Upload
others
View
4
Download
0
Embed Size (px)
Citation preview
Sustainable mobility | New energies | Responsible oil and gas
© 2
016
- IFP
Ene
rgie
s no
uvel
les
Decay test transient simulation setup o Mooring included as 3 linear “springs” o Restrained motion due to the tension legs:
• surge, heave and pitch
Meshing and boundary conditions o 7M cells mesh, refined around free surface and float (layers) o 4 x floater diameter domain, 100 m water depth, o Wall functions for turbulence modeling, slip conditions for bottom and
side walls, atmospheric pressure condition at the top
State of the art for Oil & Gas o Large body: Potential flow theory o Small body: Morison empirical formula
Morison coefficient calibration o Impact on fatigue and extreme loads impact on FOWT design o Existing coefficient tables for O&G New databases for FOWTs ?
o Radiated waves less dissipated with stabRAS models than with conventional k-Ω SST
Turbulence modeling analysis o Conventional k-Ω SST too diffusive at free surface o k-Ω SST stabRAS [2] could be more adapted (compatibility with
morphing mesh ?)
Surge Decay CFD Simulations of a Tension Leg Platform Floating Wind Turbine
A. Borràs Nadal1, P. Bozonnet2 1IFP Energies nouvelles, 1 et 4 avenue de Bois-Préau, 92852 Rueil-Malmaison, France
2IFP Energies nouvelles, Rond-point de l'échangeur de Solaize, BP 3, 69360 Solaize, France
Floater development with : Tension Leg Platform (TLP) Design checked with DeepLinesWind™ software o Aero-servo-hydro-elastic engineering tool
• Blade-Element-Momentum theory • Morison theory / potential flow theory • Finite elements
CFD benefits in the design process
Floater development with : Tension Leg Platform (TLP)
Design checked with DeepLinesWind™ softwareo Aero-servo-hydro-elastic engineering tool
• Blade-Element-Momentum theory• Morison theory / potential flow theory• Finite elements
CFD benefits in the design process
State of the art for Oil & Gaso Large body: Potential flow theoryo Small body: Morison empirical formula
Morison coefficient calibrationo Impact on fatigue and extreme loads impact on FOWT designo Existing coefficient tables for O&G New databases for FOWTs ?
OpenFOAM features and libraries in constant evolution o Meshing with integrated OpenFOAM tool snappyHexMesh o Diphasic Flow: Volume of Fluid (VOF) MULES & isoAdvector [1] o Moving Float:
• CFD coupled with motion equation: sixDoFRigidBody library • Mesh morphing: interDyMFoam solver • Overset mesh: OverInterDyMFoam solver
o RANS turbulence modeling using k-Ω SST / stabRAS [2]
1. Context 2. Hydrodynamics Background
Decay test transient simulation setup o Mooring included as 3 linear “springs”o Restrained motion due to the tension legs:
• surge, heave and pitch
Meshing and boundary conditionso 7M cells mesh, refined around free surface ano 4 x floater diameter domain, 100 m water depo Wall functions for turbulence modeling, slip conditions for bottom and
side walls, atmospheric pressure condition at
nddddddnd float ((layey rs))th,
onditions for bottom athe top
OpenFOAM features and libraries in constant evolutiono Meshing with integrated OpenFOAM tool snappyHexMesho Diphasic Flow: Volume of Fluid (VOF) MULES & isoAdvector [1]o Moving Float:
• CFD coupled with motion equation: sixDoFRigidBody library• Mesh morphing: interDyMFoam solver• Overset mesh: OverInterDyMFoam solver
o RANS turbulence modeling using k-Ω SST / stabRAS [2]
Basin 3.24 m surge test comparison
4. Numerical Wave Tank Setup
6. Conclusions & On-going Work
o Radiated waves less dissipated with stabRAS models than with conventional k-Ω SST
Turbulence modeling analysiso Conventional k-Ω SST too diffusive at free surfaceo k-Ω SST stabRAS [2] could be more adapted (compatibility with
morphing mesh ?)
Basin 3.24 m surge test comparison
5. Results on Surge Decay & Numerical Parameters
[1] Roenby J., Bredmose H., Jasak H. IsoAdvector: Geometric VOF on general meshes, OpenFOAM - Selected papers of the 11th Workshop [2] Larsen, B. E. and Fuhrman, D. R., On the over-production of turbulence beneath surface waves in Reynolds-averaged Navier–Stokes models , Journal of Fluid Mechanics, October 2018 [3] Burmester S., Vaz G., “Investigation of a semi-submersible floating wind turbine in surge decay using CFD”, Ship technology research. November 2018 [4] I. Rivera-Arreba I. “Computation of Nonlinear Wave Loads on Floating Structures, Master Thesis, August 2017
NWT is implemented in OpenFOAM to study Hydrodynamics for FOWT design Decay tests are simulated but still under investigation CFD post-processing of brace and buoy loads to compare with DeepLinesWind™ Next step: combining wave generation and float motion
k-Ω SST k-Ω SST stabRAS
k-Ω SST stabRAS k-Ω SST
turbulent viscosity at free surface
Dynamic pressure at free surface
Step-by-step methodology 1. Moving body / no wave 2. Fixed body / wave 3. Moving body with wave Experimental validation case: TLP decay test in wave tank o Free damped oscillations from initial offset, back to equilibrium position o To identify rigid body modes (frequency & damping)
o Damping fairly well predicted, strongly dependent on numerical parameters.o Natural period difference of 8% between experiment and CFD. No influence of numerical parameters observed. o Some period differences have also been noticed by Burmester [3] and Rivera [4].
They are attributed to experimental uncertainties (inertia, mooring system characterisation, COG location).
Step-by-step methodology1. Moving body / no wave2. Fixed body / wave3. Moving body with wave
Experimental validation case: TLP decay test in wave tank o Free damped oscillations from initial offset, back to equilibrium positiono To identify rigid body modes (frequency & damping)
St b t th d l3. Numerical Wave Tank Implementation
FOWT Design
Experimental data
DeepLinesWind
CFD
In-situ data
Costly, limited Extrapolation to full scale (Froude similitude / drag effects ?)
Difficult to access and to process Design tool to
be calibrated for FOWT
An alternative: Numerical Wave Tank