CN-115629313-B - Pulse current prediction method and device for lithium ion battery and computer equipment

CN115629313BCN 115629313 BCN115629313 BCN 115629313BCN-115629313-B

Abstract

The application provides a pulse current prediction method, a device and computer equipment of a lithium ion battery, wherein the method comprises the steps of obtaining pulse curve data generated by discharging the lithium ion battery under the limit of at least three preset pulse currents, and obtaining variation trend information of each pulse curve data; the method comprises the steps of establishing an initial pulse curve prediction model based on various change trend information, analyzing various pulse curve data through the initial pulse curve prediction model to determine parameter values of the target model parameters to obtain a target pulse curve prediction model with known values, and carrying out discharge pulse prediction on a lithium ion battery through the target pulse curve prediction model to obtain the maximum pulse current which is about to be generated after the lithium ion battery is discharged under the limit of preset pulse time. By adopting the method, the pulse current testing efficiency of the lithium ion battery can be improved, and the accurate current prediction of the lithium ion battery under different pulse times can be realized.

Inventors

LIU YAOJUN
YU WENJUN
ZHAI XIUMEI
HE YONGWU
XU ZHONGLING

Assignees

欣旺达电动汽车电池有限公司

Dates

Publication Date: 20260508
Application Date: 20221020

Claims (9)

1. The pulse current prediction method of the lithium ion battery is characterized by comprising the following steps of: Based on a preset test temperature, a charge state and a cut-off voltage, performing discharge pulse test on the lithium ion battery to obtain test data of the lithium ion battery, wherein the test data of the lithium ion battery is changed along with time from an initial voltage to the cut-off voltage under the limit of at least three preset pulse currents, and pulse curve data generated by discharge is obtained; According to the pulse curve data, obtaining a voltage drop rate, and analyzing the voltage drop rate to obtain change trend information of the pulse curve data, wherein the change trend information comprises a first change trend of short-time rapid drop, a second change trend of slow drop and a third change trend of near-end rapid drop; constructing an initial pulse curve prediction model based on the change trend information, wherein the initial pulse curve prediction model comprises target model parameters with unknown numerical values; Analyzing each pulse curve data through the initial pulse curve prediction model to determine the parameter value of the target model parameter, and obtaining a target pulse curve prediction model with known value; And predicting discharge pulse of the lithium ion battery through the target pulse curve prediction model to obtain the maximum pulse current which is about to be generated after the lithium ion battery discharges under the limit of preset pulse time.
2. The method of claim 1, wherein, The change trend information includes a first change trend, a second change trend and a third change trend, and the constructing an initial pulse curve prediction model based on each change trend information includes: Constructing a first prediction model comprising the target model parameter and a first power function based on the first variation trend and the second variation trend, wherein the base number of the first power function is a first pulse time smaller than or equal to the minimum moment of the voltage variation rate, and the minimum moment of the voltage variation rate is determined according to the second derivative of the pulse curve data; constructing a second prediction model comprising the target model parameter, a second power function and an exponential function based on the third variation trend, wherein the base of the second power function is a second pulse time which is larger than the minimum moment of the voltage variation rate, and the exponent of the exponential function is determined according to the difference between the second pulse time and the minimum moment of the voltage variation rate; and taking the first prediction model and the second prediction model as the initial pulse curve prediction model.
3. The method of claim 1, wherein, The initial pulse curve prediction model comprises a first prediction model and a second prediction model, and the analyzing each pulse curve data through the initial pulse curve prediction model to determine the parameter value of the target model parameter, so as to obtain a target pulse curve prediction model with a known value, and the method comprises the following steps: performing second-order derivative processing on each pulse curve data to obtain a corresponding voltage value when a derivative result is zero, wherein the voltage value is used as a target voltage value; obtaining the average value of each target voltage value to obtain a curve segment voltage value; fitting and analyzing each pulse curve data which is larger than or equal to the curve segment voltage value through the first prediction model so as to determine the parameter value of the target model parameter contained in the first prediction model; fitting and analyzing each pulse curve data smaller than the curve segment voltage value through the second prediction model to determine a parameter value of a target model parameter contained in the second prediction model; And taking an initial pulse curve prediction model with known values as the target pulse curve prediction model.
4. The method of claim 3, wherein, The target model parameters include a first model parameter, a second model parameter and a third model parameter, and the fitting analysis of each pulse curve data greater than or equal to the curve segment voltage value by the first prediction model is performed to determine the parameter value of the target model parameter included in the first prediction model, and the method includes: fitting and analyzing each pulse curve data which is larger than or equal to the curve segment voltage value through the first prediction model to obtain parameter values of the first model parameter, the second model parameter and the third model parameter which are related to each preset pulse current; performing exponential function fitting processing on the parameter values of the first model parameters and preset pulse current to obtain a first function expression of the first model parameters, and Performing mean value calculation on the parameter values of the second model parameters to obtain second parameter values of the second model parameters, and Performing linear fitting processing on the parameter values of the third model parameters and preset pulse currents to obtain third function expressions of the third model parameters; and taking the first function expression, the second parameter value and the third function expression as parameter values of target model parameters contained in the first prediction model.
5. The method of claim 4, wherein, The target model parameters further include a fourth model parameter and a fifth model parameter, and the fitting analysis of each pulse curve data smaller than the curve segment voltage value by the second prediction model to determine the parameter value of the target model parameter included in the second prediction model includes: fitting and analyzing each pulse curve data smaller than the curve segment voltage value through the second prediction model to obtain parameter values of the fourth model parameter and the fifth model parameter associated with each preset pulse current; Performing exponential function fitting processing on the parameter values of the fourth model parameters and preset pulse current to obtain fourth function expressions of the fourth model parameters, and Performing linear fitting processing on the parameter values of the fifth model parameters and preset pulse currents to obtain fifth function expressions of the fifth model parameters; And taking the first function expression, the second parameter value, the third function expression, the fourth function expression and the fifth function expression as parameter values of target model parameters contained in the second prediction model.
6. The method of claim 4, wherein, After the determining the parameter values of the target model parameters contained in the first prediction model, the method further comprises: Determining a pulse time corresponding to the target voltage value in each pulse curve data as a time with the minimum voltage change rate; Extracting constant parameter values in the first function expression and the third function expression; and determining a time function expression of the moment with the minimum voltage change rate according to the curve segment voltage value, the constant parameter value and the second parameter value, wherein the time function expression is used for carrying out discharge pulse prediction on the lithium ion battery by combining the target pulse curve prediction model.
7. A pulse current prediction apparatus for a lithium ion battery, comprising: The system comprises a data acquisition module, a data processing module and a control module, wherein the data acquisition module is used for carrying out discharge pulse test on a lithium ion battery based on preset test temperature, charge state and cut-off voltage to acquire test data of the lithium ion battery, wherein the test data of the lithium ion battery is changed along with time after initial voltage is discharged to the cut-off voltage under the limit of at least three preset pulse currents, so as to acquire pulse curve data generated by discharge, and according to the pulse curve data, the voltage dropping rate is acquired, and the change trend information of the pulse curve data is acquired by analyzing the voltage dropping rate; the model construction module is used for constructing an initial pulse curve prediction model based on the change trend information, wherein the initial pulse curve prediction model comprises target model parameters with unknown numerical values; the curve analysis module is used for analyzing each pulse curve data through the initial pulse curve prediction model so as to determine the parameter values of the target model parameters and obtain a target pulse curve prediction model with known values; And the current prediction module is used for predicting the discharge pulse of the lithium ion battery through the target pulse curve prediction model to obtain the maximum pulse current which is about to be generated after the lithium ion battery is discharged under the limit of preset pulse time.
8. A computer device, comprising: One or more processors; A memory; And one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the method of pulse current prediction for a lithium ion battery of any of claims 1-6.
9. A computer-readable storage medium comprising, A computer program stored thereon, the computer program being loaded by a processor to perform the steps in the pulse current prediction method of a lithium ion battery as defined in any one of claims 1 to 6.

Description

Pulse current prediction method and device for lithium ion battery and computer equipment Technical Field The application relates to the technical field of lithium ion batteries, in particular to a pulse current prediction method and device for a lithium ion battery and computer equipment. Background Along with the popularization of new energy automobiles, a power lithium ion battery has entered a rapid development stage. The battery pulse charge and discharge capability directly determines the performance of the whole vehicle under different working conditions, so that the battery pulse charge and discharge capability and the pulse discharge current prediction are necessary. At present, the existing prediction technology for the pulse charge and discharge capacity and the pulse discharge current of the lithium battery generally adopts a mode of testing the power performance in various states point by point or adopts a mode of establishing an electrochemical and solid heat transfer coupling model of a battery core to calibrate a pulse model, although the required result can be obtained, a large amount of time and testing resources are still consumed in practice due to complex flow and a large amount of basic parameters, and the prediction requirement of the performance of the lithium ion battery cannot be met. Therefore, the existing lithium battery pulse current prediction technology has the technical problem of low test efficiency. Disclosure of Invention Based on the above, it is necessary to provide a method, a device and a computer device for predicting the pulse current of a lithium ion battery, so as to realize accurate prediction of the maximum current of the lithium ion battery under different pulse times by analyzing the relationship between the pulse voltage, the pulse current and the pulse time of the lithium ion battery, so that the test efficiency of the pulse current of the lithium ion battery can be effectively improved without point-by-point testing or a large number of basic parameters. In a first aspect, the present application provides a method for predicting pulse current of a lithium ion battery, including: Acquiring pulse curve data generated by discharging the lithium ion battery under the limit of at least three preset pulse currents, and obtaining variation trend information of each pulse curve data; constructing an initial pulse curve prediction model based on the change trend information, wherein the initial pulse curve prediction model comprises target model parameters with unknown values; Analyzing each pulse curve data through an initial pulse curve prediction model to determine parameter values of target model parameters, and obtaining a target pulse curve prediction model with known values; And predicting discharge pulse of the lithium ion battery through a target pulse curve prediction model to obtain the maximum pulse current which is about to be generated after the lithium ion battery discharges under the limit of preset pulse time. In some embodiments of the application, pulse curve data generated by discharging the lithium ion battery under at least three preset pulse current limits are obtained to obtain variation trend information of the pulse curve data, wherein the variation trend information comprises a first variation trend of short-time rapid drop, a second variation trend of slow drop and a third variation trend of temporary drop, wherein the discharge pulse test is performed on the lithium ion battery based on preset test temperature, charge state and cut-off voltage to obtain test data of the lithium ion battery, the test data of the voltage variation with time is obtained after the initial voltage is discharged to the cut-off voltage under the preset pulse current limits, the voltage drop rate is obtained according to the pulse curve data, and the variation trend information is obtained by analyzing the voltage drop rate. In some embodiments of the application, the change trend information comprises a first change trend, a second change trend and a third change trend, an initial pulse curve prediction model is built based on the change trend information, the method comprises the steps of building a first prediction model comprising target model parameters and a first power function based on the first change trend and the second change trend, wherein the base of the first power function is a first pulse time smaller than or equal to the minimum moment of a voltage change rate, the minimum moment of the voltage change rate is determined according to the second derivative of pulse curve data, building a second prediction model comprising target model parameters, a second power function and an exponential function based on the third change trend, wherein the base of the second power function is a second pulse time larger than the minimum moment of the voltage change rate, the exponent of the exponential function is determined according to the difference between t