Application of Two Phase Eulerian CFD Model to Simulate High Velocity Jet Induced Scour

Submerged water jet causing soil excavation is a typical water-soil interaction process that occurs widely in many engineering disciplines. In hydraulic engineering for instance, a typical example would be scour downstream of headcuts, culverts, or dam spillways. In port and waterway engineering, erosion of the channel bed or quay wall by the propellers of passing ships are also typical water jet/soil interaction problems. In ocean engineering, trenching by impinging high-velocity water jets has been used as an efficient method for cable and pipeline burial. At present, physical modelling and simple prediction equations have been the main practical engineering tool for evaluating scour in these situations. However, with the increasing computational power of modern computers and the development of new Computational Fluid Dynamics (CFD) solvers, scour prediction in such engineering problems has become possible. In the present work three-dimensional (3D) numerical modelling has been applied to reproduce the capability of a pair of water jets to backfill an excavated trench. The simulations are carried out using a state-of-the-art three-dimensional Eulerian two-phase scour model based on the open source CFD software OpenFOAM. The fluid phase is resolved by solving modified Navier-Stokes equations, which take into consideration the influence of the solid phase, i.e., the soil particles. This paper first presents a validation of the numerical model against vertical jet erosion tests from the literature and conducted at HR Wallingford. The results of the model show good agreement with the experimental tests, with the numerical model predicting the scour hole depth and extent with good accuracy. The paper then presents a validation of the model’s ability to reproduce deposition which is evaluated through a comparison with settling velocity data and empirical formulations found in literature, again with the model showing good agreement. Finally, the model is applied to a prototype cable burial problem using a commercially available controlled flow jet excavator. The study found that the use of water jets can be effective (subject to confirmation of the time-scale required for real operations) for performing backfill operations but that the effectiveness is closely related to the type of sediment and selection of an appropriate jet discharge. As a result, in order for the water jet method to be effective for backfill, there is a requirement for a good description of the variation in sediment type along the trench and a requirement for the jet discharge to be varied as different sediment types are encountered.

Experimental Investigation on Vortex-Induced Vibration of Dual Pipes With Unequal Diameters in Tandem

With the massive use of buddle risers and pipelines in deep-water oil production industries, the demand to focus on the research of interference effects of dual pipe has been greatly enhanced. This paper presents interference experiments of dual flexible pipes with unequal diameters under uniform flow with Reynolds numbers ranging from 1.8E3 to 1.1E4. The pipe with larger diameter was set to be the upstream pipe. Various tandem arrangements with wall surface-to-wall surface distances being 3D to 8D were tested, where D is the smaller pipe diameter. Fiber Bragg grating (FBG) strain sensors were used to measure both in-plane and out-of-plane strain responses. Modal superposition method was applied to reconstruct IL mean displacement. Significant interference effect was found under the condition that wall-to-wall gap is smaller than 8D, where CF and IL vibration frequency ratio of downstream pipe equals to 1.0 and IL mean displacement gets smaller compared to those of single pipe in isolation. Moreover, a special ‘capture’ phenomenon, that the dominant vibration frequency and mode of downstream pipe were as the same as that of the upstream pipe subjecting to the uniform flows, was found when wall-to-wall distances were 4D and 8D.

Study on the Suppression of Vortex Induced Oscillation of Cylinder by Forward Jets

Oscillation induced by vortex shedding often causes fatigue damage of slender cylindrical structures in many engineering fields, so it is significant to study the suppression of vortex induced oscillation. We focus on the methods of experimental test to study on the suppression effect of vortex induced oscillation of cylinder with single freedom at Reynolds number of 2.4 × 104. We designed a set of device with forward water jet to suppress vortex induced oscillation. We found the most notable observation from a serials of experiments: reduction of displacement of cylinder exhibits two distinct modes separated by a sudden and very pronounced decrease in the extent of cross-flow displacement at a critical value of the geometric and flow variables followed by what appears to be a saturation regime where no further decrease is observed. These results will provide some meaningful and evaluable references for practical application in ocean engineering.

Numerical Modeling of Turbulent Wall-Bounded Oscillatory Flow and its Effect on Small-Diameter Pipelines

This study investigates the effect of wave-induced boundary layer on the on-bottom stability of small-diameter pipelines laid on the seabed. An ω-based wall boundary condition is adopted, owing to its high mesh resolution down to the viscous sublayer to resolve the flow around the pipeline. By taking into account the wave boundary layer, the present numerical simulations predict required specific gravity for small-diameter pipelines close to the theoretical estimation by Cheng et al. (2016) and, as expected, much smaller than those recommended by DNV-RP-F109.

Numerical Investigation of Heaving Hydrodynamic Behavior of a Single Cylinder and a Dual Coaxial-Cylinder System Using CFD

In this paper, we apply a CFD computer code to study the hydrodynamic behavior of a stand-alone cylinder and a dual coaxial-cylinder system (DCCS) via free-decay motion tests. The geometric proportions of a stand-alone cylinder and the inner and outer cylinder of the DCCS are chosen to be the same as those in [1] and [2], respectively, as ocean wave-energy converter (WEC) devices. Overset mesh based on the commercial code ‘Star-CCM+ 11’ is used to simulate the free-decay motion of the two systems. Five parameters chosen for the CFD implementation are: turbulence model, initial displacement, time step, number of prism layers and mesh size. Results obtained from using different values of these parameters are compared so as to confirm the validity of choices made. The hydrodynamic performance of the stand-alone cylinder and outer cylinder in the DCCS are compared with the experimental results to assess and validate the CFD models. In addition, the heave hydrodynamic coefficients, namely, the added mass and total damping, and ‘resonance’ frequency of the stand-alone cylinder and those of the inner cylinder of the DCCS, with the outer cylinder being fixed, are obtained by using the CFD procedure. The hydrodynamic coefficients of another stand-alone cylinder with the same dimensions as the inner cylinder of the dual-coaxial cylinder are also obtained by simulations. The vorticity-contour plots for the stand-alone cylinder and the outer cylinder in the DCCS in free-decay motion are presented and analyzed. Finally, the results of the three cases are compared to examine the effect of the outer cylinder on the heave hydrodynamic coefficients of the inner cylinder.

Laboratory Experiment of Two-Degree-of-Freedom Vortex-Induced Vibrations of Circular Cylinder in Regular Waves

This paper presents new laboratory experiments of two-degree-of-freedom vortex-induced vibration of a flexibly mounted vertical circular cylinder in regular waves. A new experimental model has been developed and tested in the Wind, Wave & Current Tank at Newcastle University. The system mass ratio is close to 3 and the cylinder aspect ratio based on its submerged length is close to 27. The Stokes first-order wave theory is considered to describe the depth-dependent, horizontal velocity amplitude of the wave flow in the circulating water tank. This wave theory is satisfactorily validated by the wave probe measurement. The effects of cylinder stiffness affecting system natural frequencies are also investigated by using a combination of different spring setups in in-line and cross-flow directions. For each set of springs, VIV tests are performed in regular waves, with flow frequency ranging from 0.4 to 1 Hz and amplitude from 0.01 to 0.09 m. The associated Reynolds number at the water surface is in a range of 1.7 × 103–1.5 × 104. The surface Keulegan-Carpenter number (KC) is in the range of 2 < KC < 28 while the surface reduced velocity (Vr) is in the range of 0.5 < Vr < 16 depending on the implemented spring stiffness. Combined in-line/cross-flow oscillations of the cylinder are measured using two non-intrusive Qualisys cameras and the associated data acquisition system. The spring forces are also acquired using four load cells. Results reveal that, depending on KC and Vr, the cylinder primarily oscillates at the flow frequency in the in-line direction and at an integer (mainly 2, 3 and 4) multiple of the flow frequency in the cross-flow direction. Such occurrence of multi frequencies corroborates other experimental and numerical results in the literature. Several peculiar trajectories are observed, including infinity, butterfly, S and V shapes. The present experimental data of vibration amplitudes and oscillation frequencies in in-line/cross-flow directions as well as response patterns provide new results and improved understanding of VIV in oscillatory flows. These will be useful for the development of an industrial tool in predicting offshore structural responses in waves.

Laminar and Turbulent Flow Past a Hydrofoil Predicted by a Distributed Vorticity Method

A distributed viscous vorticity equation (VISVE) method is presented in this work to simulate the laminar and turbulent flow past a hydrofoil. The current method is proved to be more computationally efficient and spatially compact than RANS (Reynolds-Averaged Navier-Stokes) methods since this method does not require unperturbed far-field boundary conditions, which leads to a small computational domain, a small number of mesh cells, and consequently much less simulation time. To model the turbulent flow, a synchronous coupling scheme is implemented so that the VISVE method can resolve the turbulent flow by considering the eddy viscosity in the vorticity transport equation, and the eddy viscosity is obtained by coupling VISVE with the existing turbulence model of OpenFOAM, via synchronous communication. The proposed VISVE method is applied to simulate both the laminar flow at moderate Reynolds numbers and turbulent flow at high Reynolds numbers past a hydrofoil. The velocity and vorticity calculated by the coupling method agree well with the results obtained by a RANS method.

Experimental and Numerical Assessment of Both Slug and Vortex Induced Vibrations on a Spool Model

The paper addresses the flow-induced response of a rigid spool/jumper. It mainly focuses on its mechanical response resulting from internal intermittent slug flows but also addresses potential coupling with an external excitation due to vortex shedding. These works provide quantitative experimental data that match quite well with existing empirical correlations in terms of slug flow properties. The repeatability of the measurement system has been experienced and underlines promising capabilities. The tests provide exhaustive complementary data regarding the slug flow properties (e.g. pocket length) which will be reused for numerical modelling purpose. The mechanical response of the spool is exhaustively addressed for different regular slug flows. Some correlations are proposed aiming at describing the mechanical responses. The coupling with an additional external current solicitation and Vortex Induced Vibrations (VIV) is discussed and characterized for some conditions. Finally, a discussion on the current Industry Best Practices is introduced in order to challenge the capability of the proposed approaches to described and recover such complex phenomenon. The comparisons underline the weak agreement between experiments and numerical models, opening discussion on the best way to address this physics and the next developments.

Benchmark Case Study of Scale Effect in Self-Propelled Container Ship Squat

A URANS CFD-based study has been undertaken to investigate scale effect in container ship squat. Initially, CFD studies were carried out for the model scale benchmarking squat cases of a self-propelled DTC container ship. In this study, a quasi-static modelling approach was adopted where the hull was fixed from sinking and trimming which is computationally more efficient than dynamic mesh methods that models actual motion directly. Instead, the quasi-static approach allows estimation of the squat base on the recorded hydrodynamic forces and moments. Propulsion of the vessel was modelled by the body-force actuator disc method. Upon successful verification and validation of the model scale self-propelled CFD model against benchmark data, full scale investigations were then undertaken. Validation of the full scale set-up was demonstrated by computing the full scale bare hull resistance in deep, laterally unrestricted water and comparing against the extrapolated resistance of model scale benchmark resistance data. Upon validating the setup, it was used to predict full scale ship squat in confined waters. The credibility of the full scale confined water model was checked by comparing vessel resistance in confined water against the Landweber empirical prediction. To quantify scale effect in ship squat predicitons, the benchmarking squat cases were computed by adopting the validated full scale CFD model with body-force propulsion. Comparison between the full scale CFD, model scale CFD and model scale benchmark EFD squat results demonstrates that scale effect is negligible.

An Uncertainty Evaluation of Different Fidelity Methods to Predict Ship Motions and Structural Loading in Waves

In this study, four approaches are investigated to predict the motions and structural loads on a containership in waves. The Flockstra (1974) containership model is used as the benchmark for this study as extensive experimental data is available to compare to the predictions. The hydrodynamic loads and motions are predicted using strip theory, a zero speed Green’s functions panel method with forward speed correction, a fully unsteady 3D panel method and unsteady RANSE simulations for limited cases. Simulations are performed at Fn = 0.245 in head, stern quartering, and bow quartering seas for wave length to ship length ration λ/L of 0.35–1.40. The accuracy of each method, relative to experimental results, in predicting the amplitudes of heave, pitch, and roll are investigated. Vertical and horizontal bending moments, shear forces, and the torsional moment on the hull at midships and 0.25LBP forward and aft of midships are also calculated and compared with the measured values. Through comparison with experimental data, the relative uncertainty of all four methodologies in predicting both motions and structural loads are assessed and discussed. Overall, all linearized potential flow methods show a large discrepancy with the experimental loads, motivating the need for further studies on non-linear effects for this particular ship type. This paper has been prepared in the framework of the ISSC-ITTC special joint committee on uncertainty quantification in wave load estimation.