Battery state of charge estimation based on ensemble machine learning techniques (Gradient boosting)
Version 1.0.0
Description: The present repository shares the the Python script to generate a gradient boosting machine learning model to estimate the state of charge of a Li-ion battery. No data has been shared due to confidentiality agreement. However, the structure of the code implemented has been provided.
Motivation: The available capacity of a battery, called the state of charge, is a fundamental characteristic for energy storage applications or electric vehicles. In order to model the state of charge of a lithium-ion battery using data-driven techniques, complex algorithms should be used so the dynamic behaviours of the battery are captured. An ensemble decision tree called gradient boosting tree was fitted with the information extracted from different experiments based on dynamic and constant discharge profiles at different temperatures and implemented in a laboratory. Despite the capacity and and the battery’s state of health being below its theoretical life expectancy, results of the model showed an adequate behaviour. The model’s performance was validated using previously unused data.