Online SOC Estimation of Li-FePO4 Batteries through a New Fuzzy Rule-Based Recursive Filter with Feedback of the Heat Flow Rate

A new method for State of Charge estimation of LiFePO4 batteries through recursive filtering is presented. A fuzzy rule-based system is used to compute the nonlinear gain of the filter, and computational intelligence techniques are used to evolve the definition of the rule base. The estimation of the charge is based on a novel battery model that accurately predicts battery voltage and temperature. It will be shown that feeding back the difference between modelled and measured heat flow rates noticeably shortens the time needed to reach a correct estimation of the battery charge when initial conditions are unknown. An empirical study has been carried over data gathered at the Battery Laboratory at Oviedo University. The results show improvements in speed and stability. An accuracy of 95% was reached five times faster than linear filters with feedback of the cell voltage error.