the cell with the largest terminal voltage and the cell with the smallest terminal voltage as the characteristic cells; Establish a circuit model for characteristic cells; Adaptive extended Kalman filter and adaptive unscented Kalman filter are used to construct hybrid Kalman filter, and the SOC and SOH of characteristic cells are jointly estimated based on the circuit model of characteristic cells, including: constructing state space equations and measurement equations for SOC estimation based on the circuit model of characteristic cells; using the ampere-hour integration algorithm to Estimation time characteristic battery cell Estimation results, establish the state space equation and measurement equation for SOH estimation; use adaptive extended Kalman filter to estimate the SOC of characteristic cells, and get the first The estimated SOC value of the characteristic cell at the estimation moment is the first input in the SOH estimation of the characteristic cell based on the adaptive unscented Kalman filter. The observation value of the measurement equation at the estimation moment; the SOH estimation of the characteristic cell based on the adaptive unscented Kalman filter is obtained The posterior estimated value of the characteristic cell capacity at the estimation time is the first input in the SOC estimation of the characteristic cell based on the adaptive extended Kalman filter. Estimation of the input values of the state space equations at the moment of evaluation; Based on the circuit model of each battery cell, the SOC and SOH of each battery cell are estimated in a fusion manner; The SOC estimated value and SOH estimated value of the characteristic battery cell obtained by the joint estimation are used to correct the SOC estimated value and SOH estimated value of each battery cell obtained by the fusion estimation, and the result is the joint estimation result of the SOC and SOH of all battery cells in the energy storage system.

2. The method for jointly estimating the SOC and SOH of all cells in the energy storage system according to claim 1, characterized in that: The circuit model of the characteristic cell is a second-order RC circuit model, which includes: The first RC link and the second RC link.

3. The method for jointly estimating the SOC and SOH of all cells in the energy storage system according to claim 1, characterized in that: Based on the circuit model of the characteristic cell, the adaptive extended Kalman filter is used to estimate the SOC of the characteristic cell, including: After initialization, the estimated values of the measurement noise covariance matrix and the estimated values of the noise covariance are improved; A priori estimation of the state vector and error covariance is performed based on the estimated value of the improved noise covariance; Determine the Kalman gain matrix based on the estimated value of the improved measurement noise covariance matrix, and use the Kalman gain matrix to perform a posteriori estimation of the state vector and error covariance based on the state space equation and measurement equation for SOC estimation; From The state vector estimation at the estimation time is obtained Estimation time characteristic battery cell Estimation results.

4. The method for jointly estimating the SOC and SOH of all cells in the energy storage system according to claim 3, characterized in that: The state space equation and measurement equation for SOC estimation are as follows: Where, For the The state vector at the estimation time, For the System inputs for estimation time, is the process noise of the state space equation, For the The terminal voltage measurement value of the battery cell at the estimation time, is the measurement noise of the measurement equation, For the The state matrix at the estimation time, For the The input matrix at the estimation moment, For the The observation matrix at the estimation moment, For the Direct transfer matrix at the estimation moment; Among them, the use of characteristic cells in the Estimated time value , the voltage of the first RC link , the voltage of the second RC link Construction The state vector at the estimation time satisfies , with the first Estimated current of characteristic cells at the time of evaluation As system input, satisfy .

5. The method for jointly estimating the SOC and SOH of all cells in the energy storage system according to claim 4, characterized in that: The matrices in the state space equation and the measurement equation satisfy the following relationship: Where, is the equation iteration time step, = , and are the resistance and capacitance of the first RC link, = , and are the resistance and capacitance of the second RC link respectively, For the Estimated capacity value of the characteristic cell at the estimation time, is the charge and discharge efficiency of the characteristic battery cell, For the Estimation of the open circuit voltage of the characteristic cell at the time of evaluation, is the internal resistance of the characteristic cell, and is the polarization internal resistance of the characteristic cell.

6. The method for jointly estimating the SOC and SOH of all cells in the energy storage system according to claim 5, characterized in that: The measurement noise covariance matrix The estimated value and noise covariance of The estimated value of is improved to satisfy the following relationship: Where, is the forgetting factor, and its value range is , 、 All are Estimation of the intermediate parameters constructed at the moment, For the The estimated value of the measurement noise covariance matrix at the estimation moment, For the The estimated value of the noise covariance at the estimation time, For the The Kalman gain matrix at the estimation moment, For the The estimated value of the state vector at the estimation time.

7. The method for jointly estimating the SOC and SOH of all cells in the energy storage system according to claim 1, characterized in that: The SOH estimation of characteristic cells is performed based on the adaptive unscented Kalman filter, including: After initialization, the capacity of the characteristic cell is estimated a priori, the error covariance of the output estimate is calculated, and the error covariance of the capacity a priori estimate of the characteristic cell and the output estimate is calculated. Estimated time and Estimated time The estimated value of the difference between the estimated results is used as the output estimate; The Kalman gain matrix is determined using the error covariance of the output estimation and the prior estimation of the characteristic cell capacity and the error covariance of the output estimation. The Kalman gain matrix is then used to perform a posteriori estimation of the characteristic cell capacity and covariance based on the state space equation and measurement equation for SOH estimation. Get the first After estimating the a posteriori estimate of the capacity of the characteristic cell at the time of estimation, the SOH estimate is calculated using the following relationship: Where, For the Estimation time characteristic battery cell Estimation results, For the The posterior estimate of the capacity of the characteristic cell at the estimation time, is the rated capacity of the characteristic battery cell.

8. The method for jointly estimating the SOC and SOH of all cells in the energy storage system according to claim 7, characterized in that: The state space equation and measurement equation for SOH estimation are as follows: Where, For the Estimating the capacity of the characteristic battery cell at the time of evaluation, is the system noise, For the Estimated time and Estimated time The difference between the estimated results, For the Estimation time characteristic battery cell Estimation results, is the charge and discharge efficiency of the characteristic battery cell, For the Estimating the current of the characteristic cell at the moment of evaluation, is the equation iteration time step, To measure noise.

9. The method for jointly estimating the SOC and SOH of all cells in the energy storage system according to claim 1, characterized in that: Based on the circuit model of each cell, the SOC and SOH of each cell are estimated, including: Identify the internal resistance of each battery cell based on the cell circuit model; The internal resistance of each battery cell obtained by identification is used to correct the SOC of each battery cell calculated by combining the ampere-hour integration and the open circuit voltage. The SOC difference of each battery cell is used to estimate the SOH of each battery cell based on the open circuit voltage characteristics of the battery cell. The estimated capacity value determined based on the estimated SOH value of each battery cell is used to obtain the estimated SOC value of each battery cell using the ampere-hour integration method.

10. The method for jointly estimating the SOC and SOH of all cells in the energy storage system according to claim 9, characterized in that: The calculation formula for the estimated value of SOH of each battery cell is as follows: Where, For the Estimated time Estimated capacity of each battery cell, For the Estimated time The accumulated charge capacity of each battery cell, For the Estimated time The SOC difference of each battery cell, For the Estimated time Estimated SOH value of each battery cell, is the rated capacity of the battery cell; Where, is the first Estimated time The SOC of each cell is obtained by identifying the Internal resistance of a cell , calculate the The open circuit voltage value of each battery cell is obtained, and the SOC value corresponding to the open circuit voltage is obtained by the open circuit voltage lookup table method. .

11. The method for jointly estimating SOC and SOH of all cells in an energy storage system according to claim 10, characterized in that: The estimated SOC value of each battery cell satisfies the following relationship: Where, For the Estimated time The estimated SOC value of each battery cell, is the charge and discharge efficiency of the battery cell, For the Estimated current of all cells at the moment, For the Estimated time Estimated capacity of each battery cell, is the time step for the calculation iteration.

12. The method for jointly estimating the SOC and SOH of all cells in the energy storage system according to claim 1, characterized in that: Taking the SOC estimated value of the characteristic cell obtained by the joint estimation as a constraint, the SOC estimated value of each cell obtained by the fusion estimation is corrected using the following relationship: Where, For the Estimated time Correction value of the estimated SOC value of each battery cell, For the The estimated SOC value of the characteristic cell corresponding to the maximum terminal voltage at the estimation moment, For the The estimated SOC value of the characteristic cell corresponding to the minimum terminal voltage at the estimation moment, For the Estimated time Estimated SOC value of each battery cell.

13. The method for jointly estimating the SOC and SOH of all cells in an energy storage system according to claim 1, characterized in that: Taking the SOH estimated value of the characteristic cell obtained by the joint estimation as a constraint, the SOH estimated value of each cell obtained by the fusion estimation is corrected by the following relationship: when hour, when hour: Where, For the Estimated time Correction value of the estimated SOH value of each battery cell, For the The minimum value of the estimated SOH value of each battery cell at the estimation time, For the The maximum value of the estimated SOH value of each cell at the estimation time, For the Estimated time Estimated SOH value of each battery cell, For the The estimated SOH value of the characteristic cell corresponding to the maximum terminal voltage at the estimation moment, For the The estimated SOH value of the characteristic cell corresponding to the minimum terminal voltage at the estimation moment.

14. A system for jointly estimating the SOC and SOH of all cells in an energy storage system, for implementing the method for jointly estimating the SOC and SOH of all cells in an energy storage system according to any one of claims 1 to 13, characterized in that: include: The SOC and SOH joint estimation module of the characteristic battery cell is used to measure the terminal voltage of each battery, and the battery cell with the largest terminal voltage and the battery cell with the smallest terminal voltage are used as the characteristic battery cell; Establish a circuit model for the characteristic cell; based on the circuit model of the characteristic cell, jointly estimate the SOC and SOH of the characteristic cell; The SOC and SOH fusion estimation module for each battery cell is used to construct a hybrid Kalman filter using an adaptive extended Kalman filter and an adaptive unscented Kalman filter. Based on the circuit model of the characteristic battery cell, the SOC and SOH of the characteristic battery cell are jointly estimated. This includes: constructing the state space equation and measurement equation for SOC estimation based on the circuit model of the characteristic battery cell; Based on the ampere-hour integration algorithm, using the Estimation time characteristic battery cell Estimation results, establish the state space equation and measurement equation for SOH estimation; use adaptive extended Kalman filter to estimate the SOC of characteristic cells, and get the first The estimated SOC value of the characteristic cell at the estimation moment is the first input in the SOH estimation of the characteristic cell based on the adaptive unscented Kalman filter. The observation value of the measurement equation at the estimation moment; the SOH estimation of the characteristic cell based on the adaptive unscented Kalman filter is obtained The posterior estimated value of the characteristic cell capacity at the estimation time is the first input in the SOC estimation of the characteristic cell based on the adaptive extended Kalman filter. Estimation of the input values of the state space equations at the moment of evaluation; The SOC and SOH joint estimation module for all battery cells is used to correct the SOC and SOH estimated values of each battery cell obtained by fusion estimation based on the SOC estimated values and SOH estimated values of the characteristic battery cells obtained by the joint estimation, and use them as the SOC and SOH joint estimation results of all battery cells in the energy storage system.

Images