A Regression-Based Li-Ion Battery State of Charge Estimation

Improvement in lifetime, reliability of the battery will show various results proper functioning of the system and a reduction in maintenance costs. The life of a Li-ion battery can be extended by avoiding over charge and discharge. The battery powered system is designed in such a way, where it requires charge-cycle counting, battery monitoring, remaining run-time information. The dynamic complexity and nonlinear battery behavior create a need for development of model for state of charge estimation. The state of charge estimation algorithm using Artificial Neural Network for Lithium-ion battery to ensure that optimum use is made of the energy inside the battery powered system and battery damage is prevented. In the present work, a comparison between a polynomial model and data driven neural network architecture has been made for SOC estimation. DC resistance and impedance-based regression has been studied to estimate suitable attribute.