Data-Driven decision support tool for power quality measures in marine vessel power system

Efficiency and reliability are important properties for a marine vessel power system, both from an economical point of view and for safety reasons. Efficiency and reliability correlate with power quality, as a poor power quality may endanger devices in the system as well as reducing the power factor(s). In this initial work bi-linear methods such as Principal Component Analysis (PCA) and Partial Least Squares Regression (PLSR) are explored as tools for analysing and predicting harmonic pollution depending on the load demands in the system. The analysis and prediction are data-driven, giving the data a chance to speak for itself.