Useful Energy Prediction Model of a Lithium-ion Cell Operating on Various Duty Cycles

Abstract

The paper deals with\nthe subject of the prediction of useful energy during the cycling of a\nlithium-ion cell (LIC), using machine learning-based techniques. It was\ndemonstrated that depending on the combination of cycling parameters, the\nuseful energy (RUEc) that\ncan be transfered during a full cycle is variable, and also three different\ntypes of evolution of changes in RUEc\nwere identified. The paper presents a new non-parametric RUEc prediction model based on Gaussian process\nregression. It was proven that the proposed methodology enables the RUEc prediction for LICs discharged,\nabove the depth of discharge, at a level of 70% with an acceptable error, which\nis confirmed for new load profiles. Furthermore, techniques associated with\nexplainable artificial intelligence were applied, for the first time, to\ndetermine the significance of model input parameters – the variable importance\nmethod – and to determine the quantitative effect of individual model\nparameters (their reciprocal interaction) on RUEc – the accumulated local effects model of the first\nand second order. Not only is the RUEc\nprediction methodology presented in the paper characterised by high prediction\naccuracy when using small learning datasets, but it also shows high application\npotential in all kinds of battery management systems.