Advanced lead-acid battery models for the state-of-charge estimation in an isolated microgrid

Abstract

In order to compare and improve battery energy storage system (BESS) models, this work exhibits a comparative study of the Coppeti, the Coulomb counting, and the Thevenin based models of the State of Charge (SoC) for lead-acid batteries. Additionally, a multi-model estimator that merges and exploits the benefits of aforementioned models is proposed by using an extended Kalman filter (EKF). The performance of the different SoC models are evaluated by means of simulations in a economic dispatch problem of an isolated microgrid with an Energy Management System (EMS). The Comparative study of the SoC models is done using real operation data from the Huatacondo's microgrid with an EMS based on a rolling horizon strategy. As shown in the results, the operating costs of the microgrid are considerably reduced when the multi-model estimator is used, showing a direct impact in the economical costs of the system.