ASME Conference Presenter Attendance Policy and Archival Proceedings

2018;():V002T00A001. doi:10.1115/OMAE2018-NS2.

This online compilation of papers from the ASME 2018 37th International Conference on Ocean, Offshore and Arctic Engineering (OMAE2018) represents the archival version of the Conference Proceedings. According to ASME’s conference presenter attendance policy, if a paper is not presented at the Conference by an author of the paper, the paper will not be published in the official archival Proceedings, which are registered with the Library of Congress and are submitted for abstracting and indexing. The paper also will not be published in The ASME Digital Collection and may not be cited as a published paper.

Commentary by Dr. Valentin Fuster

CFD and FSI: Advanced Computation: Optimization, Big Data, Machine Learning

2018;():V002T08A001. doi:10.1115/OMAE2018-77649.

Computational Fluid Dynamics (CFD) is widely used in industry and academic research to investigate complex fluid flow. The bottleneck of a realistic CFD simulation is its long simulation time. The simulation time is generally reduced by massively parallel Central Processing Unit (CPU) clusters, which are very expensive. In this paper, it is shown that the CFD simulation can be accelerated significantly by a novel hardware called General Purpose Computing on Graphical Processing Units (GPGPU). GPGPU is a cost-effective computing cluster, which uses the Compute Unified Device Architecture (CUDA) of NVIDIA devices to transform the GPU into a massively parallel processor.

The paper demonstrates the faster computing ability of GPU compared to a traditional multi-core CPU. Two scenarios are simulated; one is a 2-dimensional simulation of regular wave and another one is a 3-dimensional motion of a floating ship on a regular wave. A smoothed particle hydrodynamics (SPH) based CFD solver is used for simulating the complex free-surface flow. The performance of a single GPU is compared against a commonly used 16 core CPU. For a large simulation of 6 degrees of freedom (DOF) ship motion simulation, the comparative study exhibits a speedup of more than an order of magnitude, reducing simulation time from 30 hours to about 2 hours. This indicates a CUDA enabled GPU card can be used as a cost-effective computing tool for a reliable and accurate SPH-based CFD simulation. The cost-benefit analysis of GPU over a CPU cluster is also discussed.

Commentary by Dr. Valentin Fuster
2018;():V002T08A002. doi:10.1115/OMAE2018-78179.

Design of offshore oil platforms requires accurate prediction of the maximum wave loads due to slamming on horizontal decks. The physical processes that influence the load are the propagation of irregular short-crested wind-driven storm seas, wave breaking, and wave-structure interaction. Furthermore, the ocean is a stochastic environment, so the load and its maximum can be considered as random variables. Ideally, the designer would like to know not only the most probable extreme load, but also the extreme load distribution.

In this paper we will use a novel technique to prescribe wave environments that lead to extreme responses so that high-fidelity simulations of the highly-nonlinear process can be investigated in detail. Specifically, the dynamics of the relative motion of the sea surface and the platform will be assumed via the selection of a sea spectrum, and the extreme-value probability distribution function (PDF) will be calculated for a given exposure window. The novel aspect of the work is in the way that a set of deterministic sea environments will be generated that are amenable for simulation with a state-of-the-art computational-fluid dynamics (CFD) software. The resulting wave environments will be simulated to estimate the extreme-value PDF.

Topics: Stress , Waves
Commentary by Dr. Valentin Fuster
2018;():V002T08A003. doi:10.1115/OMAE2018-78338.

Unsteady separated flow behind a bluff body causes fluctuating drag and transverse forces on the body, which is of great significance in many offshore and marine engineering applications. While physical experimental and computational techniques provide valuable physics insight, they are generally time-consuming and expensive for design space exploration and flow control of such practical scenarios. We present an efficient Convolutional Neural Network (CNN) based deep-learning technique to predict the unsteady fluid forces for different bluff body shapes. The discrete convolution process with a non-linear rectification is employed to approximate the mapping between the bluff-body shape and the fluid forces. The deep neural network is fed by the Euclidean distance function as the input and the target data generated by the full-order Navier-Stokes computations for primitive bluff body shapes. The convolutional networks are iteratively trained using a stochastic gradient descent method to predict the fluid force coefficients of different geometries and the results are compared with the full-order computations. We have extended this CNN-based technique to predict the variation of force coefficients with the Reynolds number as well. Within the error threshold, the predictions based on our deep convolutional network have a speed-up nearly three orders of magnitude compared to the full-order results and consumes an insignificant fraction of computational resources. The deep CNN-based model can predict the hydrodynamic coefficients required for the well-known Lighthill’s force decomposition in almost real time which is extremely advantageous for offshore applications. Overall, the proposed CNN-based approximation procedure has a profound impact on the parametric design of bluff bodies and the feedback control of separated flows.

Commentary by Dr. Valentin Fuster
2018;():V002T08A004. doi:10.1115/OMAE2018-78415.

In this paper, a general data-driven approach to construct a reduced-order model (ROM) for the coupled fluid-structure interaction (FSI) problem of a transversely vibrating bluff body in an incompressible flow is presented. The proposed data-driven approach relies on the Eigensystem Realization Algorithm (ERA) to design ROM models in a state-space format. The stability boundaries of the coupled FSI system are obtained by examining the eigenvalue trajectories of the ERA-based ROM. These stability boundaries provide us valuable quantitative insights into the lock-in phenomenon of the bluff-body vibration. We demonstrate the present ERA-based ROM technique for various configurations of bluff bodies such as an isolated single cylinder, the side-by-side and the tandem cylinder arrangements. A comparative study on the effect of different appendages to suppress the VIV of a cylinder is also presented using the ERA-based stability analysis. The validity of the proposed method for the FSI stability analysis on such variety of configurations has not been presented before and is the novel contribution of this paper. Overall, the proposed data-driven framework is found to be much more effective in terms of computational costs and the predicted lock-in regions are comparable to high-fidelity full-order simulations. This work has a potential for a profound impact on the design optimization and control of bluff body structures used in offshore industry.

Commentary by Dr. Valentin Fuster
2018;():V002T08A005. doi:10.1115/OMAE2018-78425.

Fluctuating wave force on a bluff body is of great significance in many offshore and marine engineering applications. We present a Convolutional Neural Network (CNN) based data-driven computing to predict the unsteady wave forces on bluff bodies due to the free-surface wave motion. For the full-order modeling and high-fidelity data generation, the air-water interface for such wave-body problems must be captured accurately for a broad range of physical and geometric parameters. Originated from the thermodynamically consistent theories, the physically motivated Allen-Cahn phase-field method has many advantages over other interface capturing techniques such as level-set and volume-of-fluid methods. The Allen-Cahn equation is solved in the mass-conservative form by imposing a Lagrange multiplier technique. While a tremendous amount of wave-body interaction data is generated in offshore engineering via both CFD simulations and experiments, the results are generally underutilized. Design space exploration and flow control of such practical scenarios are still time-consuming and expensive. An alternative to semi-analytical modeling, CNN is a class of deep neural network for solving inverse problems which is efficient in parametric data-driven computation and can use the domain knowledge. It establishes a model with arbitrarily generated model parameters, makes predictions using the model and existing input parametric settings, and adjusts the model parameters according to the error between the predictions and existing results. The computational cost of this prediction process, compared with high-fidelity CFD simulation, is significantly reduced, which makes CNN an accessible tool in design and optimization problems. In this study, CNN-based data-driven computing is utilized to predict the wave forces on bluff bodies with different geometries and distances to the free surface. The discrete convolution process with a non-linear rectification is employed to approximate the mapping between the bluff-body shape, the distance to the free-surface and the fluid forces. The wave-induced fluid forces on bluff bodies of different shapes and submergences are predicted by the trained CNN. Finally, a convergence study is performed to identify the effective hyper-parameters of the CNN such as the convolution kernel size, the number of kernels and the learning rate. Overall, the proposed CNN-based approximation procedure has a profound impact on the parametric design of bluff bodies experiencing wave loads.

Commentary by Dr. Valentin Fuster

CFD and FSI: Advanced Computation: Software

2018;():V002T08A006. doi:10.1115/OMAE2018-77108.

The marine propeller is regarded as critical component with regard to the performance of the ships and torpedoes. Traditionally marine propellers are made of manganese-nickel-aluminum-bronze (MAB) or nickel-aluminum-bronze (NAB) for superior corrosion resistance, high-yield strength, reliability, and affordability. Since the composite materials can offer the potential benefits of reduced corrosion and cavitation damage, improved fatigue performance, lower noise, improved material damping properties, and reduced lifetime maintenance cost, Many researches on the application of the composite materials for marine propeller had been conducted. In this work, the INSEAN 1619 large screw 7 bladed propeller is analyzed, to explore the hydrodynamic and structural performance of composite materials effect on propeller’s performances, The commercial software ANSYS Workbench was used in this research. The coupled FSI method was used to analysis the dynamic performance of INSEAN 1619 large screw 7 bladed propeller made of different materials. The simulation results show that the effect of fluid–structure interaction in the analysis of flexible composite propellers should be considered.

Commentary by Dr. Valentin Fuster
2018;():V002T08A007. doi:10.1115/OMAE2018-77293.

Bubble breakup and coalescence is a phenomenon which occurs within a developing subsea gas plume. A Computational Fluid Dynamics (CFD) model incorporating bubble breakup and coalescence was developed to describe the behaviour of a subsea gas release and the subsequent rising gas plume. The model was assessed for its suitability in capturing the characteristic behaviour of a rising gas plume by comparing the CFD results with experimental data obtained from underwater gas release experiments.

The study shows bubble breakup and coalescence plays a key role in determining the shape and the behaviour of a subsea gas release. Without the bubble breakup and coalescence included in the CFD model a narrower plume width and higher rising velocity is observed when compared to the experimental data. With bubble breakup and coalescence included the results obtained from the CFD model more accurately match the experimental data. Breakup and coalescence is a mechanism which redistributes the energy within the core of the gas plume towards the edge of the plume. This has a significant impact on the plume characteristics and is vital to be included in the CFD model to describe the behaviour of the released gas.

The study was carried out using air as the released gas. This was done to compare with the available experimental data where air was used as the source. However the CFD model developed is applicable for hydrocarbon subsea gas releases.

Commentary by Dr. Valentin Fuster
2018;():V002T08A008. doi:10.1115/OMAE2018-77995.

A novel formulation for marine propellers based on adaptations from wing lifting-line theory is presented; the method is capable of simulating propellers with skewed and raked blades. It also incorporates the influence of viscosity on thrust and torque from hydrofoil data through a nonlinear scheme that changes the location of the control points iteratively. Several convergence studies are conducted to verify the different aspects of the numerical implementation and the results indicate satisfactory convergence rates for Kaplan, KCA, and B-Troost propellers. It is expected that the method accurately describes thrust, torque, and efficiency under the moderately loaded propeller assumption.

Topics: Design , Propellers
Commentary by Dr. Valentin Fuster
2018;():V002T08A009. doi:10.1115/OMAE2018-78281.

Precise position and motion control of offshore vessels is often challenging, especially in harsh environment due to highly nonlinear dynamic loads from free-surface ocean waves and currents. In addition, coupled nonlinear effects of risers and mooring cables connected to the vessel can lead to unexpected responses, thus justifying the significance of modeling these nonlinear coupled effects for safer and cost-effective design and operation of offshore structures. In this study, a fully coupled multi-field fluid-structure-interaction (FSI) solver is developed to simulate the wave- and flow-induced vibration of the flexible multibody system with constraints (viz., vessel-riser system) in a turbulent flow. The structural domain with multibody systems is solved using nonlinear co-rotational finite element method, whereas the fluid domain is solved using Petrov-Galerkin finite element method for moving boundary Navier-Stokes solutions. A partitioned iterative scheme based on non-linear interface force corrections is employed for coupling of the turbulent fluid-flexible multibody system with nonmatching interface meshes. Delayed Detached Eddy Simulation (DDES) via the Positivity Preserving Variational (PPV) method is employed for modeling turbulence effects at high Reynolds number. The free-surface ocean waves are modeled by the Allen-Cahn based phase-field method. We address two key challenges in the present variational coupled formulation. Firstly, the coupling of the incompressible turbulent flow with a system of nonlinear elastic bodies described in a co-rotated frame. Secondly, the two-phase coupling based on the phase-field approach to model the air-water interface. We then present the dynamics of coupled vessel-riser system studied in harsh environmental conditions with a view of developing a robust station keeping system. The proposed fully-integrated methodology based on the first principles of variational continuum mechanics removes many assumptions and empirically assigned parameters (e.g. drag and inertia coefficients) for modeling the surrounding fluid flow at high Reynolds number.

Commentary by Dr. Valentin Fuster
2018;():V002T08A010. doi:10.1115/OMAE2018-78515.

This paper presents the construction and solution algorithm for a projection-based reduced model emphasising on integration with sensing data from a physical modelling for a coupled physical-numerical simulation. The reduced model is constructed based on the well-known Galerkin projection of the equations governing the physical processes on reduced subspaces obtained by the Proper Orthogonal Decomposition method. The projection is done using numerical discretisation in the existing OpenFOAM® platform. A simple yet efficient way to integrate physical sensing data into the reduced model to enhance the model accuracy is presented. The reduced model is demonstrated for classical incompressible flows at low and intermediate Reynolds numbers driven by various boundary conditions. Model results show excellent agreement with the full CFD simulations while only a fraction of computational resource is required.

Topics: Simulation
Commentary by Dr. Valentin Fuster

CFD and FSI: Advanced Computation: Verification, Validation, and Best Practices

2018;():V002T08A011. doi:10.1115/OMAE2018-78522.

To research the flexible hydrofoils’ hydroelastic response, the fluid-structure interaction (FSI) characteristic investigation is conducted on the basis of the analysis of a rigid hydrofoil’s hydrodynamic performance. For a rigid cantilevered rectangular hydrofoil, the pitching hydrodynamic performance is calculated using boundary motion with remeshing strategy. The Laminar Separation Bubble (LSB) and turbulent transition are captured. Numerical flow analysis revealed that the LSB occurs at 0.8c when pitching at initial angle of attack. As the angle increases to 5.1°, the laminar to turbulent transition occurs and the lift presents an inflection. For a geometric equivalent flexible hydrofoil, the static FSI characteristic is researched using oneway and two-way FSI method. The lift decreases and the drag increases using two-way compared to one-way FSI. The center of pressure and the maximum deformation move from trailing edge to leading edge as the angle of attack increases, showing the necessary of two-way FSI calculation. The transient FSI characteristic of the flexible hydrofoil is then studied using LES model. The lift fluctuation at 8° in frequency domain is calculated . The dry mode and wet mode natural frequency of the flexible hydrofoil are calculated to simulate the vibration performance, which meet the experiment data quite well, laying foundation for further research on the hydroelastic vibration response.

Commentary by Dr. Valentin Fuster
2018;():V002T08A012. doi:10.1115/OMAE2018-78531.

A common structural element encountered in semisubmersible designs is a rectangular vertical column with rounded corners. The time-averaged drag and oscillating lift and drag forces on such columns are strongly influenced by the location of the lines of flow separation on the column and hence the angle of attack of the incoming flow and the corner radius. In this paper we examine published wind tunnel data to illustrate these effects which include angle of attack and Reynolds number effects. This examination suggests that care must be exercised modeling flows around these elements. Also, the data suggest that Reynolds number effects and surface roughness effects may distort the results of scaled experiments. We use CFD simulations first to model the existing data and then to explore the possible changes in hydrodynamic properties due to Reynolds number and boundary layer effects. Recommendations are made regarding the physical and CFD modeling of the flow over these structures.

Commentary by Dr. Valentin Fuster
2018;():V002T08A013. doi:10.1115/OMAE2018-78598.

This paper presents the assessment of the modelling error (Validation) of a Navier-Stokes solver using Volume of Fluid (VOF) and moving grid techniques in the simulation of a free falling wedge into calm water. This problem has been studied experimentally to determine the time histories of six pressure probes located on the wedge surface and the acceleration of the wedge. The simulation is restricted to the first 100ms after the impact of the wedge on the water (t = 0 at the impact) and the mathematical model uses the following assumptions: incompressible fluid; two-dimensional, laminar flow, negligible shear-stress at the surface of the wedge and deep water. The selected quantities of interest are the peak pressures at the six sensors, time intervals between peak pressures at the sensors, sensors pressures and acceleration of the wedge at six different time instants and integrated pressure signals for 80ms after the pressure peak at the first sensor.

The application of the ASME V&V 20 standard to local quantities is presented, including the estimation of experimental and numerical uncertainties. Furthermore, a multivariate metric is used to evaluate quantitatively the overall performance of the mathematical model. The results show significant comparison errors (mismatches between simulations and measurements) for the accelerations, which may be a consequence of the assumptions of a deep water boundary condition at the bottom. However, such conclusion is hampered by some doubts about the accuracy of the experimental data. On the other hand, modeling errors are significantly smaller for the pressure measurements at the six sensors for which the main challenge is to reduce the validation uncertainty Uval. In many of the selected flow quantities, Uval is dominated by the experimental uncertainty.

Topics: Water , Wedges
Commentary by Dr. Valentin Fuster

CFD and FSI: Free Surface Flows

2018;():V002T08A014. doi:10.1115/OMAE2018-77207.

The aim of this paper is to evaluate how much affects the presence of gravity and free-surface to a flexible structure in a classical fluid structure interaction (FSI) problem typically found in off-shore problems and other oceanic applications. The base problem selected is the Turek benchmark case where a deformable plate is attached to the wake of a circular cylinder. To focus on the differences of considering free surface, a simple geometry has been selected and two different situations have been studied: the first one is the classical Turek benchmark, the second is a similar geometry but adding gravity and free surface. The free surface problem was studied placing the structure at different depths and monitoring the deformation and forces on the structure.

Commentary by Dr. Valentin Fuster
2018;():V002T08A015. doi:10.1115/OMAE2018-77309.

In the past, the CFD simulation method ComFLOW has been successfully applied in a wide range of offshore applications, involving wave simulations and impact calculations. In many of these calculations the area of interest comprises a small part of the domain and remains fixed in time, which allows for efficient grid refinement by means of grid stretching or static local refinement. However, when trying to accurately resolve the surface dynamics and kinematics of irregular and breaking waves, the resolution requirements are strongly time-dependent and difficult to predict in advance. Efficient grids can only be obtained by means of time-adaptive refinement. A Cartesian block-based refinement approach is followed which allows for efficient grid adaptation, with moderate overhead. An array-based data structure is employed which exploits the semi-structured nature of the Cartesian block grid. Currently we are testing the method with the simulation of lifeboat drops in regular and irregular wave conditions. This poses several challenges such as accurately imposing the incoming waves and modifying the absorbing boundary conditions to support two-phase flow. To reduce the wall-clock time, the simulation method has been parallelized.

Commentary by Dr. Valentin Fuster
2018;():V002T08A016. doi:10.1115/OMAE2018-77511.

Ocean waves are random by nature and can be regarded as a superposition of a finite number of regular waves travelling in different directions with different frequencies and phases. Cylinder-shaped objects are commonly present in most coastal structures. An irregular bottom topography has a significant influence on the wave behaviours and therefore the wave forces on the coastal structures. A numerical approach that is able to calculate the wave forces on a cylinder in a multi-directional irregular wave field over an irregular bottom is desired. As Computational Fluid Dynamics (CFD) is able to represent most of the wave behaviour with few assumptions, it is considered to be an attractive option to address these issues. The open-source CFD wave model REEF3D has shown good performances in simulating wave propagation over irregular bottoms and a good prediction of wave forces on a cylinder in a uni-directional wave field, yet the ability to calculate the wave force in a multi-directional irregular sea needs to be validated. Therefore, this paper attempts to simulate the multi-directional random sea interaction with a large cylinder using REEF3D and validate the results. A novel approach of multi-directional irregular wave generation method in a CFD-based numerical wave tank is introduced. Only even-bottom tanks are considered in this study, leaving the irregular bottom simulation for future studies. Furthermore, among many factors that influence the wave forces, this paper focuses particularly on the effect of the wave steepness. The effects of wave steepness in regular waves, uni-directional irregular waves and multi-directional irregular waves are investigated. Goda’s JONSWAP frequency spectrum and the frequency-independent Mitsuyasu directional spreading function are used to generate the multi-directional irregular waves. The wave forces due to the multi-directional irregular waves in the numerical tank are compared with experimental data. The performance of the CFD simulation is analysed and discussed.

Commentary by Dr. Valentin Fuster
2018;():V002T08A017. doi:10.1115/OMAE2018-77646.

A series of experiments were carried out with a flat plate towed normal to the flow in quiescent fluid. The focus was given to the analysis of the drag force seen by the plate as a function of its aspect ratio and hydraulic diameter. The effect of towing the plate near the water free surface was also investigated thoroughly. Plates of aspect ratio ranging from 0.25 to 4 were towed in a still water tank at different Reynolds numbers in the range from 15000 to 60000. Submergence depth was measured from the upper edge to the free surface and varied from zero to the centre of the tank. Forces on the plates were measured using a submersible bending beam load cell and the carriage motion was monitored by a rotary potentiometer. It was found that the drag increases abruptly prior subsiding with increasing submergence depth, with this effect being more dominant in lower aspect ratio plates. The abrupt rise in the drag is due to the interaction of the upper edge of the plate with the free surface resulting in a large shrinkage of the recirculation zone. The non-unit low aspect ratio plates also showed another drag peak around 50% depth, especially at lower speeds. Overall, the trends were Reynolds number independent, except when the aspect ratios was in the range from 0.75 to 1.33 and the plate was near the free surface.

Commentary by Dr. Valentin Fuster
2018;():V002T08A018. doi:10.1115/OMAE2018-77752.

The design of new offshore structures requires the calculation of the wave-induced loads. In this regard, the Computational Fluid Dynamics (CFD) methodology has shown to be a reliable tool, in the case of breaking waves especially. In this paper, two CFD models are tested in the reproduction of an experimental spilling wave impacting a circular cylinder. The numerical results from the models are shown together with the experimental measurements.

Commentary by Dr. Valentin Fuster
2018;():V002T08A019. doi:10.1115/OMAE2018-77848.

Plastic pollution in the marine environment is an increasing problem with severe impacts on ecosystems and economies around the globe. The Ocean Cleanup (TOC) Foundation develops a floating barrier able to intercept, concentrate and extract plastic from the marine environment.

TOC has conducted several experiments and numerical studies to determine the capture efficiency of its system. One of the phenomena leading to its decrease is wave overtopping or under-flowing when the system cannot properly follow the waves, this issue is amplified by the use of a stiffer barrier than the original deep-water moored concept. When such events occur, plastic debris won’t be captured by the system and will escape into the open ocean. Such an event will therefore be decreasing the capture efficiency.

To model and quantify plastic loss due to wave overtopping and under-flowing, the ideal approach would be to use a nonlinear 3D CFD method including hydro-elasticity of the barrier structure. Given the size of the problem and the number of conditions that need to be simulated to characterize the design space of the system, the use of such a method is computationally very expensive and therefore unrealistic. Therefore, the objective of this work is to propose an alternative method.

This paper presents a method of quantifying plastic loss by coupling a hydrodynamic solver to a 2D CFD solver. A hydrodynamic model is set up to predict the dynamics of the boom. A 2D CFD model with imposed motion is used to analyze the local effects of wave overtopping. From there, wave overtopping events along the barrier system are analyzed and quantified using the results found in the 2D CFD study.

Commentary by Dr. Valentin Fuster
2018;():V002T08A020. doi:10.1115/OMAE2018-78077.

The capability of wave generation and absorption in a viscous flow solver becomes important for achieving realistic simulations in naval and offshore fields. This study presents an efficient generation of nonlinear wave fields in the viscous flow solver by using a nonlinear potential solver called higher-order spectral method (HOS). The advantages of using a fully nonlinear potential solver for the generation of irregular waves are discussed. In particular, it is shown that the proposed method allows the CFD simulation to start at the time and over the space of interest, retrieved from the potential flow solution. The viscous flow solver is based on the open source library OpenFOAM. The potential solvers used to generate waves are the open source solvers HOS-Ocean and HOS-NWT (Numerical Wave Tank). Several simulation parameters in the CFD solver are investigated in the present study. A HOS wrapper program is newly developed to regenerate wave fields in the viscous flow solver. The wrapper program is validated with OpenFOAM for 2D and 3D regular and irregular waves using relaxation zones. Finally, the extreme waves corresponding to the 1000 year return period condition in the Gulf of Mexico are simulated with the viscous flow solver and the wave elevation is compared with the experiments.

Commentary by Dr. Valentin Fuster
2018;():V002T08A021. doi:10.1115/OMAE2018-78158.

For the estimation of wave loads on offshore structures, relevant extreme wave events need to be identified. In order to achieve this, long term wave simulations of relatively large scales need to be performed. Computational Fluid Dynamics (CFD) based Numerical Wave Tanks (NWT) with an interface capturing two-phase flow approach typically require too large computational resources to achieve this efficiently. They are more suitable for the near-field hydrodynamics of steep and breaking wave impacts on the structures. In the current paper, a three-dimensional non-hydrostatic wave model is presented. While it also solves the Navier-Stokes equations, it employs an interface tracking method for the calculation of the free surface location. The algorithm for the simulation of the free surface is based on the continuity of the horizontal velocities along the vertical water column. With this approach, relatively fewer cells are needed in the vicinity of the air-water interface compared to CFD based NWTs. With coarser grids and larger time steps, the wave propagation can be accurately predicted. The numerical model solves the governing equations on an rectilinear grid, which allows for the employment of high-order finite differences. For time stepping, a fractional step method with implicit treatment of the diffusion terms is employed. The projection method is used for the calculation of the non-hydrostatic pressure. The resulting Poisson equation is solved with Hypres geometric multigrid preconditioned conjugated gradient algorithm. The numerical model is parallelized following the domain decomposition strategy and MPI communication between the individual processors.

In the current paper, the capabilities of the new wave model are presented by comparing the wave propagation in the tank with the CFD approach in a 2D simulation. Further, a 3D simulation is carried out to determine the wave forces on a vertical cylinder. The calculated wave forces using the new approach is compared to that obtained using the CFD approach and experimental data. It is seen that the new approach provides a similar accuracy to that from the CFD approach while providing a large reduction in the time taken for the simulation. The gain is calculated to be about 4.5 for the 2D simulation and about 7.1 for the 3D simulation.

Commentary by Dr. Valentin Fuster
2018;():V002T08A022. doi:10.1115/OMAE2018-78288.

Using the computational fluid domain for propagation of ocean waves have become an important tool for the calculation of highly nonlinear wave loading on offshore structures such as run-up, wave slamming and non-linear breaking wave kinematics.

At present, there are many computational fluid dynamics (CFD) codes available which can be employed to calculate water wave propagation and wave induced loading on structures. For practical reasons, however, the use of these codes is often limited to the propagation of regular uni-directional waves initiated very close to the structure, or to investigating the properties and loading due to measured waves by fitting a numerical wave to a measured wave profile.

The present paper focuses on the propagation of steep irregular and short crested wave groups up to the point of breaking. Indeed, this is challenging because of the highly nonlinear behavior of irregular wave groups as steepness increases and they approach the point of breaking. The complexity further increases with the introduction of short-crestedness requiring computation in a large 3-dimentional domain.

Two CFD codes are used in this comparison study which are believed to be well conditioned for wave propagation, the commercial code ComFLOW (available through the ComFLOW JIP project) and the open-source code BASILISK. The primary objective of this paper to show the two CFD codes capability of recreating measured irregular wave groups, using the known linear wave components from the model test as input to fluid domain. Wave elevation is measured at several locations in the close vicinity of the focus point. The comparison is carried out for a selection of events with variation in steepness, wave spreading and wave spectrum.

Commentary by Dr. Valentin Fuster
2018;():V002T08A023. doi:10.1115/OMAE2018-78369.

The Moving Particle Semi-implicit (MPS) method has been proven effective to simulate violent flows such as dam-break flow, liquid sloshing and so on. But the low computational efficiency is one disadvantage of MPS. In the field of scientific computations, GPU based acceleration technique is widely applied to reduce the computation time of various numerical methods. In this paper, an in-house solver MPSGPU-SJTU is developed based on modified MPS method and GPU acceleration technique. A three-dimensional (3-D) dam-break flow is simulated by present solver and the validity and accuracy of GPU code are investigated by comparing the results with those by other researches. By comparisons, the flow field of GPU-based calculation is in better agreement with the experiment. In addition, the computation times of GPU and CPU solvers are compared to demonstrate the effect of GPU acceleration technique on the computational efficiency of MPS method.

Commentary by Dr. Valentin Fuster
2018;():V002T08A024. doi:10.1115/OMAE2018-78387.

In the present study, three-layer-liquid sloshing in a rigid tank is simulated based on the newly developed multiphase MPS method. Firstly, the multiphase MPS method is introduced in detail, including the basic particle interaction models and the special interface treatments employed to extend single phase MPS solver to multiphase flows simulations. The new multiphase MPS method treats the multifluid system as the multi-density and multi-viscosity fluid, thus only a single set of equations needs to be solved for all phases. Besides, extra density smoothing technique, interparticle viscosity model and surface tension model are included in the present method for interface particles. The new multiphase MPS method is then applied to simulate three-layer-liquid sloshing in a rigid tank and verified through comparison with the experiment conducted by Molin et al. [1]. The predicted motion of interfaces by the present method shows a good agreement with the experimental data and other numerical results.

Commentary by Dr. Valentin Fuster

CFD and FSI: Risers and Pipelines

2018;():V002T08A025. doi:10.1115/OMAE2018-77063.

Fairings have historically been known to achieve in-line drag coefficients (Cdx) of approximately 0.60 across the Reynolds number (Re) range of 100,000 to 1,000,000, typical for the offshore environment [1]. The recent development of helically grooved drill riser buoyancy was shown to achieve Cdx values of 0.65 for this Re range [2], presenting a strong alternative to fairing products especially considering the additional installation, storage and maintenance requirements of fairings.

Therefore it is the purpose of this paper to investigate possible fairing designs capable of achieving even lower Cdx values where fairings can still be beneficial in further reducing drag loading. This paper proposes a non-parallel reduced chord horseshoe (RCH) fairing design and is analysed using computational fluid dynamics (CFD) in 3-d using the transient k-epsilon (Reynolds-averaged Navier-Stokes) turbulence model. The modelling approach is validated against tow tank test data of a previous teardrop-shaped (TD) fairing design which showed good agreement with published, peer-reviewed literature.

It was found CFD simulations with axially continuous fairings provide artificially low Cdx values due to the absence of fairing end-effects and gaps between fairing sections. In essence, an infinitely long and uninterrupted fairing in the riser axial dimension is not realistic. Incorporation of this discontinuity sees a significant increase in Cdx compared to the axially continuous fairing configuration. Although this is the case, it was found Cdx of approximately 0.48 or lower is achievable for the entire offshore Re range for the discontinuous fairing configuration (assuming a chord/diameter ratio of 2.0). Larger chord/diameter ratios would provide lower Cdx at the cost of a longer chord length which may impact fairing installation efficiency. Longer axial lengths would also achieve lower Cdx but with the risk of flutter instability.

This development in RCH fairing design sees a possible option for further fairing applicability to offshore drilling operations where lower drag is desirable beyond that offered by the helically grooved buoyancy.

Commentary by Dr. Valentin Fuster
2018;():V002T08A026. doi:10.1115/OMAE2018-77133.

Flow interference between two identical circular cylinders in side-by-side arrangement with one stationary and the other forced to oscillate in the transverse direction are studied. Direct numerical simulations are performed by Lattice Boltzmann Method (LBM) with a constant Reynolds number of 100. We consider four representative pitch ratios, T/D, ranging from 1.2 to 4, corresponding to four distinct flow patterns for two stationary side-by-side cylinders. The forced oscillation is fixed at a constant small amplitude of A/D = 0.1. A wide range of dimensionless oscillating frequency (fe/fs = [0.5, 2]) is examined. The results show that the response state of flow around two side-by-side cylinders when one cylinder is forced to vibrate is quite different from that of the corresponding stationary system. Four response states are identified according to the different characteristics on the power spectra and phase portrait of lift forces on cylinders. In addition, hydrodynamic forces on the cylinders are analyzed in terms of root-mean-square and time-averaged quantities. It is found that the pitch ratio, oscillating frequency and response state play different roles in determining the force quantities.

Commentary by Dr. Valentin Fuster
2018;():V002T08A027. doi:10.1115/OMAE2018-77298.

Slug flow appearance in a multiphase-carrying riser with a long tie-back distance and deeper water is inevitable, depending on the operational and environmental conditions. Several state-of-the-art technologies in mitigating the effects of internal slug flows might not be completely effective or cost-efficient. In addition to the slug excitation, the external current flows can also affect the riser structural behavior and integrity by the presence of vortex-induced vibration (VIV). This study aims to investigate and understand the behavior of slug-conveying catenary riser under uniform and random slug excitations, in combination with VIV. The steady-state slugs are considered and modelled by a series of liquid and gas phases flowing at certain rates inside the riser pipe. Each slug unit consists of a slug liquid (oil, water or their mixture) and gas pocket. In the uniform slug flow cases, all slug units have their equal slug liquid lengths. Time-domain simulations are conducted for different slug units of 20D, 30D, 40D and 50D, where D is the pipe internal diameter, and for different internal flow rates. The non-uniform slug flow case is considered by randomly generating the time-varying slug liquid and unit lengths. Multi-frequency oscillations of the catenary riser are observed, triggered by the transient slug excitations rendering the fundamental vibration mode which is sustained over the ensuing steady-state slugging period. The random slug-induced vibration (SIV) entails larger response amplitudes which are critical from the fatigue life viewpoint, especially when VIV is also accounted for. For riser SIV analysis, only in-plane response is observed; nevertheless, the interaction of riser SIV and VIV generates both in-plane and out-of-plane responses with larger 3-D dynamic responses, deformations and stresses. Such combined SIV and VIV should be specially considered during the riser analysis and design by also taking into consideration the travelling random-like or intermittent slug flows.

Commentary by Dr. Valentin Fuster
2018;():V002T08A028. doi:10.1115/OMAE2018-77299.

Subsea jumper is the steel pipe structure to connect wellhead and subsea facilities such as manifolds or processing units in order to transport the produced multiphase flows. Generally, the jumper consists of a goalpost with two loop structures and a straight pipe between them, carrying the multiphase oil and gas from the producing well. Due to the jumper pipe characteristic geometry and multi-fluid properties, slug flows may take place, creating problematic fluctuating forces causing the jumper oscillations. Severe dynamic fluctuations cause the risk of pipe deformations and resonances resulting from the hydrodynamic momentum/pressure forces which can lead to unstable operating pressure and decreased production rate. Despite the necessity to design subsea jumper with precise prediction on the process condition and the awareness of slug flow risks, it is challenging to experimentally evaluate, identify and improve the modified design in terms of the facility scale, time and cost efficiency. With increasing high computational performance, numerical analysis provides an alternative approach to simulate multiphase flow-induced force effects on the jumper. The present paper discusses the modelling of 3-D flow simulations in a subsea jumper for understanding the development process of internal slug flows causing hydrodynamic forces acting on the pipe walls and bends. Based on the fluctuating pressure calculated by the fluid solver, dynamic responses of the jumper pipe are assessed by a one-way interaction approach to evaluate deformation and stress. A potential resonance is discussed with the jumper modal analysis. Results from the structural response analyses show dominant multi-modal frequencies due to intermittent slug flow frequencies. Numerical results and observed behaviors may be useful for a comparison with other simulation and experiment.

Commentary by Dr. Valentin Fuster
2018;():V002T08A029. doi:10.1115/OMAE2018-77305.

The problem of Vortex-Induced Vibrations (VIV) on spool and jumper geometries is known to present several drawbacks when approached with conventional engineering tools used in the study of VIV on risers. Current recommended practices can lead to over-conservatism that the industry needs to quantify and minimize within notably cost reduction objectives.

Within this purpose, the paper will present a brief critical review of the Industry standards and more particularly focus on both experimental and Computational Fluid Dynamic (CFD) approaches. Both qualitative and quantitative comparisons between basin tests and CFD results for a 2D ‘M-shape’ spool model will be detailed. The results presented here are part of a larger experimental and numerical campaign which considered a number of current velocities, heading and geometry configurations. The vibratory response of the model will be investigated for one of the current velocities and compared with the results obtained through recommended practices (e.g. Shear7 and DNV guidelines).

The strategy used by the software K-FSI to solve the fluid-structure interaction (FSI) problem is a partitioned coupling solver between fluid solver (FINE™/Marine) and structural solvers (ARA).

FINE™/Marine solves the Reynolds-Averaged Navier-Stokes Equations in a conservative way via the finite volume method and can work on structured or unstructured meshes with arbitrary polyhedrons, while ARA is a nonlinear finite element solver with a large displacement formulation.

The experiments were conducted in the BGO FIRST facility located in La Seyne sur Mer, France. Particular attention was paid towards the model design, fabrication, instrumentation and characterization, to ensure an excellent agreement between the structural numerical model and the actual physical model. This included the use of a material with low structural damping, the performance of stiffness and decay tests in air and in still water, plus the rationalization of the instrumentation to be able to capture the response with the minimum flow perturbation or interaction due to instrumentation.

Commentary by Dr. Valentin Fuster
2018;():V002T08A030. doi:10.1115/OMAE2018-77393.

Migration of particles in pipe flow is commonly seen in offshore engineering, such as vertical transport of ores in deep sea mining. As the basis of the investigation on fluid-particle two-phase flow, the interaction of two spheres in upward pipe flow is studied by direct numerical simulations in this paper. The pipe flow is set as Poiseuille flow; the Reynolds number is no more than 1250. The dynamic responses of the spheres and the flow pattern are analyzed at different flow velocity. When compared to the sedimentation of two spheres in quiescent flow, the trailing sphere in Poiseuille flow will never surpass the leading one in Poiseuille flow. As the flow velocity increases in the pipe, the spheres are easier to separate after collision. When the flow velocity exceeds a critical value, the spheres will never collide.

Topics: Pipe flow
Commentary by Dr. Valentin Fuster
2018;():V002T08A031. doi:10.1115/OMAE2018-77751.

Mixing in pipe junctions can play an important role in exciting force and distribution of flow in pipe network. This paper investigated the cross pipe junction and proposed an improved plan, Y-shaped pipe junction. The numerical study of a three-dimensional pipe junction was performed for calculation and improved understanding of flow feature in pipe. The filtered Navier–Stokes equations were used to perform the large-eddy simulation of the unsteady incompressible flow in pipe. From the analysis of these results, it clearly appears that the vortex strength and velocity non-uniformity of centerline, can be reduced by Y-shaped junction. The Y-shaped junction not only has better flow characteristic, but also reduces head loss and exciting force. The results of the three-dimensional improvement analysis of junction can be used in the design of pipe network for industry.

Commentary by Dr. Valentin Fuster
2018;():V002T08A032. doi:10.1115/OMAE2018-78440.

Water waves play an important role in local scour around subsea pipelines laid on the sandy seabed, especially in shallow water regions. In this paper, a two-dimensional numerical model is employed to predict local scour around submarine pipelines under water waves in shoaling condition. The motion of water under waves is simulated by solving the Reynolds Averaged Navier-Stokes (RANS) equations. The evolution of the seabed surface near the pipeline is predicted by solving the conservation of the sediment mass, which transport in the water in the forms of bed load and suspended load. The main aim of this study is to investigate the effect of the seabed slope on the scour profiles and scour depth. To achieve this aim, numerical simulations of scour around a pipeline on a flat seabed and on a slope seabed with a slope angle of 15° are conducted for various wave conditions.

Topics: Pipelines
Commentary by Dr. Valentin Fuster

CFD and FSI: Ship and Floating Systems

2018;():V002T08A033. doi:10.1115/OMAE2018-77191.

Predicting ship maneuverability is one of the important topics in ship engineering. However because of the huge difference between model and full scale Reynolds number (Re), it is almost impossible to predict full scale ship maneuverability using conventional methods such as model test. On the other hands, with the developments of computational technologies and computational fluid dynamics (CFD) techniques, CFD simulations are widely applied on ship maneuvering problems (e.g. Stern et al., 2011). Moreover some of the researchers start the CFD simulation with full scale Re especially on propulsion problems (e.g. Tezdogan et al., 2015) which showing reasonable results.

Therefore, in this paper, captive maneuvering simulations (rudder angle test) in model/full scale Re on KVLCC2 are carried out using Reynolds-averaged Navier–Stokes (RANS) solver NAGISA (Ohashi et al., 2014) with the overset gird method UP_GRID (Kodama et al., 2012). And the results between model and full scale simulations are compared in maneuvering coefficients and flow field to reveal the scale effect on ship maneuverability.

Commentary by Dr. Valentin Fuster
2018;():V002T08A034. doi:10.1115/OMAE2018-77216.

This paper presents the application of Computational Fluid Dynamics (CFD) simulations to the heaving and rolling motion of the planing craft under different speeds and centers of gravity. Comparing the flow lines, the pressure distribution at the bottom of the boat, the heave and the trim angle before instability with those elements after instability, a critical trim angle results in the early separation of the air in the bow. Meanwhile, due to effect of aerodynamic lift, the bow is lifted, which eventually leads to instability of the hull. Forward or upward movement of the center of gravity may eliminate or postpone the porpoising, the backward center of gravity may result in the unstablity of the ship. Serious porpoising is random and irregular. It will damage the structure of the hull, affect the maneuverability of the ship and threaten the safety of the crew.

Commentary by Dr. Valentin Fuster
2018;():V002T08A035. doi:10.1115/OMAE2018-77218.

Experiments regarding vortex-induced vibration (VIV) on floating circular cylinders with low aspect ratio, L/D = 0.5, and different free-end conditions were carried out in a recirculation water channel. The floating circular cylinders were elastically supported by a set of linear springs to provide low structural damping on the system. Four different free-end corner shape conditions were tested, namely r/R = 0.0, 0.25, 0.5 and 1.0; where r/R is the relation between chamfer rounding radius, r, and the radius of cylinder, R. These different free-end conditions were selected to promote changes in the structures shedding around the free end of the cylinder. The aims were to understand the free-end effects on the VIV of floating circular cylinders with very low aspect ratio. The range of Reynolds number covered 2,800 < Re < 55,400. All the results presented here complement the work presented previously for a floating circular cylinder with L/D = 2.0 by Gambarine et al. (2016) [6] - Experimental study of the influence of the free end effects on vortex-induced vibration of floating cylinder with low aspect of ratio, OMAE2016-54623. The present results showed that the amplitudes in both directions were the highest for the semi-sphere case, r/R = 1.0. The amplitudes were almost the same for the other radius values, 0.0 < r/R ≤ 0.5; in which the maximum amplitudes decreased with increasing the corner radius. A critical value, L/Dcrit = 0.5, in which only the free-end structures affect the VIV behavior of the cylinder piercing the free-surface could be stated. The conclusion was that the cylinder free-end affects the VIV behavior for cylinders with very low-aspect ratio.

Commentary by Dr. Valentin Fuster
2018;():V002T08A036. doi:10.1115/OMAE2018-77241.

This paper presents CFD to study the hydrodynamic performance for the high-speed, multi-hull Catamaran advancing in calm water. It uses inhouse computational fluid dynamics (CFD) code to solve RANS equation coupled with six degrees of freedom solid body motion equations. RANS equations are solved by finite difference method and PISO arithmetic. Computations have been made using structured grid with overset technology. Turbulence models used the anisotropic two equations Shear Stress Transport (SST) k-ω model. Single phase level set was used for free surface simulation. A good agreement on the resistance prediction between CFD and experimental fluid dynamics (EFD) has been observed (on the resistance prediction of about 4.0%). Differences between CFD and EFD have been seen for the 3 degrees of freedom (3DOF) motion, whereas larger discrepancy is observed for the sinkage and trim estimation (about 8.0%).

Commentary by Dr. Valentin Fuster
2018;():V002T08A037. doi:10.1115/OMAE2018-77284.

Numerical modelling of floating bodies is still being a very challenging issue, especially for large body displacements. Despite of the good performance of potential flow models in predicting floating body dynamics, there are still physical processes which are not well reproduced with that approximation. Their strong assumptions yield a lack of accuracy when high non-linear effects become predominant. In addition, the presence of restrictions to motion induced by mooring elements also introduces additional non-linear features which are sometimes out of the framework of the use of potential flow models. The use of CFD approach overcomes potential model limitations especially for non-linear effects. When CFD models are applied to solve waves and current interaction with floating bodies, several issues arise such as the numerical treatment of the floating element, mooring implementation and also the computational cost. Although several approaches are available in literature regarding the numerical implementation of floating bodies, the use of the Overset mesh appears as the more suitable one for large body displacement. Although accurate results have been obtained with re-meshing or even morphing techniques, large mesh deformation can yield into non-acceptable skewness and aspect ratio for the cells, consequently inducing numerical instabilities. In this work, we will present a numerical analysis of wave and current interaction with floating bodies. The objective of the work is to present a set of numerical implementations performed in OpenFOAM environment with the use of the Overset mesh method to study moored floating body dynamics due to the combined action of waves and current.

The implementations, included in IHFOAM (www.ihfoam.ihcantabria.com) are a new set of boundary conditions to generate waves and current without the use of relaxation zones. The main consequence is that the computational cost can be reduced due to the use of smaller domains. In addition, the implementation of mooring will be also presented in order to extend the use of the model to realistic conditions. Numerical model predictions compared with laboratory data of wave interaction with moored floating bodies have been performed showing a high degree of agreement. Comparison of floating body displacement and mooring tension will be presented. The combined effect of waves and current, traveling in the same and in opposite directions than waves, and their interaction with floating bodies and mooring will be also studied. Results will show the applicability of current method to model floating bodies.

Commentary by Dr. Valentin Fuster
2018;():V002T08A038. doi:10.1115/OMAE2018-77308.

Simulating the hydrodynamics of deformable, floating structures using a partitioned strategy poses a major challenge when the ratio of the added mass to the structural mass is considerate. Existing computational procedures for fluid-structure interaction become less efficient or even unstable. In these situations, it is advisable to modify the coupling to allow the fluid to respond better to the solid motions. A simultaneous solution of the equations governing fluid and solid-body would be a stable choice but is often not feasible. Usually the numerical problems are taken care of with subiterations between fluid and structure, but their convergence can be slow. In this paper we present a more powerful, quasi-simultaneous approach, which tries to mimic a fully simultaneous coupling in an affordable way. It makes use of a simple approximation of the body dynamics, based on the (6 DOF) solid-body modes and the main elastic modes of the structure. The method will be demonstrated in offshore practice, with a falling life boat, a floating CALM buoy, an elastic membrane and a rubber gate.

Commentary by Dr. Valentin Fuster
2018;():V002T08A039. doi:10.1115/OMAE2018-77327.

The motion of surface ship in wave environments is fully three-dimensional unsteady motion and includes complex coupling with hydrodynamic force and dynamic motion of the rigid body. This paper presents simulations of the KCS model with motions involve pitch and heave in regular head waves. Computations were performed with an in-house viscous CFD code to solve RANS equation coupled with six degrees of freedom (6DOF) solid body motion equations and dynamic overset grids designed for ship hydrodynamics. RANS equations are solved by finite difference method and PISO arithmetic. Level-set method is used to simulate the free surface flow. The simulation geometry includes KCS hull and rudder under three conditions with three wave length and wave height combinations and two velocities (Fr = 0.26 and 0.33). Total resistance coefficient CT, heave motion z and pitch angle θ have been compared between CFD and EFD. Comparisons show that pitch and heave are much better predicted than the resistance. In the first section, simulations considered only 2 degrees of freedom (heave and pitch), for the second section, numerical simulation added the rolling motion to study the KCS in regular head waves. The second simulation cases were carried out with the same velocity and wave length and amplitude combination as the first cases. Comparisons of heave and pitch motion between 2DOF simulations and 3DOF simulations were presented in this paper. Results show the difference of heave motion z and pitch angle θ between the 2DOF and 3DOF-simulasions. In both cases the free surface were studied as an example of the flow generated by the ship pitching and heaving.

Commentary by Dr. Valentin Fuster
2018;():V002T08A040. doi:10.1115/OMAE2018-77330.

It is critical to be able to estimate a ship’s response to waves, since the added resistance and loss of speed may cause delays or course alterations, with consequent financial repercussions. Traditional methods for the study of ship motions are based on potential flow theory without viscous effects. Results of scaling model are used to predict full-scale of response to waves. Scale effect results in differences between the full-scale prediction and reality. The key objective of this study is to perform a fully nonlinear unsteady RANS simulation to predict the ship motions and added resistance of a full-scale KRISO Container Ship. The analyses are performed at design speeds in head waves, using in house computational fluid dynamics (CFD) to solve RANS equation coupled with two degrees of freedom (2DOF) solid body motion equations including heave and pitch. RANS equations are solved by finite difference method and PISO arithmetic. Computations have used structured grid with overset technology. Simulation results show that the total resistance coefficient in calm water at service speed is predicted by 4 .68% error compared to the related towing tank results. The ship motions demonstrated that the current in house CFD model predicts the heave and pitch transfer functions within a reasonable range of the EFD data, respectively.

Commentary by Dr. Valentin Fuster
2018;():V002T08A041. doi:10.1115/OMAE2018-77331.

Making CFD with the capability of predicting ship scale design performance, rather than relying on scale model tests will help reduce design costs and provide a greater opportunity to develop more energy efficient ship designs. The key objective of this paper is to perform a fully nonlinear unsteady RANS simulation to predict the ship motions and resistance of a full scale DTMB 5415 ship model. The analyses are performed at design speeds, at a certain Fr number, using in-house computational fluid dynamics (CFD) to solve RANS equation coupled with six degrees of freedom (6DOF) solid body motion equations. RANS equations are solved by finite difference method and PISO arithmetic. Computations have been made using structured grid with overset technology. Simulation results shown that the total resistance coefficient in calm water at service speed is predicted by 2.36% error compared to the related towing tank results. The ship resistance for different scale demonstrated that the current in-house CFD model could predict the resistance in a reasonable range of the EFD data. The comparison of flow field for wave pattern for different scale model were analyzed and discussed.

Commentary by Dr. Valentin Fuster
2018;():V002T08A042. doi:10.1115/OMAE2018-77368.

Burner booms, one of the most important pieces of equipment for well testing procedures, are used to burn associated gas or oil-and-gas mixture. This paper first conducts a mesh sensitivity analysis to find a proper grid size. Grid independence is evaluated by the correlation value in different monitoring points. Then, the heat radiation of the burner boom on the semi-submersible drilling platform is analyzed using FDS. Without water curtain, it researches and compares the impact of low, medium and high speed wind condition on heat radiation. Without the wind influence, the simulation on heat radiation is done on the optimized water curtain design. The results show that the water curtain design can efficiently reduce the heat radiation on the platform, which has guiding significance for engineering design.

Commentary by Dr. Valentin Fuster
2018;():V002T08A043. doi:10.1115/OMAE2018-77380.

The effect of wedge angle at a constant submerged volume for a plunger type wave maker on the wave height, wave amplitude ratio and the quality of generated wave is studied numerically for a range of linear wave conditions in this research. The commercial ANSYS-FLUENT finite volume code is used to solve the Navier-Stokes equations using dynamic meshes and a Volume of Fluid (VOF) scheme is used to capture the air-water interface. A second order upwind numerical scheme is used to discretize the convective terms of the momentum equations and the standard SIMPLE algorithm is used for coupling the pressure and velocity based equations. At first the plunger-type wedge shaped wave-maker of Wang is considered numerically for the conditions used in his experiments, over a range of linear wave conditions (H/λ less than or equal to 0.03). After validating the numerical method, the effect of plunger wedge angle on the quality of the generated waves and on the power which is needed to run the wave maker are investigated. From these results, we conclude that the quality of the generated waves reduces with increasing wedge angle, when the submerged volume is fixed.

Topics: Waves , Wedges
Commentary by Dr. Valentin Fuster
2018;():V002T08A044. doi:10.1115/OMAE2018-77767.

The paper proposes a series of numerical investigations performed to test and demonstrate the capabilities of a RANS solver in the area of complex ship flow simulations. Focus is on a complete numerical model for hull, propeller and rudder that can account for the mutual interaction between these components. The paper presents the results of a complex investigation of the flow computations around the hull model of the 3600 TEU MOERI containership (KCS hereafter). The resistance for the hull equipped with rudder, the POW computations as well as the self-propulsion simulation are presented. Comparisons with the experimental data provided at the Tokyo 2015 Workshop on CFD in Ship Hydrodynamics are given to validate the numerical approach in terms of the total and wave resistance coefficients, sinkage and trim, thrust and torque coefficients, propeller efficiency and local flow features. Verification and validation based on the grid convergence tests are performed for each computational case. Discussions on the efficiency of the turbulence models used in the computations as well as on the main flow features are provided aimed at clarifying the complex structure of the flow around the stern.

Commentary by Dr. Valentin Fuster
2018;():V002T08A045. doi:10.1115/OMAE2018-77776.

The application of a discrete mooring model for floating structures is presented in this paper. The method predicts the steady-state solution for the shape of an elastic cable and the tension forces under consideration of static loads. It is based on a discretization of the cable in mass points connected with straight but elastic bars. The successive approximation is applied to the resulting system of equations which leads to a significant reduction of the matrix size in comparison to the matrix of a Newton-Raphson method. The mooring model is implemented in the open-source CFD model REEF3D. The solver has been used to study various problems in the field of wave hydrodynamics and fluid-structure interaction. It includes floating structures through a level set function and captures its motion using Newton and Euler equations in 6DOF. The fluid-structure interaction is solved explicitly using an immersed boundary method based on the ghost cell method. The applications show the accuracy of the solver and effects of mooring on the motion of floating structures.

Commentary by Dr. Valentin Fuster
2018;():V002T08A046. doi:10.1115/OMAE2018-78046.

Accurately modelling a self-propelled vessel in a large amplitude seaway with CFD is very expensive and practically out of reach. The expense is due to the very small numerical time-step required for the propeller rotation and the large mesh size. A method for accurately modelling a propeller while reducing computational cost is desirable. This paper describes the first step towards developing a body force propeller model for unsteady conditions. The purpose of this study is to train a semi-empirical algorithm to accurately prescribe the unsteady body force to model the propeller. The MOERI Container Ship propeller is analyzed with RANS CFD. Open water test data is compared to the RANS CFD results of a steady Moving Reference Frame approach. Harmonic surge is applied to a transient rotating mesh model in open water and the behind condition.

Topics: Propellers , Surges
Commentary by Dr. Valentin Fuster
2018;():V002T08A047. doi:10.1115/OMAE2018-78097.

The large-scale presence of debris is a recurrent issue in the Madeira River, located on Amazon rainforest, North of Brazil, and it is a major concern for the Santo Antonio hydropower plant, located at this region. In order to avoid the abundant amount of debris, floating structures called log booms are installed across the river to retain and deflect them. This paper aims to present the methods used to investigate the structural characteristics of a truncated scale model of a log boom line, through water proof strain-gauges and load cells in hydrodynamic experiments. For that, the model was towed along the model basin of the Institute for Technological Research and wooden scale logs were included to simulate the log jam phenomenon. The paper covers experiment methods, from model design to setting of data acquisition devices and system, characteristics of the experimental runs, and further data analysis. The influence of the scale debris on the structural elements are presented, which can leads to develop a correlation model to scale the fluid-structure interactions in the real prototype.

Commentary by Dr. Valentin Fuster
2018;():V002T08A048. doi:10.1115/OMAE2018-78106.

Debris containment grid is an important part of hydroelectric power plant, since it retains objects, preventing damage to the turbine. In the case of the Santo Antonio hydropower plant, located in the Amazon rainforest, in the north of Brazil, the most significant debris are logs. This paper aims to analyze the interaction between several log boom modules (type of debris containment grids developed specifically for containing logs) present in a debris containment line present in Santo Antonio hydropower plant, as well as its interactions with the fluid, varying the advance velocity and side-slip angle. The analysis of the fluid-body interaction is performed using CFD software with Finite Volume Method approach. The problem is divided into steps. Firstly, one log boom module is simulated with several velocities and side-slip flow angle, obtaining a relation between forces, moments and movements. Next, in order to save the expected computational cost, the module is analyzed and compared through the porosity approach. Finally, the analysis of a line with several log boom modules, including the interaction between each module, is carried out. The results of the simulations will allow to perform an analysis of the line stability, obtaining the forces, moments and movements of each log boom module, observing its influence in the log boom line. With a fluid-body hydrodynamic analysis of several modules in a line, data are provided for a structural analysis. Since the porosity approach is used to reduce the computational cost, this paper also contributes to similar cases, with a main interest in larger scales of forces and movements.

Commentary by Dr. Valentin Fuster
2018;():V002T08A049. doi:10.1115/OMAE2018-78372.

This paper describes a set of VIM CFD simulations for a semi-submersible with and without helical strakes. The numerical investigations are conducted under low Reynolds number (Re) using naoe-FOAM-SJTU, a solver developed based on the open source framework OpenFOAM. The self-developed six degree-of-freedom (6DoF) motion module and mooring system module are applied to model motions of semi-submersible and the constraint of mooring lines, respectively. To carry out the calculations, turbulence closure has been chosen the Shear Stress Transport (SST) based Delay Detached eddy simulation (DDES), which uses the RANS model inside the boundary region and LES model outside the boundary area. This allows a realistic simulation within the boundary region where the vortex shedding is taking place, while not using unnecessary amounts of computational power. The Vortex Induced Motion (VIM) of semi-submersible with and without helical strakes was compared against each other for different reduced velocities (Ur). The flow characteristics of the semi-submersible platform is studied based on the characteristics of vortex shedding. For different current incident angles, time histories, trajectories and vorticity of the semi-submersible at different reduced velocities are reported. The result shows our CFD solver naoe-FOAM-SJTU is applicable and reliable to study VIM of semi-submersibles.

Commentary by Dr. Valentin Fuster
2018;():V002T08A050. doi:10.1115/OMAE2018-78676.

In the exceeding water process of underwater vehicles, the existing of trailing cavity will have distinct effects on the hydrodynamic characteristics of vehicles. Recent researches mostly leave gravity effect out of consideration, while the gravity will affect trailing cavity characteristics and then affect the hydrodynamic characteristics of vehicles. In this study, we research the effect of gravity on the trailing cavity of underwater vehicles. Firstly, a complex boundary model which taken partial cavity into consideration is established based on potential flow theory and a program according to this model is written. Because all flow parameter has nothing to do with the radial location, the research problem can be simplified as a two-dimensional problem and studied in polar coordinates. With regularization of the length of the navigation calculation model, infinity to flow velocity and the distance pressure, research domain can be represented by plane in the containing slit. The program is proved to be effective by comparison the results with the data in existing papers. Finally, we calculate the trailing cavity forms of a cone and an underwater vehicle under different cavitation numbers and Froude numbers to study the rules of trailing cavity forms changing with cavitation number and Froude number. Under the same number of Froude, the cavity size of the rear part of vehicle gradually decreases with the increasing cavitation number, and the maximum radius of the cavity equals to the biggest size of the tail radius of the vehicle. Under the same cavitation number bodies, vehicle trailing cavity length gradually increases with the increase of Froude number, but radius of the cavity of the vehicle changed little, the largest radius is equivalent to the tail radius of the vehicle.

Commentary by Dr. Valentin Fuster

CFD and FSI: VIV Physics and Suppression

2018;():V002T08A051. doi:10.1115/OMAE2018-77107.

Suppression of vortex-induced motion (VIM) of multi-columns semisubmersibles is an important factor for the safety design and operating environment of floating platforms. Vortex generation around square columns is a key issue to cause VIM, and thus, an essential point for suppressing VIM is to desynchronize vortex shedding frequency and natural frequency of the system. In our work, we investigate the effect of a square column with the twisted surface on the suppression of VIM. The numerical tool we used in this work is an open source package, OpenFOAM, to solve transient flow pattern around a vibrating square column. The numerical validation of the present study is done by benchmarking with the reported experimental work at moderate Reynolds numbers (Re). In real sea state, the flow around floating structure is turbulent and is in high Re region. We implemented k-ω SST DES turbulence model in the present numerical model to solve complex turbulent flow around the square column with the twisted surface at high Re. By comparison with VIM of a square column, the VIM reduction of a twisted column is significant, which is up to 86% of VIM reduction. The vortex structure, flow separation points and vortex shedding frequency are varied in the spanwise direction of the twisted column, which causes the effect of vortex desynchronization on the VIM. The detailed flow pattern, 3D vortex structure (Q criterion) around the twisted column is demonstrated as well. A remarkable conclusion is made based on the present numerical findings.

Topics: Vortices
Commentary by Dr. Valentin Fuster
2018;():V002T08A052. doi:10.1115/OMAE2018-77142.

The time-varying effect of axial tension has recently attracted increasing focus when investigating vortex-induced vibration (VIV) for flexible cylinders. This paper applies an alternative time domain force–decomposition model to predict VIV response, in which the structural stiffness will be updated at each time step to take the tension variation into account. Firstly, the adopted numerical model is compared against the latest published experimental results of a small-scale cylinder with constant and time-varying tensions. Then, extensive cases of a long flexible cylinder are designed to investigate the tension time-varying effect on structural response and fatigue damage respectively. Several new response characteristics different from the constant tension case are analyzed from the VIV mechanism level. Fatigue analysis also reveals the influence laws of the amplitude and frequency of varying tension. Mathieu-type resonance between VIV and time-varying tension excitation is captured, under which structural response as well as fatigue damage will enlarge significantly. Some conclusions drawn by this research can provide reference at the engineering design stage of marine slender structures.

Commentary by Dr. Valentin Fuster
2018;():V002T08A053. doi:10.1115/OMAE2018-77251.

Here, we experimentally studied the vortex-induced motion (VIM) of a free-standing riser (FSR; 1:65 scale model) with and without a porous metal screen (‘sheath’) placed co-centrically around the buoyancy can (BC). Specifically, we investigated the effects of mesh orientation (square and square rotated 45° in its own plane) and screen-BC diameter ratio (1.1 and 1.2) over a range of flow velocities. BC motions were recorded with a submersible camera; and inline (IL) and cross-flow (CF) amplitudes were then estimated with a motion tracking software. As expected, the installation of the screen changed the natural frequency of the models. Furthermore, the screen increased the reduced velocity at which the lock-in occurred, delaying it by a factor of ∼1.2 and ∼1.4 for the CF and IL respectively. All sheathed models had a prominent reduction in IL amplitudes compared to the bare/unsheathed BC; and at smaller flow velocities, the sheathed models also exhibited significantly lower CF motions, particularly those with a greater screen-BC diameter ratio.

Topics: Buoyancy
Commentary by Dr. Valentin Fuster
2018;():V002T08A054. doi:10.1115/OMAE2018-77451.

Prediction of vortex induced vibration (VIV) for a long-flexible pipe has always been an important concern for the design of risers. Currently, VIV prediction methods are mainly based on the linear beam theory, where the axial tension is treated as time-independent, and the couples between VIV and axial tension are totally ignored. However, experimental results have illustrated strong couples between the axial tension and VIV [1–2]. The purpose of this paper is to develop a time domain VIV prediction model. This model consists of pipe’s structural non-linearity, couplings between axial force, cross-flow/in-line (CF/IL) VIV responses, and the hydrodynamic forces. The hydrodynamic forces are further divided into vortex-induced force in CF and IL directions, and drag force in IL direction. The former one is determined via empirical force model based on forced oscillation test of rigid cylinders. The IL drag coefficients model considering the effects of VIV developed by Song [3] is adopted. VIV responses under these hydrodynamic forces at each time step are solved by Newton-Raphson method. Comparison between present method and the experimental results under uniform flows and shear flows are conducted, which verified the feasibility and reliability of the proposed method. In addition, by comparing the results under constant tension and time-varying tension, it is proved that the time-varying tension has a significant effect on VIV responses, especially under the case of high flow velocity and high vibration mode.

Commentary by Dr. Valentin Fuster
2018;():V002T08A055. doi:10.1115/OMAE2018-77665.

Concern over the Vortex-induced Motions (VIM) acting on offshore structures, with special focus on monocolumn and spar platforms, mooring systems have crucial importance on system movements; the system has thus been transformed into a concept study herein. A floating and rigid circular cylinder with low aspect ratio (L/D = 2) was used in the experiments carried out to investigate the influence of stiffness ratio (kx/ky) on Vortex-Induced Vibration (VIV). The cylinder was mounted in an elastic base composed of four springs with differences in in-line and transverse stiffness, defining: kx/ky ≅ 0.3, 0.5, 1.0, 2.0 and 3.0. The Reynolds number analysed belongs to a range between 0.2 · 104 and 2 · 104. Some good qualitative and quantitative agreements are found for in-line amplitudes, and higher kx/ky systems demonstrate significant oscillation for low relative velocities. This variation can be seen and justified when the XY-plane trajectories were plotted. When kx/ky is defined as 2 and 3, the traditional VIV 8-shape is illustrated for reduced velocities between 3 and 6. In contrast, the other stiffness systems do not show significant movements and, consequently, a negligible XY shape. Roll and pitch degrees of freedom have shown the motions coupled with the transverse and the in-line motions respectively. Moreover, the yaw motion did not present considerable angles. kx/ky = 2 has presented the highest lift force coefficients, without a great difference from the other aspects ratios, though. The drag force coefficient showed constant values for kx/ky = 2 and 3, the smallest results were observed for the system kx/ky = 3.

Commentary by Dr. Valentin Fuster
2018;():V002T08A056. doi:10.1115/OMAE2018-77689.

This paper presents a numerical phenomenological model for a two-degree-of-freedom VIV of a flexibly mounted circular rigid cylinder subject to sinusoidal oscillatory flows. This prediction model is based on the use of double Duffing-van der Pol (structure-wake) oscillators which capture the structural geometrical coupling and fluid-solid interaction effects through system cubic-quadratic nonlinearities. Empirical coefficients are calibrated based on computational fluid dynamics results in the literature for the Keulegan-Carpenter numbers (KC) of 10, 20 and 40, satisfying a reasonable correspondence in amplitude and frequency responses. For KC = 10, the cross-flow vibrations present a single-frequency response. For KC = 20 and 40, cross-flow vibrations have multi-frequency responses. The primary frequency of the response in the cross-flow direction decreases with increasing reduced velocity, except for small values of the reduced velocities. In all KC cases, the in-line vibrations exhibit mostly a single frequency. Overall, parametric studies capture the dependence of response characteristics on the KC, reduced velocity, mass ratio, frequency ratios and empirical coefficients.

Commentary by Dr. Valentin Fuster
2018;():V002T08A057. doi:10.1115/OMAE2018-77708.

Induced vibrations are three-dimensional oscillations in a structure, whereby maximum amplitude is mostly perpendicular to sustained action. In this paper, we discuss the specific physics for how induced-vibrations evolve with space and time in a few example structures. We demonstrate how a sustained action (particularly fluid drag and gravity loading) rotates and reshapes these slender structures. We demonstrate how this then shifts and expands the dynamic nature of the structure, making the structure more receptive to vibrational inducements of any kind. Contrary to historical focus, the structure (not the fluid) primarily determines the physical nature of any induced vibrations, including fluid-induced vibrations.

Commentary by Dr. Valentin Fuster
2018;():V002T08A058. doi:10.1115/OMAE2018-77837.

The Ocean Cleanup designs floating systems to intercept and accumulate plastic marine debris by having a relative velocity towards the plastic drifting with the sea-surface current of the oceans. One of the solutions to create such a velocity difference is by the mean of drift anchors located in a deeper and slower water layer therefore slowing down the system. As a basic design, two main shapes are being investigated: a punctured cylindrical pipe and a cross shape.

In order to compare the general hydrodynamic behavior and the influence of the holes pattern of the different anchor designs, a CFD study is carried out using a Lattice Boltzmann approach in order to characterize the in-line and cross-flow hydrodynamic coefficients.

Commentary by Dr. Valentin Fuster
2018;():V002T08A059. doi:10.1115/OMAE2018-78010.

The helical strakes are now widely being used in offshore riser design for the suppression of vortex induced vibrations (VIV). The purpose of this paper is to investigate the responses and suppression effectiveness of a straked pipe in a more real working conditions of the risers which will endure a kind of “oscillatory flow” due to the relative motions between the fluid around and the risers induced by the top platform motions. Experiments are performed on a flexible straked pipe with pitch length/height of 15D/0.25D in ocean basin. The pipes are forced to harmonically oscillate in various combinations of amplitude and period with Keulegan-Carpenter (KC) number varying between 5 and 165, and maximum reduced velocities from 4 to 12. Responses in both in direction are firstly investigated. Inverse analysis method and the Least Square method are adopted to identify drag coefficients and added mass coefficients. The results show that strakes can reduce the higher frequency responses in both CF and IL direction. Suppression efficiencies of the strakes and are not ideal as expected in oscillatory flow. Moreover, the hydrodynamic coefficients change dramatically under the small KC number and stabilize under the large KC number. The drag coefficients obviously magnify at KC ∼ 20.

Commentary by Dr. Valentin Fuster
2018;():V002T08A060. doi:10.1115/OMAE2018-78316.

This paper presents an investigation of the suppression of vortex shedding of a larger circular cylinder by the interference of smaller rotating wake-control cylinders positioned around its center. Three-dimensional numerical simulations have been conducted at a moderate Reynolds number of 10,000, thus complementing the previous experimental results by offering a better understanding of the physical mechanisms behind the suppression. Visualization of the vortex wakes revealed a complex disruption of the vortex tubes for the higher rotation speeds, with consequent reduction in the mean drag of almost 52% when compared with that of a bare cylinder. Fluctuating lift has also been drastically reduced in 90%. Configurations of control cylinder that can suppress vortex shedding may produce more efficient suppressors for flow-induced vibrations.

Commentary by Dr. Valentin Fuster
2018;():V002T08A061. doi:10.1115/OMAE2018-78443.

In this study, we conducted numerical simulations to compute the hydrodynamic forces acting on a circular cylinder undergoing bidirectional oscillations in still fluid. The simulations correspond to the regime of attached laminar two-dimensional flow at low values of the Keulegan-Carpenter number (KC ≤ 5) and Reynolds numbers from 35 to 1000 based on the primary motion of the cylinder. The effect of a secondary motion transverse to the primary motion having twice the frequency and a fifth of the amplitude of the latter is investigated and the results are compared with the corresponding case of unidirectional motion and theoretical predictions from Stokes–Wang theory. The results for unidirectional motion show that the computed force in-line with the motion agree well with theory for KC < 1 and KCRe > 100. The agreement between computations and theory improves as KC decreases and Re increases. The addition of a secondary motion with different phase angles with respect to the primary motion did not have any observable effect on the force acting along the direction of the primary motion compared to that for the same unidirectional motion, although it had a marked effect on the distribution of vorticity around the cylinder. The forces on the cylinder undergoing bidirectional oscillations could be well predicted from Stokes–Wang theory applied in each individual direction for the range of parameters examined in this study. The present study provides insight into the relationship between the generation of vorticity around an oscillating cylinder and the fluid forces acting on it.

Commentary by Dr. Valentin Fuster
2018;():V002T08A062. doi:10.1115/OMAE2018-78472.

Flexible cylinders, such as marine risers, often experience sustained vortex-induced vibration (VIV). Both helical strakes and fairings are demonstrated to be effective in suppressing VIV, while, helical strakes result in large drag, which increases the rotational angle and bending moment at the riser hang-off location and, fairings are cumbersome in term of storage, installation and maintenance. This study was inspired by the giant Saguaro Cacti which grow in desert region. Saguaro Cacti have shallow root system, but can grow up to fifty feet in height and can withstand very high wind velocities. In this study, numerical simulations of flow past a stationary cactus-shaped cylinder are performed in two-dimensional field at a low Reynolds number of 200. The hydrodynamic coefficients and the vortex-shedding patterns of a cactus-shaped cylinder are compared with those of a circular cylinder. In addition, a set of two cactus-shaped cylinders of tandem arrangement are also studied to investigate the effects of wake. Results showed that a cactus-shaped cylinder can reduce the drag, lift, and Strouhal number, which suggests its potential as an alternative technology to suppress VIV of a riser.

Commentary by Dr. Valentin Fuster
2018;():V002T08A063. doi:10.1115/OMAE2018-78605.

A newly developed Tri-Helically Grooved drilling riser buoyancy module design was tested in the towing tank of SINTEF Ocean in June 2017. This new design aims to reduce riser drag loading and suppress vortex-induced vibrations (VIV). Objectives of the test program were two-fold: to assess the hydrodynamic performance of the design allowing for validation of previous computational fluid dynamics (CFD) studies through empirical measurements, and, to develop a hydrodynamic force coefficient database to be used in numerical simulations to evaluate drilling riser deformation due to drag loading and fatigue lives when subjected to VIV.

This paper provides the parameters of the testing program and a discussion of the results from the various testing configurations assessed.

Tests were performed using large scale, rigid cylinder test models at Reynolds numbers in the super-critical flow regime, defined as starting at a Reynolds number of Re = 3.5 × 105 – 5.0 × 105 (depending on various literatures) and continuing until Re = 3 × 106. Towing tests, with fixed and freely oscillating test models, were completed with both a bare test cylinder and a test cylinder with the Tri-Helical Groove design. Additional forced motion tests were performed on the helically grooved model to calculate lift and added mass coefficients at various amplitudes and frequencies of oscillation for the generation of a hydrodynamic force coefficient database for VIV prediction software.

Significant differences were observed in the hydrodynamic performance of the bare and helically grooved test models considering both in-line (IL) drag and cross-flow (CF) cylinder excitation and oscillation amplitude. For the helically grooved model, measured static drag shows a strong independence from Reynolds number and elimination of the drag crisis region with an average drag coefficient of 0.63.

Effective elimination of VIV and subsequent drag amplification was observed at relatively higher reduced velocities, where the bare test model shows a significant dynamic response. A small level of expected response for the helically grooved model was seen across the lower range of reduced velocities. However, disruption of vortex correlation still occurs in this range and non-sinusoidal and highly amplitude-modulated responses were observed.

Commentary by Dr. Valentin Fuster

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In