CN-121973629-A - Power battery internal short circuit early warning method and device and vehicle

CN121973629ACN 121973629 ACN121973629 ACN 121973629ACN-121973629-A

Abstract

The application relates to the technical field of battery safety management, in particular to a method and a device for early warning of internal short circuit of a power battery and a vehicle, wherein the method comprises the steps of obtaining charging data, standing data and discharging data of the power battery; the method comprises the steps of calculating first characteristics of all monomers in the power battery according to charging data, calculating second characteristics of all monomers in the power battery according to standing data, calculating third characteristics of all monomers in the power battery according to any one of charging data, standing data and discharging data, calculating comprehensive risk indexes of internal short circuits of the power battery according to the first characteristics, the second characteristics and the third characteristics, and early warning the internal short circuits of the power battery according to the comprehensive risk indexes. Therefore, the problems of high mechanism driving path cost, data driving path model dependence on data scale and cloud computing power, limited instantaneity and generalization capability, insufficient early risk identification, early warning lag and the like in the related technology are solved.

Inventors

CAO RANRAN
QIN ZHIDONG
YAN KANGKANG
ZHANG YANCHAO

Assignees

北京卡文新能源汽车有限公司

Dates

Publication Date: 20260505
Application Date: 20260205

Claims (10)

1. The power battery internal short circuit early warning method is characterized by comprising the following steps of: Acquiring charging data, standing data and discharging data of a power battery; calculating first characteristics of all monomers in the power battery according to the charging data; Calculating second characteristics of all monomers in the power battery according to the standing data; Calculating third characteristics of all monomers in the power battery according to any one of the charging data, the standing data and the discharging data; And calculating a comprehensive risk index of the internal short circuit of the power battery according to the first feature, the second feature and the third feature, and carrying out early warning on the internal short circuit of the power battery according to the comprehensive risk index.
2. The method for early warning of internal short circuit of a power battery according to claim 1, wherein the calculating the first characteristic of all the cells in the power battery according to the charging data comprises: Identifying a state of charge of the power battery in the charge data; intercepting a first voltage and time data segment of a single body from the charging data according to the state of charge; and calculating first characteristics of all the single units in the power battery according to the first voltage and the time data period.
3. The method for early warning of internal short circuit of a power battery according to claim 2, wherein the calculating the first characteristic of all the cells in the power battery according to the first voltage and the time data segment includes: performing curve fitting on the first voltage and the time data segment; Extracting slope characteristics and distortion characteristics from the curve fitting result; And generating a first characteristic of the corresponding monomer according to the slope characteristic and the distortion characteristic.
4. The method of claim 1, wherein calculating the second characteristics of all cells in the power cell based on the rest data comprises: identifying the starting time of the power battery entering the standing working condition in the standing data; Intercepting a second voltage and time data segment of the single body from the rest data according to the starting time; And calculating second characteristics of all the single bodies in the power battery according to the second voltage and the time data period.
5. The method of claim 4, wherein calculating the second characteristic of all cells in the power battery according to the second voltage and the time data segment comprises: performing curve fitting on the second voltage and the time data segment; Extracting a fast relaxation time constant, a slow relaxation time constant and a time constant ratio from the curve fitting result; generating a second characteristic of the corresponding monomer from the fast relaxation time constant, the slow relaxation time constant, and the time constant ratio.
6. The power cell internal short warning method according to claim 1, wherein the calculating the third characteristic of all the cells in the power cell based on any one of the charge data, the rest data, and the discharge data includes: acquiring voltage data of each single body from any one of the charging data, the standing data and the discharging data in a preset time window; Calculating standard deviation of corresponding monomers according to the voltage data, calculating a first derivative of the standard deviation with respect to time, and determining the change rate of the standard deviation according to the first derivative; and calculating the third characteristics of all the single bodies in the power battery according to the change rate of the standard deviation.
7. The method for early warning of internal short circuit of a power battery according to claim 6, wherein the calculating the third characteristic of all the cells in the power battery according to the rate of change of the standard deviation comprises: Performing moving average filtering on the change rate of the standard deviation to obtain a smooth trend signal; Extracting trend direction characteristics, trend intensity characteristics and trend persistence characteristics from the trend signals; and generating a third feature of the corresponding monomer according to the trend direction feature, the trend intensity feature and the trend persistence feature.
8. The power cell internal short warning method according to claim 1, wherein calculating an integrated risk index of the internal short of the power cell from the first feature, the second feature, and the third feature comprises: Obtaining a reference value of each monomer; normalizing the first feature, the second feature, and the third feature according to the reference value; calculating the abnormal degree of each of the normalized first feature, the normalized second feature and the normalized third feature, and obtaining the weight of each feature; and calculating the comprehensive risk index according to the degree of abnormality and the weight of each feature.
9. An internal short circuit early warning device for a power battery is characterized by comprising: the acquisition module is used for acquiring charging data, standing data and discharging data of the power battery; the first characteristic module is used for calculating first characteristics of all monomers in the power battery according to the charging data; the second characteristic module is used for calculating second characteristics of all monomers in the power battery according to the standing data; the third characteristic module is used for calculating third characteristics of all monomers in the power battery according to any one of the charging data, the standing data and the discharging data; And the early warning module is used for calculating the comprehensive risk index of the internal short circuit of the power battery according to the first feature, the second feature and the third feature, and carrying out early warning on the internal short circuit of the power battery according to the comprehensive risk index.
10. A vehicle comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor executing the program to implement the method for warning of a short circuit in a power cell of any one of claims 1-8.

Description

Power battery internal short circuit early warning method and device and vehicle Technical Field The application relates to the technical field of battery safety management, in particular to a power battery internal short circuit early warning method, a device and a vehicle. Background Internal short circuit of lithium ion battery is a main cause of thermal runaway, and is important to monitor and prevent internal short circuit and identify battery abnormality in time. In the related technology, the mechanism driving path depends on hardware to observe electrochemical characteristics, but the cost is higher, the data driving path depends on an artificial intelligent model to carry out fault diagnosis, the data scale and cloud computing power are more dependent, the real-time performance and generalization capability are limited, and the early warning lag or conclusion reliability is low. Disclosure of Invention The application provides a power battery internal short circuit early warning method, a power battery internal short circuit early warning device and a vehicle, and solves the problems that a mechanism driving path is high in cost, a data driving path model cannot be interpreted, instantaneity and reliability are insufficient and the like in the related technology. An embodiment of the first aspect of the application provides a power battery internal short-circuit early warning method, which comprises the following steps of obtaining charging data, standing data and discharging data of a power battery, calculating first characteristics of all monomers in the power battery according to the charging data, calculating second characteristics of all monomers in the power battery according to the standing data, calculating third characteristics of all monomers in the power battery according to any one of the charging data, the standing data and the discharging data, calculating comprehensive risk indexes of the internal short-circuit of the power battery according to the first characteristics, the second characteristics and the third characteristics, and early warning the internal short-circuit of the power battery according to the comprehensive risk indexes. Alternatively, in one embodiment of the application, the first characteristics of all cells in the power battery are calculated from the charge data, including identifying a state of charge of the power battery in the charge data, intercepting a first voltage and time data segment of the cells from the charge data based on the state of charge, and calculating the first characteristics of all cells in the power battery based on the first voltage and time data segment. Alternatively, in one embodiment of the application, the first characteristics of all the single cells in the power battery are calculated according to the first voltage and the time data segment, wherein the method comprises the steps of performing curve fitting on the first voltage and the time data segment, extracting slope characteristics and distortion characteristics from curve fitting results, and generating the first characteristics of the corresponding single cells according to the slope characteristics and the distortion characteristics. Optionally, in one embodiment of the application, the second characteristics of all the monomers in the power battery are calculated according to the standing data, wherein the second characteristics comprise the steps of identifying the starting time of the power battery entering the standing working condition in the standing data, intercepting the second voltage and time data segments of the monomers according to the starting time from the standing data, and calculating the second characteristics of all the monomers in the power battery according to the second voltage and time data segments. Optionally, in one embodiment of the application, calculating the second characteristics of all the cells in the power battery according to the second voltage and time data segments comprises performing curve fitting on the second voltage and time data segments, extracting a fast relaxation time constant, a slow relaxation time constant and a time constant ratio from the curve fitting result, and generating the second characteristics of the corresponding cells according to the fast relaxation time constant, the slow relaxation time constant and the time constant ratio. Optionally, in one embodiment of the application, calculating the third characteristics of all the cells in the power battery according to any one of the charging data, the standing data and the discharging data comprises obtaining voltage data of each cell from any one of the charging data, the standing data and the discharging data in a preset time window, calculating standard deviation of the corresponding cell according to the voltage data, calculating first derivative of the standard deviation with respect to time, determining the change rate of the standard deviation according to the f