Algorithms for Monitoring Rechargeable Batteries

Abstract

The battery management system is a fundamental part of the battery-based storage system due to allow use the energy in a safety way. Ones of the main tasks of this system includes the estimation and prediction of the state of charge, the aging and the remaining energy all from voltage, current and temperature measurement.
In this work, we present algorithms to perform some of these estimations using an electrochemical model based on two processes that allow the battery converts the chemical energy intothe electrical energy. In particular, the prediction of the remaining time and the estimation of the state of charge and other additionally variable that models the rate-capacity effect were carried on. The proposed algorithms were tested in a battery pack formed by four commercial Li-ion batteries connected electrically in series.
The results shows that the averaged error in the prediction of the remaining time was 4.2 minutes considering discharges lesser and equal to 1 hour of duration.