Nonlinear estimator-based state of charge estimation for lithium titanate oxide battery in energy storage systems

Fuel savings, energy savings, and reduction of CO2 emissions are the key requirements in electric and hybrid electric vehicles (EVs and HEVs). These requirements are achieved through the usage of start-stop systems, high-rate energy boost during acceleration, and recovery of the energy during braking. Unfortunately, the commonly used lithium-ion batteries are not a sufficient solution, although their performance is enhanced by supercapacitors. This hybrid system used is only a temporary solution. Therefore, lithium-titanate-oxide batteries (Li4Ti5O12-LTO), show high-rate discharging and charging performance, high power capability, excellent cycle life, and improved cycle stability at wide-rate temperatures and current rates are promising candidates for HEV and EV applications. There is a need to monitor the state of charge (SoC) for the reliability, performance, and safety of LTO batteries with the help of a battery management system. However, the conventional SoC estimation methods, such as open circuit voltage, Coulomb counting, artificial neural networks, fuzzy logic, and linear Kalman filter algorithm become insufficient. In this paper, a new nonlinear approach for the SoC estimation of an LTO battery is presented. The approach combines the static battery model and the sigma point Kalman filter (SPKF) algorithm while adapting to both constant load and dynamic load. The results show that the SPKF algorithm successfully estimated the SoC of the LTO battery under various initial SoC initialization values. It is also supported by various metric errors calculated.