A construction method and device for a prediction model of residual cycle life of lithium battery

Abstract

The invention provides a construction method and device for a prediction model of residual cycle life of lithium battery. The method includes the following steps: the feature data of N sample batteries is obtained, the feature data includes m charging feature elements, n discharging feature elements and sample battery capacitance, wherein N is a round figure higher than 1, m and n are the integerhigher than 0, feature fusion processing is performed on the feature data of N sample batteries by adopting a self-coding model, the feature data after fusion is obtained, the feature data after fusion includes L feature elements, L is the integer higher than 0, L is less than m+n+1, training is performed on a depth nerve network DNN model by adopting the feature data after fusion of N sample batteries, and the prediction model of the residual cycle life of the lithium battery is obtained. The construction of the prediction model of the residual cycle life of the lithium battery with higher accuracy is realized, the method is more simple and convenient when the residual cycle life of the lithium battery is predicted, and the accuracy of the obtained prediction result is higher.

Description

technical field [0001] The invention relates to the technical field of lithium batteries, in particular to a method and device for constructing a lithium battery remaining cycle life prediction model. Background technique [0002] Lithium batteries rely on lithium ions to shuttle back and forth between the positive and negative electrodes to realize the conversion and storage of electric energy to complete the charging and discharging process of lithium batteries. In an ideal state, this charging and discharging process can be carried out indefinitely, but in actual use, the chemical reaction inside the lithium battery will consume the materials such as electrolyte, separator and battery polarity, so that the internal lithium battery cannot be reversed. The decline of the battery affects the life of the battery. The life of lithium batteries includes three types: storage life, service life and cycle life. Among them, the remaining cycle life of lithium batteries is an impor...

AI Technical Summary

Problems solved by technology

[0004] However, due to the highly complex chemical reactions inside the lithium battery, the state data required for building the model are usually difficult to detect and collect. Therefore, it is difficult to establish a lithium battery degradation prediction model with high accuracy, so it is not convenient to carry out the remaining cycle life of the lithium battery. predict

Images