A kind of lithium-ion battery fault diagnosis method

Abstract

The invention discloses a lithium-ion battery fault diagnosis method, comprising the following steps: S 1. Obtain the battery data when the lithium-ion battery does not fail and when various types of failures occur; S 2. Use the signal noise reduction model to step S 1. The obtained data is subjected to noise reduction processing; S 3. Through the calculation of the feature extraction model, the characteristic parameters representing the chemical reactions of different frequencies inside the lithium-ion battery are obtained; S 4. Calculate the safety threshold of the lithium-ion battery; S 5. Determine the type of fault corresponding to the alarm according to the safety threshold, and establish a lithium-ion battery fault diagnosis model; S 6. Obtain the charge-discharge cycle battery data during the use of the lithium-ion battery to be diagnosed, denoise the data and calculate the characteristic parameter curve, and compare the similarity between the characteristic parameter curve and the reference parameter curve to obtain the similarity; S 7. Enter the degree of similarity into the step S 5 lithium-ion battery fault diagnosis model, if the threshold value is reached, an alarm signal of the corresponding fault type will be issued.

Description

technical field [0001] The invention belongs to the field of battery fault diagnosis, and in particular relates to a lithium ion battery fault diagnosis method. Background technique [0002] In recent years, secondary lithium-ion batteries have been widely used in C products, electric vehicles, and energy storage due to their high energy density, long service life, low self-discharge rate, and no memory effect. Especially with the increasingly prominent environmental protection issues, the use of lithium-ion batteries in the field of electric vehicles shows an almost linear growth trend. However, due to the instability and abuse of lithium-ion batteries and the more stringent technical requirements for thinner batteries and higher energy density due to technological development, frequent safety accidents have attracted more and more attention. Therefore, it is imminent to invent a method that can detect power battery faults in real time and predict the results accurately. ...

AI Technical Summary

Problems solved by technology

[0003] Lithium-ion battery is a complex nonlinear system that changes in real time. The internal chemical reaction mechanism is complex, and the external performance is affected by changes in various parameters. It is very complicated to establish an electrochemical mechanism model for prediction, and it is difficult to achieve real-time prediction.

In recent years, neural networks have been widely used in power battery fault diagnosis, but there are defects such as complex structure, huge amount of calculation, poor interpretability, and this method is purely based on data-driven, without considering the internal microscopic chemical reaction and external macroscopic performance when a fault occurs The relationship between the prediction results is poor

Images