EP-4737924-A1 - BATTERY STATE-OF-HEALTH ESTIMATION METHOD, ELECTRONIC DEVICE AND COMPUTER-READABLE STORAGE MEDIUM

EP4737924A1EP 4737924 A1EP4737924 A1EP 4737924A1EP-4737924-A1

Abstract

Embodiments of the present application provide a battery state of health estimation method, an electronic device, and a computer-readable storage medium. The battery state of health estimation method includes: obtaining historical operating condition data of a to-be-tested battery module; performing parameter identification on a battery parameter of the battery module based on the historical operating condition data to obtain an identification result; and predicting a state of health of a battery in the battery module based on the identification result. The above method can effectively improve the estimation accuracy of the battery state of health.

Inventors

QIAN, Mu

Assignees

Contemporary Amperex Future Energy Research Institute (Shanghai) Limited
Contemporary Amperex Technology Co., Limited

Dates

Publication Date: 20260506
Application Date: 20240110

Claims (20)

A battery state of health estimation method characterized by comprising: obtaining historical operating condition data of a to-be-tested battery module; performing parameter identification on a battery parameter of the battery module based on the historical operating condition data to obtain an identification result; and predicting a state of health of a battery in the battery module based on the identification result.
The method according to claim 1, characterized in that the performing parameter identification on a battery parameter of the battery module based on the historical operating condition data to obtain an identification result comprises: obtaining a mathematical model of the battery module, wherein the mathematical model is constructed based on the battery parameter of the battery module; and performing parameter identification on the battery parameter of the battery module based on the historical operating condition data and the mathematical model to obtain the identification result.
The method according to claim 2, characterized in that the performing parameter identification on the battery parameter of the battery module based on the historical operating condition data and the mathematical model to obtain the identification result comprises: discretizing the mathematical model based on the historical operating condition data to obtain a discretized mathematical model; and performing parameter identification on the battery parameter of the battery module based on the historical operating condition data and the discretized mathematical model to obtain the identification result.
The method according to claim 3, characterized in that the discretizing the mathematical model based on the historical operating condition data to obtain a discretized mathematical model comprises: obtaining a sampling frequency of the historical operating condition data; and discretizing the mathematical model based on the sampling frequency of the historical operating condition data to obtain the discretized mathematical model.
The method according to claim 4, characterized in that the performing parameter identification on the battery parameter of the battery module based on the historical operating condition data and the discretized mathematical model to obtain the identification result comprises: establishing an identification model based on the discretized mathematical model; and performing parameter identification based on the historical operating condition data and the identification model to obtain the identification result.
The method according to claim 5, characterized in that the establishing an identification model based on the discretized mathematical model comprises: constructing a target function based on the discretized mathematical model; obtaining a constraint condition for an independent variable in the mathematical model; and establishing the identification model based on the target function and the constraint condition.
The method according to claim 6, characterized in that the independent variable comprises an initial state of charge of the battery module; and the obtaining a constraint condition for an independent variable in the mathematical model comprises: obtaining historical data of the initial state of charge from the historical operating condition data, and determining a constraint condition for the initial state of charge based on first frame data in the historical data of the initial state of charge.
The method according to claim 6, characterized in that the independent variable comprises a battery capacity of the battery module; and the obtaining a constraint condition for an independent variable in the mathematical model comprises: obtaining historical charging data of the battery module from the historical operating condition data, estimating the battery capacity of the battery module based on the historical charging data to obtain an estimated value of the battery capacity, and determining a constraint condition for the battery capacity based on the estimated value of the battery capacity.
The method according to any one of claims 6 to 8, characterized in that the constraint condition comprises a linear constraint and a nonlinear constraint; and the establishing the identification model based on the target function and the constraint condition comprises: adding a multiplier to the nonlinear constraint of the constraint condition to obtain a nonlinear constraint term, adding the nonlinear constraint term to the target function to obtain an updated target function, and establishing the identification model based on the updated target function and the linear constraint of the constraint condition.
The method according to any one of claims 6 to 9, characterized in that the independent variable of the mathematical model comprises an observable quantity and a non-observable quantity, and the historical operating condition data comprise an initial value of the non-observable quantity, historical observation data of the observable quantity, and historical observation data of a dependent variable of the mathematical model; and the performing parameter identification based on the historical operating condition data and the identification model to obtain the identification result comprises: inputting the initial value of the non-observable quantity, the historical observation data of the observable quantity, and the historical observation data of the dependent variable into the identification model to obtain a first value of the target function, adjusting the value of the non-observable quantity based on the first value to obtain a first adjusted value of the non-observable quantity, inputting the first adjusted value, the historical observation data of the observable quantity, and the historical observation data of the dependent variable into the identification model to obtain a second value of the target function, if the second value does not satisfy a preset condition, continuing to adjust the value of the non-observable quantity based on the second value until the value of the target function satisfies the preset condition, and if the second value satisfies the preset condition, determining a current value of the non-observable quantity as the identification result.
The method according to claim 10, characterized in that the adjusting the value of the non-observable quantity based on the first value to obtain a first adjusted value of the non-observable quantity comprises: determining a feasible range of the non-observable quantity based on the first value; and determining the first adjusted value from the feasible range, wherein a value of the target function corresponding to the first adjusted value is less than a value of the target function corresponding to any other adjusted value within the feasible range.
The method according to any one of claims 2 to 11, characterized in that the method further comprises: optimizing the mathematical model of the battery module according to a preset cycle.
The method according to claim 12, characterized in that a step of updating the mathematical model of the battery module each time comprises: determining a dependent variable of the mathematical model based on the battery parameter; determining an independent variable related to the dependent variable based on the battery parameter; and constructing the mathematical model based on the independent variable and the dependent variable.
The method according to claim 13, characterized in that the constructing the mathematical model based on the independent variable and the dependent variable comprises: obtaining an equivalent circuit model of the battery module; and constructing the mathematical model between the independent variable and the dependent variable based on the equivalent circuit model.
The method according to claim 14, characterized in that the equivalent circuit model comprises a first resistor, a second resistor, a first capacitor, and a battery, wherein the second resistor and the first capacitor are connected in parallel to form a parallel circuit, and the battery, the first internal resistance, and the parallel circuit are connected in series; and when the dependent variable is an output voltage of the battery module, the constructing the mathematical model between the independent variable and the dependent variable based on the equivalent circuit model comprises: constructing a first expression for an open-circuit voltage of the battery based on the independent variable, wherein the open-circuit voltage represents a potential difference between a positive electrode and a negative electrode of the battery; constructing a second expression for an internal resistance voltage of the first internal resistance based on the independent variable, wherein the internal resistance voltage represents a potential difference between two connection terminals of the first internal resistance; constructing a third expression for a parallel voltage of the parallel circuit based on the independent variable, wherein the parallel voltage represents a potential difference between two connection terminals of the parallel circuit; and constructing the mathematical model based on the first expression, the second expression, and the third expression.
The method according to claim 15, characterized in that the independent variable comprises an output current, a battery capacity, and an initial state of charge of the battery module; and the constructing a first expression for an open-circuit voltage of the battery based on the independent variable comprises: constructing a fourth expression for a real-time state of charge based on the output current, the battery capacity, and the initial state of charge of the independent variable, and constructing the first expression based on the fourth expression and a relationship between the real-time state of charge and the open-circuit voltage.
The method according to claim 16, characterized in that the independent variable comprises the first resistor, and the constructing a second expression for an internal resistance voltage of the first internal resistance based on the independent variable comprises: constructing the second expression based on the output current, the first resistor, and the fourth expression of the independent variable.
The method according to claim 15, characterized in that the independent variable comprises an output current, the first capacitor, and the second resistor of the battery module; and the constructing a third expression for a parallel voltage of the parallel circuit based on the independent variable comprises: constructing the third expression based on the output current, the first capacitor, and the second resistor among the independent variable.
The method according to any one of claims 1 to 18, characterized in that the identification result comprises a battery capacity of the battery module; and the predicting a state of health of a battery in the battery module based on the identification result comprises: predicting the state of health of the battery in the battery module based on a value of the battery capacity in the identification result.
An electronic device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that when the computer program is executed by a processor, the battery state of health estimation method according to any one of claims 1 to 19 is implemented.

Description

This application claims priority to Chinese Patent Application No. 202310808644.2, filed on July 3, 2023, at the China Patent Office, and entitled "BATTERY STATE OF HEALTH ESTIMATION METHOD, ELECTRONIC DEVICE, AND COMPUTER-READABLE STORAGE MEDIUM," which is incorporated herein by reference in its entirety. TECHNICAL FIELD The present application relates to the field of batteries, and in particular, to a battery state of health estimation method, an electronic device, and a computer-readable storage medium. BACKGROUND With the development of electric vehicles, the service life of batteries has gained increasing attention. Battery state of health (state of health, SOH), also referred to as a battery degradation coefficient, is used to reflect an aging degree and health degree of batteries. A higher SOH value indicates a lower aging degree and a higher health degree of batteries. A lower SOH value indicates a higher aging degree and lower health degree of batteries. However, related SOH value estimation methods typically exhibit low estimation accuracy, thereby affecting users' understanding of battery conditions. TECHNICAL PROBLEM One of the objectives of embodiments of the present application is to provide a battery state of health estimation method, an electronic device, and a computer-readable storage medium, to address the issue of low estimation accuracy of the battery state of health. TECHNICAL SOLUTION To address the foregoing technical problem, the technical solutions adopted in the embodiments of the present application are as follows. According to a first aspect, a battery state of health estimation method is provided, including: obtaining historical operating condition data of a to-be-tested battery module;performing parameter identification on a battery parameter of the battery module based on the historical operating condition data to obtain an identification result; andpredicting a state of health of a battery in the battery module based on the identification result. In the embodiments of the present application, performing parameter identification on the battery parameter based on the historical operating condition data of the battery module is equivalent to an offline identification approach. Compared with a real-time identification approach, the offline identification approach can avoid instability in identification results caused by recursive solving. Additionally, since parameter identification is conducted based on the historical operating condition data of the battery module, the battery parameter of the battery module can be accurately fitted, thereby effectively improving the estimation accuracy of the battery state of health. In an implementation of the first aspect, the performing parameter identification on a battery parameter of the battery module based on the historical operating condition data to obtain an identification result includes: obtaining a mathematical model of the battery module, where the mathematical model is constructed based on the battery parameter of the battery module; andperforming parameter identification on the battery parameter of the battery module based on the historical operating condition data and the mathematical model to obtain the identification result. In the embodiments of the present application, parameter identification is performed in combination with the mathematical model to identify the parameter of the battery module, and then the state of health of the battery is predicted based on the identified parameter of the battery module. Since the mathematical model is constructed based on the battery parameter of the battery module, the mathematical model can accurately reflect the state of the battery module, providing a reliable model foundation for subsequent parameter identification. Additionally, the parameter identification method can provide accurate fitting of the battery parameter of the battery module. Combining the mathematical model and the parameter identification method can effectively improve the estimation accuracy of the battery state of health. In an implementation of the first aspect, the performing parameter identification on the battery parameter of the battery module based on the historical operating condition data and the mathematical model to obtain the identification result includes: discretizing the mathematical model based on the historical operating condition data to obtain a discretized mathematical model; andperforming parameter identification on the battery parameter of the battery module based on the historical operating condition data and the discretized mathematical model to obtain the identification result. In practical applications, sampled data of an observable parameter of the battery module are typically discrete. In the embodiments of the present application, discretizing the mathematical model enables the mathematical model to more accurately reflect the actual state of the battery module, thereby facilitating