A review on flow interactions analysis using machine learning-driven CFD modelling for offshore structures and marine applications

The integration of Machine Learning (ML) and Computational Fluid Dynamics (CFD) could become a paradigm shift in marine engineering applications. This review deals with the current state-of-the-art on the recent progress in the implementation of ML methods on marine-related CFD problems. This paper presents the use of CFD in the marine application world, such as the flow analysis over rectangular cylinders, wind turbine foundations and submerged structures under wave action. However, the supervised, unsupervised and semi-supervised methods are also discussed critically. The paper also discusses the use of ML in the prediction of vessel shaft power, fuel consumption, propeller analysis, structural damage inspection and possibly autonomous hull maintenance. The present overview is also focused on the CFD-based ML models to simulate the turbulent flows, predict fluid field and pressure drops, analyse turbine wake and blade loads, generate pump performance maps, and assess offshore structure response, namely vortex-induced vibrations.