Session: 08-01-01 Waves and Loads

Paper Number: 79208

79208 - Full Scale CFD Validation Using Ship Performance and Wave Pattern Measurements of a Mega Cruise Ship 

The International Maritime Organization (IMO) fixed the objective of a minimum 50% reduction in total annual greenhouse gas (GHS) emissions by 2050 compared to 2008. The shipping industry is directly impacted by this challenge and must improve ships fuel consumption and reliability. Hull form optimization is therefore essential when designing and improving new ships. Computational Fluid Dynamics (CFD) is nowadays a common practice in the ship design procedure, from basic resistance simulations to state of the art simulations such as self-propulsion with rotating propeller or manoeuvring in waves. Historically, CFD solvers were validated extensively against towing tanks experiments carried out at model scale. Differences in flow characteristics, mainly focussing  on the wave pattern and boundary layer thickness at the ship’s stern, occur due to different Reynolds numbers between model and full scale. These nonlinear phenomena, called scale effects, are difficult to include in predicting full scale performance from model scale values. Therefore, it is increasingly common to perform full scale CFD calculations when designing new ships. However, validation of full scale CFD is essential for the use and interpretation of CFD results. Unfortunately, the available full scale data are very limited and generally only include time series of speed, propeller rate, torque and thrust. Flow velocity and radiated wave pattern data are very rare at full scale because of the complexity of such on board measurements. Nevertheless, in the current work, full scale wave pattern measurement were performed on a 330 meter long cruise ship with a width of 43 meters and a height of 65 meters, sailing at 20 knots. During the sea trials, the waves were measured using the Digital Image Correlation technique. This is an image analysis method capable of measuring displacements and deformation of a surface in space. It required placing two computer vision cameras on the top deck, 40 meters above the waterline, overlooking the stern waves of the ship. The approximate size of the field of view was 100 by 30 meters. This allowed to measure three-dimensional free surface elevations of two entire wave lengths behind the ship. The ship’s speed, shaft power, propeller rate, motions and environmental waves were measured as well. After the trials, Reynolds Averaged Navier-Stokes (RANS) simulations were performed using STARCCM+. The propeller was modelled both as an actuator disk as well as a rotating propeller with a sliding interface. The turbulence closure model was k-ω SST and the free surface was modelled using the Volume of Fluid method. The balance between hull resistance and propeller thrust was verified as a first step, showing less than 2.5% difference. The power determined by CFD was validated against the one measured during the sea trials with less than 2.3% difference. Finally, the comparison of the stern wave pattern resulting from full scale CFD simulations with the pattern measured using DIC showed excellent agreement with good accuracy.

 

Keyword :

Sea trials validation, stern wave measurements, full scale CFD, STARCCM+, Digital Image Correlation, Vic3D

Presenting Author: D.R. Schouten MARIN

Authors:

D.R. Schouten MARIN
Aurelien Drouet Chantiers de l'Atlantique
Miloš Birvalski MARIN
Loic Morand Chantiers de l'Atlantique