<![CDATA[Optimization of battery usage, including VRLA battery which is often used for large amounts of energy storage at low prices, is usually pursued by implementing Battery Management System (BMS). To carry out BMS, information about the condition of charge and health is needed. The State of Charge (SoC) is defined as the ratio of the current remaining capacity of the battery to the capacity of the battery before discharge, while the State of Health (SoH) is the ratio between the measured full capacity of a battery to its nominal capacity when it is still in a new condition. SoC and SoH estimation can be held indirectly by using the voltage and current at the battery terminals. This study uses the Coulomb Counting (CC) method and Support Vector Regression (SVR) to estimate SoC and SoH of VRLA batteries which are used as backup energy for the nanogrid system in the laboratory. This study uses a Python machine learning module which enables the implementation of SVR with various types of kernels including linear kernels, polynomial kernel, and RBF kernel. The tests carried out in this research using the grid search module show that the best performance is obtained when using the RBF kernel.]]>
Copyrights © 2021
