Abstract
Abstract
Highlights
At sufficiently high Reynolds numbers all surfaces are hydrodynamically rough, as is almost always the case for flows past the surfaces of naval vessels
The construction of a predictive model from a large ensemble of datasets for the equivalent sand-grain height ks of a surface of arbitrary roughness, as a function of many different measures of surface topography, is a labelled regression problem that is well-suited to machine learning (ML) techniques
Machine learning techniques are well suited to this modelling problem because: (i) it is complex in so far as different kinds of surface roughness yield different flow phenomena which are modelled most accurately in different ways, making the prospect of a general physical model very remote; and (ii) the dependent surface-roughness variables upon which ks is modelled are a large non-orthogonal set for which robust multivariable regression techniques are required
Summary
At sufficiently high Reynolds numbers all surfaces are hydrodynamically rough, as is almost always the case for flows past the surfaces of naval vessels. Moody 1944) that can predict accurately the surface drag coefficient is not known a priori and does not appear to be equivalent to any single geometrical length scale, such as an average or a root mean square (r.m.s.) of roughness height (Flack 2018) It is well-established that ks can depend on many geometrical parameters such as the effective slope (Napoli, Armenio & De Marchis 2008; Yuan & Piomelli 2014a) and the skewness of the roughness height distribution (Flack & Schultz 2010). The small number of roughness parameters used to devise ks correlations tended to limit their application to a narrow range of surface roughness Since it appears that many geometrical parameters, such as porosity, moments of roughness height (e.g. r.m.s., skewness and kurtusis), effective slope and surface inclination angle might affect ks, it is useful to employ a data science approach suited to modelling large multivariate/multioutput systems. We describe the ML models, their predictions of ks and their uncertainty
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have