Clustering and Outlier Analysis for Key Performance Indicators in Battery Energy Storage Systems applications

This research work focuses on implementing outlier analysis and clustering to provide an assessment of the charging and discharging processes of Battery Energy Storage Systems (BESSs). K-Means, Density-based spatial clustering of applications with noise (DBSCAN), and Local Outlier Factor (LOF) are the main algorithms executed to illustrate Key Performance Indicators (KPIs) and their corresponding critical values during battery operation. Additional Data Mining methods are implemented to provide feature selection, correlation analysis, parameter estimation, and validation. Implemented algorithms show that there is a strong correlation between specific variables at certain operation stages, which is complemented by the lifetime period of BESSs.