Application of data-driven models in the analysis of marine power systems

Abstract

Utilization of measurements from on-board monitoring systems of marine vessels is a part of shipbuilding industry’s digitalization phase. The data collected can be used to verify and improve vessel’s power system design. Deployment of data-driven statistical models can enhance the knowledge about the power requirements. In this study, we describe a data-driven statistical model that can be used to study and analyze the power requirement of a vessel, which might help to understand the key factors that influence the power and to quantify their contribution. We propose a powerful tool namely, generalized additive model (GAM), which allows us to model nonlinearities. We build the GAM to see the relationship between power consumed and the key influential factors for a power system based on real data from a platform supply vessel (PSV) in a dynamic positioning (DP) mode with diesel-electric configuration. We also describe the importance of feature extraction based on Hilbert Transform to improve the model. In addition, we fit the linear regression (LR) model as a reference model. In the last phase we verify the results of GAM, LR with simulation model from ShipX to show that the data-driven model is within the boundaries of power requirement from simulations.