Lithium ion battery remaining life prediction method based on fusion of gating circulation unit neural network and Kalman filtering model

Abstract

The invention discloses a lithium ion battery remaining life prediction method based on fusion of a gating circulation unit neural network and a Kalman filtering model, and relates to the technical field of lithium ion battery health state detection. The objective of the invention is to solve the problems of poor fitting capability and low adaptability to different working states in a nonlinear degradation process of an existing lithium ion battery residual life prediction method based on a fusion model. According to the method, through establishing a GRU-RNN deep network model, the lithium ion battery capacity degradation characteristics are extracted by using the strong characteristic extraction capability of the GRU deep learning model on the time sequence, so that a more accurate battery capacity prediction model is obtained, and finally, noise is reduced and a more accurate prediction value is obtained through a KF filtering method.

Description

technical field [0001] The invention belongs to the technical field of lithium-ion battery health state detection, and in particular relates to battery life prediction technology. Background technique [0002] At present, the methods for predicting the Remaining Useful Life (RUL) of lithium-ion batteries are roughly divided into two categories: physical model-based and data-driven model. Among them, the data-driven method does not need to clarify the degradation mechanism of the battery, so the related research is more in-depth. Data-driven methods include one type of statistical data-driven method based on statistical filtering and another type of data-driven method based on machine learning methods. The lithium-ion battery degradation model used in the data-driven method based on statistical filtering is relatively simple, and has poor adaptability to life prediction problems of different types of batteries and different usage conditions. The data-driven method based on ...

AI Technical Summary

Problems solved by technology

[0003] The present invention aims to solve the problems of poor fitting ability of nonlinear degradation process and low adaptability to different working states in the existing remaining life prediction method of lithium-ion battery based on fusion model, and now provides a method based on Gated Recurrent Unit (Gated Recurrent Unit , GRU) neural network and Kalman filter (Kalman Filter, KF) model fusion lithium-ion battery remaining life prediction method

Images