Towards a standardized statistical correction between RANSe-based full-scale CFD and sea trial measurement

Master thesis (2021)

Authors

B.J.H. Vendeloo Mechanical Engineering

Contributors

T.J.C. van Terwisga Ship Hydromechanics and Structures - Mechanical, Maritime and Materials Engineering (supervisor 1)
C.H. Thill Ship Hydromechanics and Structures - Mechanical, Maritime and Materials Engineering (supervisor 2)
A. Coraddu Ship Design, Production and Operations - Mechanical, Maritime and Materials Engineering (supervisor 2)
Faculty
Mechanical Engineering
Copyright: © 2021 Bauke Vendeloo
Uncertainty Full-scale CFD Sea trial measurements Statistical correction
More Info
expand_more

Published Date

10-12-2021

Language

English

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Faculty

Mechanical Engineering

Abstract

This research considers Amels LE and SX super yachts. Bare hull, appended hull, and actuator disk simulations were performed using CFD, and compared to sea trial results. The results of the CFD are unique because they were generated after the sea trials were conducted, resulting in the ship speed in CFD and sea trial being similar. The results were compared using CF, CFD the relative error between measured sea trial and CFD computed resistance. The uncertainty was defined as a function of accuracy of precision, represented by the mean μ and standard deviation σ of CF, CFD. First, it was investigated if the uncertainty of CF, CFD could be reduced by including more physics in the simulations and what the impact of more variety in ship dimensions was. Subsequently, an investigation to the driving factors behind the uncertainty was conducted by performing a linear regression. After which it was investigated if the total uncertainty can be reduced by replacing experimental uncertainty by numerical uncertainty. Finally, a statistical correction to adjust the resistance was used to predict the power-speed relationship. Two correction factors were investigated. The first is based on the sample mean X of CF, CFD, and the second is CF, CFD derived from a CDF.

It was seen that including more physics reduced the uncertainty of CF, CFD. The uncertainty decreased whilst comparing bare hull to appended hull simulations, and it increased whilst comparing appended hull to actuator disk simulations. Thus, it was concluded that the actuator disk does not lead to a more realistic simulation of the flow. More variety in the ship dimensions lead to a larger uncertainty of CF, CFD, which was expected. The variance was found to be caused by the ship speed, the mean wave height and mean wind speed, and also by the block coefficient if different ship types are considered. By computing the hull efficiency ηH numerically instead of experimentally, the total uncertainty of CF, CFD was reduced. The power-speed relationship can be predicted accurately, but not precise with CF, CFD based on X. While the prediction with a CF, CFD derived from a CDF is not accurate and not precise.