Three-level fault diagnosis method for power battery

Abstract

The invention discloses a three-level fault diagnosis method for a power battery. The method comprises a comprehensive performance evaluation level, a fault feature extraction level and a fault mechanism analysis level, the comprehensive performance evaluation level evaluates whether current battery performance attenuation possibly triggers an internal fault mechanism or not according to sensor data, if the overall performance attenuation index of the battery reaches a threshold value, it is considered that the battery has a fault risk, and the fault feature extraction level is entered; the fault feature extraction level is used for extracting early symptoms and field fault data according to the sensor data and judging a fault mode or a potential risk of the current power battery; and thefault mechanism analysis level is used for carrying out fault diagnosis and decision based on fuzzy reasoning by utilizing the fault feature vector after confirming that the fault occurs, and determining an internal fault mechanism causing the fault. By adopting the three-level fault diagnosis method provided by the invention, the service life and the safety characteristic of the corresponding battery system are obviously improved.

Description

technical field [0001] The invention belongs to the field of new energy vehicle power lithium battery management, and in particular relates to a three-level fault diagnosis method for power batteries. Background technique [0002] The power lithium battery provides the power source for the components assembled on the new energy vehicle, and its safety directly affects the entire system. Although lithium batteries, especially lithium iron phosphate batteries, are considered to be a relatively safe energy storage device, a series of practical problems such as low-temperature charging, overcharge and overdischarge conditions will still bring safety hazards to battery applications. . Therefore, it is necessary to establish a battery management system that must have evaluation and diagnosis functions for battery safety performance, so as to monitor the health status of the battery in real time, evaluate the potential failure risk of the battery, diagnose and deal with sudden bat...

AI Technical Summary

Problems solved by technology

The current lithium battery fault diagnosis system is often based on sensor data, such as the invention patent "A Power Battery Fault Detection Method and System Based on High-dimensional Data Diagnosis" with the application number CN201811336819.X. Carry out clustering processing, and then identify abnormal values ​​to achieve the purpose of fault diagnosis. However, such a diagnosis method lacks the analysis of the fault mechanism, cannot effectively identify whether the battery is normally aging or has a fault, and lacks a mechanism analysis of the internal root cause of the fault. As a result, the diagnostic method cannot effectively detect the early symptoms of battery failure, and then realize the early warning of failure

Images