Development of an Aging Model for Lithium Titanate Oxide Batteries

Abstract

Lithium-ion batteries are becoming the go to option for many automotive industries and applications, partly because of its high-power density and high efficiency as compared to other traditional batteries. The surging interest in electric vehicles (EVs) has resulted in a heightened need for advanced battery management systems (BMS) that can reliably determine the state of health (SOH) of batteries. The SOH is a vital parameter that defines the performance and durability of Lithium-ion batteries. Precise assessment of these parameters is crucial for maximizing battery usage and enhancing the overall effectiveness and dependability of EVs. Despite its high applications, it still put in to question the batteries aging lifespan. Several technological approaches with the very best innovation available are being implemented to reduce the battery aging lifetime, which will directly have an impact on the cost in the energy storage market. Precisely determining the state of a Li-ion batteries under real-world driving conditions is a demanding challenge due to the variability of driving patterns, which can lead to considerable fluctuations in battery performance. To tackle these issues, a semi-empirical based modelling technique is implemented. The objective of this thesis is to describe an aging model for lithium titanate oxide (LTO) batteries, by investigating systematically its cycling and calendar aging processes. The model will give a detailed understanding of the state of health (SOH), that is the capacity loss and resistance growth of the batteries under different equivalent full cycle, state of charge, depth of discharge, current rate and temperature conditions. A model equation and parameters will be formulated based on the available measuring dataset. A prediction of the End-of-Life of the batteries will be made using a machine learning algorithm by examining their capacity loss and the degradation processes occurring during use and storage. A precise prediction facilitated by the model can contribute to boosting EVs in the market and ensuring the safe utilization of the LTO batteries.