Comparison of Stochastic capacity estimation tools applied on remaining useful life prognosis of Lithium ion batteries

##plugins.themes.bootstrap3.article.main##

##plugins.themes.bootstrap3.article.sidebar##

Published Jun 30, 2018
Mikel Arrinda

Abstract

Lithium ion batteries suffer a performance decrease in normal use, which leads to an end to the usability of those batteries under the designed application. Accurate estimation of the useful lifetime of the batteries is important for achieving high energy efficiency and cost reduced designs. However, the aging of a lithium ion battery has a non-linear behavior and the models available nowadays are far from completely describing this aging behavior. Thus, capacity estimation models and tools are applied in order to improve the accuracy of current models. In this sense, this paper evaluates different stochastic tools applied to typical aging models so as to increase the accuracy of the useful life estimations at different aging states. The chosen aging models are semi-empirical models based on power laws, exponentials and polynomials which represent in a simple way the aging behavior of lithium ion cells under concrete conditions. The degradation data used in this paper comes from LiFePO4-graphite and LiNi0.8Co0.1Al0.1O2-graphite aged cells. Then, the estimations are improved applying Particle Filter (PF) and Gaussian Process Regression (GPR) based stochastic tools, which adds uncertainty evaluation and historical evaluation to the model. The benefits and limitations of different configuration of those tools will be discussed in terms of the needed computer resources and future prediction accuracy improvement.

How to Cite

Arrinda, M. (2018). Comparison of Stochastic capacity estimation tools applied on remaining useful life prognosis of Lithium ion batteries. PHM Society European Conference, 4(1). https://doi.org/10.36001/phme.2018.v4i1.369
Abstract 436 | PDF Downloads 483

##plugins.themes.bootstrap3.article.details##

Keywords

modeling and simulation

Section
Technical Papers