SOC Estimation of battery in Hybrid Vehicle Using Adaptive Neuro-Fuzzy Technique
P. R. Shabarish, D. V. S. Aditya, V. V. S. S Phani Pavan, P. V. Manitha
Abstract
The advent of Electric vehicles is a major step in building a sustainable energy model. Battery management system is the crux of a hybrid and electric vehicles in market. Online estimation of State of charge is proved to be a challenge in battery management for an electric vehicle. Various Kalman filters with combination of better estimating methods increases accuracy in estimation of state of charge. Improper knowledge of battery management of batteries which leads to performance reduction and it can even lead to practical divergence. Hence adaptive estimation of State of Charge of batteries plays important role in battery management. This paper does a comparative analysis of Extended Kalman filter, Unscented Kalman Filter and Adaptive neuro fuzzy inference system. A new hybrid method combining extended Kalman filter and adaptive neuro fuzzy inference system is been proposed. Adaptive neuro fuzzy inference system, a data driven approach, is given the input from extended Kalman filter for minimizing the error in the estimate value of State of charge. Results show that the error decreased by 70% by using proposed hybrid method over extended Kalman filter to estimate state of charge.