Healthy lithium nickel manganese cobalt oxide (NMC) battery using machine-learning method

Abstract

Nowadays, batteries are experiencing fast development and have a wide range of applications. However, due to their complex characteristics, it is still challenging for battery health estimation. The purpose of this project was to determine the correlations between the parameters and battery health. Despite the fact that there are some methods for predicting battery health such as physics-based models and empirical models. While machine-learning-based method has good accuracy in estimation of battery health management. This project use machine-learning techniques to predict the health state of the Lithium Nickel Manganese Cobalt Oxide (NMC) battery. A variety of circuit models for battery modelling that are equivalent have been given and analysed. In addition, a battery modelling framework is presented for estimating Lithium-Ion Battery modelling parameters. The efficiency of the model will be demonstrated using NMC battery pack experiment data. Besides, this project also analyse the relationships between the parameters and the health state of battery.