CN-121186631-B - Battery SOC estimation method combining current integration and open-circuit voltage correction

CN121186631BCN 121186631 BCN121186631 BCN 121186631BCN-121186631-B

Abstract

The invention relates to the technical field of battery electric quantity calculation, in particular to a battery SOC estimation method combining current integration and open-circuit voltage correction, which comprises the steps of obtaining an equivalent circuit model of a target battery from a database, and extracting parameters of the equivalent circuit model; the parameters of the equivalent circuit model comprise open-circuit voltage, battery internal resistance, polarization capacitance and polarization resistance, the SOC of the target battery is estimated and corrected through the SOC estimation model based on the equivalent circuit model of the battery, the final estimated value of the SOC of the target battery is output, and the parameters of the equivalent circuit model of the battery are updated on line to match the current battery condition, so that the adaptability of the SOC estimation model to battery aging is improved.

Inventors

ZHAO YANAN
LIU JI
CAO SHENGJI
XU GUIJIA
YANG WEI
SHAO CHUAN
HUA CHUNLIANG

Assignees

小洋电源股份有限公司

Dates

Publication Date: 20260508
Application Date: 20251104

Claims (8)

1. A battery SOC estimation method combining current integration and open circuit voltage correction, the method comprising: Obtaining an equivalent circuit model of a target battery from a database, and extracting parameters of the equivalent circuit model, wherein the parameters of the equivalent circuit model comprise open circuit voltage, battery internal resistance, polarization capacitance and polarization resistance; Based on an equivalent circuit model of the battery, estimating and correcting the SOC of the target battery through an SOC estimation model, and outputting a final estimated value of the SOC of the target battery; Updating parameters of an equivalent circuit model of the battery on line to match current battery conditions and improve the adaptability of the SOC estimation model to battery aging, comprising: providing continuous effective excitation, enabling the SOC estimation model to identify parameters of the equivalent circuit model and update on-line, comprising: judging battery working conditions, wherein the battery working conditions comprise controllable working conditions and uncontrollable dynamic working conditions; if the battery working condition is a controllable working condition, actively injecting a continuous excitation signal, and updating parameters of an equivalent circuit model; If the battery working condition is an uncontrollable dynamic working condition, the excitation sufficiency is detected, and if the excitation sufficiency is enough, the corresponding excitation is utilized to update the parameters of the equivalent circuit model, otherwise, the parameters are not updated, namely the parameters of the current equivalent circuit model are continuously used; constructing a new equivalent circuit model of the target battery by using the updated parameters, and matching the current actual state of the battery so as to improve the adaptability of the SOC estimation model to battery aging; And outputting the estimated value of the target battery SOC through the SOC estimation model by using the new equivalent circuit model.
2. The method for estimating the SOC of the target battery by combining current integration and open circuit voltage correction according to claim 1, wherein estimating and correcting the SOC of the target battery by the SOC estimation model, outputting a final estimated value of the SOC of the target battery, comprises: The method comprises the steps of constructing an SOC estimation model, wherein the SOC estimation model comprises a current integration model, an open-circuit voltage calibration model and a Kalman filter; predicting the SOC of the target battery through a current integration model, and outputting an SOC predicted value; Calibrating the SOC predicted value through an open circuit voltage calibration model; and correcting the SOC predicted value through a Kalman filter to obtain the optimal SOC estimated value at the current moment.
3. The battery SOC estimation method combining current integration and open circuit voltage correction as claimed in claim 2, wherein the predicting the target battery SOC by the current integration model and outputting the SOC prediction value comprises: based on the SOC prediction value at the previous time And the current value of the currently measured battery Predicting a priori value of SOC at current time Wherein, the method comprises the steps of, Indicating the rated capacity of the battery, The time interval is represented by a time interval, Representing coulombic efficiency; The calibrating the SOC prediction value by the open circuit voltage calibration model includes: Calculating a predicted terminal voltage , Indicating the open circuit voltage (open circuit voltage), Represents the internal resistance of the device, Representing a polarization voltage; Calculating voltage residual Wherein, the method comprises the steps of, Representing the actual measured terminal voltage; If it is It means that the current integration model predicts the battery SOC accurately if Indicating that the current integration model has accumulated errors; correcting the SOC predicted value through a Kalman filter to obtain an optimal SOC estimated value at the current moment, wherein the method comprises the following steps: Computing a Kalman gain matrix Wherein, the method comprises the steps of, The covariance matrix is represented by a matrix of covariance, Representing the observation matrix of the image of the object, Representing a measurement noise covariance matrix; correcting the predicted SOC by using the calculated Kalman gain matrix and the voltage residual error to obtain an SOC estimated value at the current moment 。
4. A method of estimating battery SOC in combination with current integration and open circuit voltage correction as claimed in claim 3, wherein if the battery condition is a controllable condition, the active injection of the continuous excitation signal updates the parameters comprising: generating a pseudo-random binary sequence current signal of limited amplitude range ; Acquiring the state of the battery, and if the battery is in a constant voltage charge state or a static state, namely a controllable working condition, obtaining Superimposed to a reference charging current Obtaining a mixed current signal I.e. ; Introducing a forgetting factor algorithm, comprising: during active injection excitation, the equivalent circuit model of the battery is reduced to a linear regression model: ; Wherein, the Representing the output of the equivalent circuit model, ; Represents the open circuit voltage of the equivalent circuit model output, The open-circuit voltage value obtained by inquiring the OCV-SOC curve after representing the SOC predicted value obtained based on the current integration method; The regression vector is represented as a function of the regression vector, ; Representing parameters to be identified; wherein, the method comprises the steps of, Represents the internal resistance of the battery, Representing the polarization resistance of the battery; the initialization is performed, i.e. at time 0, Covariance matrix , wherein, Representing the identity matrix, the constant ; For each sampling instant Calculating a priori error ; ; At this time, the kalman gain matrix may be expressed as: wherein, the method comprises the steps of, Representing the forgetting factor, ; Updating parameters, wherein the updated parameters are ; Updating covariance matrix, wherein the updated covariance matrix is 。
5. The method for estimating battery SOC combining current integration and open circuit voltage correction as claimed in claim 4, wherein said detecting of excitation sufficiency if the battery condition is an uncontrollable dynamic condition and updating the parameters with the corresponding excitation if the excitation is sufficient comprises: obtaining excitation, namely obtaining a battery current signal which changes due to battery load change under uncontrollable dynamic working conditions; the initialization is performed, i.e. at time 0, Covariance matrix , wherein, Representing the identity matrix, the constant ; For each sampling instant Calculating a priori error ; ; An adaptive forgetting factor algorithm is introduced, comprising: Representing the forgetting factor as a variable that varies with time, i.e. adaptive forgetting factor ; Wherein, the Representing the sampling instant Is a self-adaptive forgetting factor of (1), Representing a preset minimum value of the forgetting factor, Represents the adjustment coefficient of the device, Representing the excitation index; ; Calculating an average excitation index within an excitation evaluation window Wherein, the method comprises the steps of, Representing the number of times employed, i.e. the length of the excitation evaluation window; If it is Or (b) Judging that the excitation is sufficient, and updating parameters by using a current signal of the change caused by the change of the battery load under the uncontrollable dynamic working condition; otherwise, judging that the excitation is insufficient and not updating the parameters; Wherein, the Representation of Is used for the track of (a), Representing an uncertainty threshold; Representing an excitation level threshold.
6. The method for estimating battery SOC combining current integration and open circuit voltage correction according to claim 5, wherein updating parameters using the varying current signal caused by the battery load variation under excitation-uncontrollable dynamic conditions comprises: By means of Calculating uncontrollable dynamic operating mode gain vector ; Updating parameters to obtain updated parameter estimation under uncontrollable dynamic working conditions ; Updating covariance matrix to obtain updated covariance matrix under uncontrollable dynamic working condition 。
7. An electronic device comprising a processor, a communication module and a memory connected to the processor, wherein the electronic device is configured to implement a battery SOC estimation method in combination with current integration and open circuit voltage correction as claimed in any of the preceding claims 1-6.
8. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program, which is executed by a processor to implement the battery SOC estimation method combining current integration and open circuit voltage correction of any of the above claims 1-6.

Description

Battery SOC estimation method combining current integration and open-circuit voltage correction Technical Field The invention relates to the technical field of battery electric quantity calculation, in particular to a battery SOC estimation method combining current integration and open-circuit voltage correction. Background SOC, the state of charge, is colloquially the "remaining charge" of a battery. I.e., the ratio of the current battery remaining capacity to the nominal capacity, is typically expressed in percent. Accurate SOC estimation is critical to the safety, performance and life of a battery system, for example, to prevent overcharge and overdischarge, to ensure that the battery operates within a safe voltage range, to prevent battery damage, fire and even explosion due to overcharge or overdischarge, to improve user experience, to provide accurate range predictions (e.g., for electric vehicles) for users, to avoid "range anxiety", to optimize energy management, to decide when to charge and discharge based on SOC in a hybrid vehicle or energy storage system, to achieve the most efficient energy utilization, and to balance a battery pack, to provide accurate SOC in a battery pack consisting of multiple cells, to provide a basis for balanced management of the cells. In practical applications, especially in high performance fields such as electric vehicles, fusion algorithms are generally used. The most common is the combination strategy of ampere-hour integration (current integration) +open-circuit voltage calibration+kalman filtering to give consideration to real-time, accuracy and robustness. However, as the battery ages, causing an increase in internal resistance and a decrease in capacity, the initially calibrated equivalent circuit model parameters and OCV curves of the battery gradually fail. If the online parameter identification or parameter update is not performed, the estimation accuracy is reduced. Therefore, how to solve the problems of the SOC estimation algorithm, such as adaptability to battery aging and ensuring accuracy of SOC estimation, is a key and difficult problem in current research and practice. Disclosure of Invention In the invention, in the SOC estimation process, the equivalent circuit model parameters are updated in real time, so that the actual state of the current battery can be reflected, and the accuracy of SOC estimation is ensured. The technical scheme provided by the invention is that the battery SOC estimation method combining current integration and open-circuit voltage correction comprises the following steps: Obtaining an equivalent circuit model of a target battery from a database, and extracting parameters of the equivalent circuit model, wherein the parameters of the equivalent circuit model comprise open circuit voltage, battery internal resistance, polarization capacitance and polarization resistance; Based on an equivalent circuit model of the battery, estimating and correcting the SOC of the target battery through an SOC estimation model, and outputting a final estimated value of the SOC of the target battery; Parameters of an equivalent circuit model of the battery are updated on line to match the current battery condition, and the adaptability of the SOC estimation model to battery aging is improved. Preferably, the estimating and correcting the SOC of the target battery by the SOC estimation model, and outputting a final estimated value of the SOC of the target battery, includes: The method comprises the steps of constructing an SOC estimation model, wherein the SOC estimation model comprises a current integration model, an open-circuit voltage calibration model and a Kalman filter; predicting the SOC of the target battery through a current integration model, and outputting an SOC predicted value; Calibrating the SOC predicted value through an open circuit voltage calibration model; and correcting the SOC predicted value through a Kalman filter to obtain the optimal SOC estimated value at the current moment. Preferably, the predicting the SOC of the target battery by the current integration model and outputting the SOC prediction value includes: based on the SOC prediction value at the previous time And the current value of the currently measured batteryPredicting a priori value of SOC at current timeWherein, the method comprises the steps of,Indicating the rated capacity of the battery,The time interval is represented by a time interval,Representing coulombic efficiency; The calibrating the SOC prediction value by the open circuit voltage calibration model includes: Calculating a predicted terminal voltage ,Indicating the open circuit voltage (open circuit voltage),Represents the internal resistance of the device,Representing a polarization voltage; Calculating voltage residual Wherein, the method comprises the steps of,Representing the actual measured terminal voltage; If it is It means that the current integration model predicts the battery SOC acc