State of charge estimation for batteries in HEV using locally linear model tree (LOLIMOT)

State of charge (SOC) estimation is one of the essential segments of hybrid electrical vehicles (HEV). By monitoring of SOC, we can optimize the consumption of fuel and decrease the pollution of air, in HEV. In this works considers the state of charge (SOC) estimation problem for lead-acid batteries for use in HEV. In this paper, the online state of charge estimation is worked using a locally linear model tree (LOLIMOT) which is a Nero-fuzzy network. The training data of LOLIMOT contain measured voltage, current and SOC data in different temperature in which voltage, current and temperature are used as inputs and SOC is used as output. The SOC estimation results using LOLIMOT is compared with the results of ANFIS [1] in this paper.