1

SPHERIC newsletter 15th issue – December 2012 SPH European Research Interest Community http://wiki.manchester.ac.uk/spheric/ Contact: david.letouze@ec-nantes.fr

D. Le Touzé, Chairman N. Quinlan, Secretary B. Rogers, Webmaster D. Violeau, Newsletter Steering Committee Ecole Centrale de Nantes (France) National University of Ireland, Galway University of Manchester (UK) University of Vigo (Spain) L’Aquila University (Italy) EDF R&D / LNHE (France) VA TECH Hydro (Switzerland) University of Plymouth (UK) Swiss National Supercomputing Centre INSEAN (Italy) ESI-Group Netherlands Technical University of Munich (Germany) Members Ecole Centrale de Lyon (France) Novosibirsk State University (Russia) Technische Universitaet Muenchen (Germany) Univer. Nacional de Educacion a Distancia (Spain) Johns Hopkins University (USA) University of Nottingam (UK) University of Bradford (UK) University of Lancaster (UK) Université de Montpellier (France) Technical University of Madrid (Spain) Shanghai Jiao Tong University (China) Ecole Polytechnique Fédérale de Lausanne (Switzerland) Université du Havre (France) Swiss Federal Inst. of Technology (Switzerland) ERSE SpA (Italy) CIMNE Barcelona (Spain) University of Palermo (Italy) University of Genova (Italy) CEDEX (Spain) University of Pavia (Italy) Dublin Institute of Technology (Ireland) Imperial College London (UK) Institute for Plasma Research (India) BAE Systems (UK) University of Umeå (Sweden) Institut Français du Pétrole (France) University of West Bohemia (Czech Republic) Tarbiat Modares University (Iran) University of West Bohemia (Czech Republic) Universidade de Santiago de Compostela (Spain) CUGRI (Italy) University of Hamburg (Germany) Iran University of Science & Technology University of Cambridge (UK) ASR Limited (New Zealand) Cranfield University (UK) City University London (UK) HydrOcean (France) Laboratório Nacional de Engenharia Civil (Portugal) Aristotle University of Thessaloniki (Greece) Catholic University Leuven (Belgium) University of Calabria (Italy) University of Ljubljana (Slovenia) Virginia Tech (USA) SINTEF Materials and Chemistry (Norway) Manchester Metropolitan University (UK) Aberystwyth University (UK) J P KENNY Sdn. Bhd. (Malaysia) Politecnico di Milano (Italy) Bangladesh Univ. of Engineering and Technology Ritsumeikan University (Japan) Institut fuer Thermische Stroemungsmaschinen (Germany) Sulzer Markets & Technology Ltd (Switzerland) University of Heidelberg (Germany) Fraunhofer-Chalmers Res. Centre for Ind. Math. (Sweden) Simulation Innovation Lab (South Korea) Mustafa Kemal University (Turkey) National Cheng Kung University (China) Amir Kabir University of Technology (Iran) University Technology Malaysia National Nuclear Laboratory (UK) Texas A&M University (USA) Istituto Nazionale di Geophisica e Vulcanologia (Italy) Glasgow Caledonian University (UK) National Technical University of Athens (Greece) University of Exeter (UK) University of Parma (Italy) Kyoto University (Japan) CRS4 (Italy) University of Regina (Canada) University of Auckland (New-Zealand) Alstom Hydro (France) Mara University of Technology (Malaysia) Institute of Mechanics (Vietnam) Instituto Superior Tecnico (Portugal) Kitware (USA)

Editorial: 8th SPHERIC workshop The 8th international workshop organized by the Smoothed Particle Hydrodynamics European Research Interest Community will be the major international event of 2013 focusing entirely on SPH techniques and applications. It will be held in Trondheim (Norway) and organized by the research foundation SINTEF. The workshop will be arranged on June 4-6 with a training day on June 3.

Trondheim is an historic town founded in the Viking era more than a thousand years ago. Today it is the centre of technological development in Norway. Trondheim offers a good selection of sights including the famous cathedral and several museums. In June the nights are bright with only a few hours of darkness. The workshop venue is at Thon Hotel Prinsen located centrally downtown in Trondheim.

The aim of this scientific workshop is to enable experienced researchers using SPH to share and contribute to the development and applications of the SPH method, and enable PhD students to present their work in a favorable atmosphere. The topics will include a large range of applications and technical improvements to SPH including:

Computational Modelling using SPH

� Free Surface and Moving Boundaries Applications

� Solids and Structure

� Multiple Continua and Multi-Phase Flows

� Representation of Boundary Conditions

� Modelling of Viscosity and Turbulence

� Modelling of Incompressible Flows

� Shallow Water SPH

Alternative Formulations, related particle methods and coupled methods

� DPD

� MPS

� DEM coupling to SPH

� Others

Numerics and Fundamentals of SPH

� Theoretical and Numerical Aspects of SPH

� High-Performance Computing

� Hardware Acceleration

� Convergence

� Stability (continued next page)

2

SPHERIC newsletter 15 th issue – December 2012

SPH Applications

� New Applications

� Hydraulic Applications

� Maritime and Naval Architecture Applications

� Process Engineering

� Geotechnical Applications

� Micro Fluidics

� Astrophysics

� Solids and Fracture Mechanics

� Biomechanics

� Disaster Simulations

The important dates are:

� Abstract Submission Deadline: 1st Feb. 2013

� Selected Abstracts: 4th March 2013

� Final Paper Submission:12th April 2013

� Early Registration Deadline: 19th April 2013

� Author Registration Deadline: 19th April 2013

� Final Participant Registration: 21th May 2013

� Training Day: 3rd June 2013

� Workshop: 4th–6th June 2013

For further information, check:

www.sintef.no/spheric2013

Jan Erik Olsen

Selected Recent Publications and Ph.D. Theses on SPH This is a small selection of references recently added to the SPHERIC community’s online catalogue of SPH literature at http://www.citeulike.org/group/3462. Anybody can access and contribute to this database.

Cummins, S. J., Silvester, T. B., Cleary, P. W. (2012), Three-dimensional wave impact on a rigid structure using smoothed particle hydrodynamics, Int. J. Num. Meth. Fluids 69(9):1566. DOI: 10.1002/fld.3690

Ferrand, M., Laurence, D. R., Rogers, B. D., Violeau, D., Kassiotis, C. (2012), Unified semi-analytical wall boundary conditions for inviscid, laminar or turbulent flows in the meshless SPH method, Int. J. Num. Meth. Fluids DOI: 10.1002/fld.3666

Monaghan, J. J. (2012), Smoothed particle hydrodynamics and its diverse applications, Ann. Review Fluid Mech. 44(1):323-346. DOI: 10.1146/annurev-fluid-120710-101220

Macià, F., González, L. M., Cercos-Pita, J. L., Souto-Iglesias, A. (2012), A boundary integral SPH formulation. consistency and applications to ISPH and WCSPH, Progress Theor. Phys. 128(3). DOI: 10.1143/PTP.128.439

Marrone, S., Colagrossi, A., Le Touzé, D., Graziani, G. (2010), Fast free-surface detection and level-set function definition in SPH solvers, J. Comput. Phys. 229:3652-3663. DOI: 10.1016/j.jcp.2010.01.019

Cherfils, J. M., Pinon, G., Rivoalen, E. (2012), JOSEPHINE: A parallel SPH code for free-surface flows, Comput. Phys. Com. 183(7):1468-1480. DOI:

10.1016/j.cpc.2012.02.007

The following theses are available to download in full at http://wiki.manchester.ac.uk/spheric:

Abdelraheem, M.A. (2012), An Improved Incompressible Smoothed Particle Hydrodynamics to Simulate Fluid-Soil-Structure Interactions, Kyushu University.

Marrone, S. (2011), Enhanced SPH modeling of free-surface flows with large deformations, University of Rome, La Sapienza.

Kajtar, J. B. (2010), Smooth lattice general relativity, and SPH simulations of swimming linked bodies, Monash University.

Vacondio, R. (2010), Shallow Water and Navier-Stokes SPH-like numerical modelling of rapidly varying free-surface flows, Università degli Studi di Parma Facoltà di Ingegneria.

Xu, R. (2010), An Improved Incompressible Smoothed Particle Hydrodynamics Method and Its Application in Free-Surface Simulations, University of Manchester.

Federico, I. (2010), Simulating Open-channel Flows and Advective Diffusion Phenomena through SPH Model, Università della Calabria.

Omidvar, P. (2010), Wave Loading on Bodies in the Free Surface Using Smoothed Particle Hydrodynamics (SPH), University of Manchester.

3

SPHERIC newsletter 15th issue – December 2012

Numerical Diffusive Terms and the δ-SPH scheme M. Antuono, A. Colagrossi, S. Marrone, CNR-INSEAN (The Italian Ship Model Basin), 00128 Rome, Italy

One of the main drawbacks of the weakly compressible SPH schemes is the high-frequency spurious numerical noise that affects the density and the pressure fields.

Over the years, different approaches have been proposed to overcome such an issue. Among those, we focus on the use of proper numerical diffusive terms to smooth out the density field. Incidentally, we recall that in weakly-compressible schemes the density field is related to the pressure one through a specific state equation. Then, a noise-free density field corresponds to a noise-free pressure field.

The aforementioned numerical diffusive terms are generally added inside the density equation and they have to satisfy the global continuity equation, that is, the integral version of the continuity equation. Further, they have to reduce to zero when the spatial resolution increases, allowing the recovery of the correct local equations. At a local spatial scale, the numerical diffusive terms smooth out the high-frequency noise and maintain larger spatial scales unaltered.

The first attempts at numerical diffusion in weakly-compressible SPH schemes were done by Ferrari et al. (2009) and Molteni and Colagrossi (2010). They both proposed diffusive terms which correspond to second-order differential operators of the density field (see, for details, Antuono et al. 2012). The use of these terms provides good results for simulations characterized by strong dynamics but, for weaker dynamics, it leads to problems close to the free surface caused by inaccuracies of the numerical diffusive terms. Specifically, it is no longer possible to attain the hydrostatic solution.

In order to eliminate such a problem, Antuono et al. (2010) proposed an improved diffusive term which allowed a correct behaviour for both strong and weak dynamics. As proven by Antuono et al. (2012), this term corresponds to a fourth-order differential operator of the density field. The weakly-compressible scheme which implements such a diffusive term has been called δ-SPH scheme and has been widely validated in different flow conditions.

Figure 1 shows a hydrostatic problem solved by using the diffusive term proposed by Ferrari et al. (2009, left panel) and by using the δ-SPH scheme (right panel). In the former case the hydrostatic solution cannot be achieved and particles start moving upwards. Figure 2 displays a dam break problem solved through the standard SPH model (top panel) and through the δ-SPH scheme (bottom panel). In the latter case, the pressure field appears much more regular than that predicted by the standard SPH scheme.

0 0.25 0.5 0.75 1

0

0.25

0.5

0.75

10.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50

x/L

y/L p/ρgL

t(g/H)½ = 90.5

H/∆x = 25

0 0.25 0.5 0.75 1

0

0.25

0.5

0.75

10.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50

x/L

y/L

t(g/H)½ = 90.5

H/∆x = 25

p/ρgL

Figure 1 – Hydrostatic problem: SPH with the diffusive term of Ferrari et al. (2009, left) and the δ-SPH model (right).

0 1 2 3 4 5

0

0.5

1

1.5 0.00 0.09 0.18 0.27 0.36 0.45 0.55 0.64 0.73 0.82 0.91 1.00

y/H

x/H

t(g/H)½ = 6.48

P1

p/ρgH

0 1 2 3 4 5

0

0.5

1

1.5 0.00 0.09 0.18 0.27 0.36 0.45 0.55 0.64 0.73 0.82 0.91 1.00

y/H

x/H

t(g/H)½ = 6.48

P1

p/ρgH

Figure 2 – Dam break problem: standard SPH (top panel) and the δ-SPH model (bottom panel).

Contact: matteo.antuono@cnr.it

References

Ferrari, A., Dumbser, M., Toro, E.F., Armanini, A. (2009), A new 3D parallel SPH scheme for free-surface flows, Computers & Fluids 38:1203–1217.

Molteni, D., Colagrossi, A. (2010), A simple procedure to improve the pressure evaluation in hydrodynamic context using the SPH, Comput. Phys. Com. 180(6):861–872.

Antuono, M., Colagrossi, A., Marrone, S., Molteni, D., (2010), Free-surface flows solved by means of SPH schemes with numerical diffusive terms, Comput. Phys. Com. 181:532–549.

Antuono, M., Colagrossi, A., Marrone, S, (2012), Numerical diffusive terms in weakly-compressible SPH schemes, Comput. Phys. Com. 183:2570–2580.

4

SPHERIC newsletter 15 th issue – December 2012

SPH Modelling of Plunging Wave Breaking, Surf Zone Turbulence and Wave-Induced Currents Christos V. Makris & Yannis N. Krestenitis, Aristotle University of Thessaloniki, Greece Constantine D. Memos, National Technical University of Athens, Greece

Following our previous work (Makris et al., 2010), SPHysics code v.2 (Gómez-Gesteira et al., 2010) has been thoroughly calibrated and validated against experimental data for wave propagation and weak plunging breaking on a smooth mild sloping beach placed inside a laboratory scale wave flume (Stansby & Feng, 2005). The LES-type Smagorinsky model is used for the viscosity treatment. Spatial resolution is based on the size of expected turbulent eddies. Discretization values ∆x approach the demarcation range between integral turbulence length scales (energy-containing eddies) and Taylor micro-scales (inertial sub-range).

Remarkable visual output (Fig. 1) is further supported by extended quantitative validation through comparison between experimental data and SPHysics results. In this framework, several classic and more sophisticated hydrodynamic features are investigated. Plausible agreement is achieved in terms of wave heights and setup, r.m.s. free-surface elevation fluctuation, wave crest and trough envelopes, throughout the whole computational domain (Fig. 2). Relevant Pearson correlation coefficients vary from 0.9 to 0.97 for most cases. Moreover, ensemble-averaged free-surface elevation and depth-averaged velocities are generally well predicted (Fig. 3).

Figure 1 – Consecutive depictions (PV-Meshless) of simulated (SPHysics) weak plunging wave breaking and consequent turbulent bore formation.

-0.05

0.00

0.05

0.10

0.15

0.20

0 1 2 3 4 5 6 7 8Horizontal Distance, x (m)

Mea

n E

leva

tion

η

, Wav

e H

eigh

t H (

m)

ME_exp H_exp ME_sim H_sim

0.000

0.005

0.010

0.015

0.020

0.025

0.030

0.035

0.040

0.045

0.050

5.5 5.7 5.9 6.1 6.3 6.5 6.7 6.9 7.1 7.3 7.5

R.M

.S.

elev

atio

n flu

ctua

tion

, η

fluct

,rm

s (m

)

ηfluct,rms exp

ηfluct,rms sim

0

0.1

0.2

0.3

0.4

0.5

0.6

5.4 5.6 5.8 6 6.2 6.4 6.6 6.8 7 7.2 7.4Horizontal Distance, x (m)

Cre

st a

nd T

roug

h E

nvel

ope

s (m

)

CE_exp TE_exp Bottom CE_sim TE_sim

Figure 2 – Comparison between experimental data (exp) and SPHysics output (sim), for wave heights and setup [bottom of previous page], r.m.s. free-surface elevation fluctuation [upper], wave crest and trough envelopes [lower].

-0.10

-0.08

-0.06

-0.04

-0.02

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25

Ens

embl

e-A

vera

ged

Sur

face

Ele

vatio

n η

(m

) η_ens exp (m)

η sim (m)

η_ens sim (m)

-0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50

Time t (sec)

Ens

embl

e-A

vera

ged

Dep

th-A

vera

ged

Vel

ocity

U

ens

(m/s

ec)

u_ens exp (m/sec)

u sim (m/sec)

u_ens sim (m/sec)

Figure 3 – Comparison of experimental (exp) against SPHysics (sim) output for ensemble-averaged (blue dash) and real-time (red dash-dot) values of free-surface elevation [upper] and depth-averaged velocity [lower].

5

The Fourier spectra of the simulated turbulent component of horizontal velocities u´ at still surface level is derived (Fig. 4), revealing a trend that follows the –5/3 slope on the log/log scale, typical of isotropic (inertial sub-range) turbulence. This is the case for turbulent wavenumber values of f = 10Hz, somehow continued until the Nyquist filter limit f = 25Hz. Improvement of our previous results (Makris et al., 2010) is clear for high frequency bands, that correspond to either the SPS-treated scales or the smaller of the resolved large eddies. Besides that, preliminary results of residual normal and shear stresses reveal a mild anisotropy in turbulence especially in the vicinity of the initial plunging breaking region. In addition, we focus on the simulation of the wave-induced mean flows in the surf zone, namely the undertow and the Stokes drift (Fig. 5). The period-averaged kinematics for the surf zone is very similar to that of Stansby & Feng (2005), for velocity vector field averaging both over the ‘wet’ period (the time for which a point is immersed in water for a wave cycle) and the actual one. Moreover, the shoreward inversion of the mean flow near the bed (streaming), is qualitatively well predicted by SPHysics. Depth-averaged horizontal volume flux over one period is close to zero, indicating an acceptable level of accuracy for our simulations.

STILL SURFACE u' SPECTRUM

1.E-10

1.E-05

1.E+00

1.E+05

1.E-02 1.E-01 1.E+00 1.E+01 1.E+02 1.E+03Frequency, f (Hz)

Vel

ocity

Spe

ctru

m,

P(f)

(m

2 /sec

3 )

Figure 4 – Fourier spectra of simulated turbulent component of horizontal velocity u´ for the incipient breaking region at still surface level.

5.6 6 6.4 6.8 7.25.4 5.8 6.2 6.6 7 7.40.1

0.2

0.3

0.4

0.5

0.15

0.25

0.35

0.45

Ver

tical

Dis

tanc

e z

(m)

Umean Vectors

BottomCrest EnvelopeTrough EnvelopeWave Setup

5.6 6 6.4 6.8 7.25.4 5.8 6.2 6.6 7 7.4Horizontal Distance x (m)

0.1

0.2

0.3

0.4

0.5

0.15

0.25

0.35

0.45

Ver

tical

Dis

tanc

e z

(m)

Umean Vectors

BottomCrest EnvelopeTrough EnvelopeWave Setup

Figure 5 – Time-averaged vertical distribution of velocity at gauges in the surf zone for ‘wet period’ [upper] and real-time [lower] data. Depiction of undertow and Stokes’ drift regions, delimited by envelopes of wave trough (red dash), crest (blue dash), and setup (green dash-dot). Recurring patterns of periodically concentrated vorticity in a 2D cross-sectional plane are investigated too (Fig. 6).

Evolution of the relevant vorticity field is similar to that of experiments (Stansby & Feng, 2005; Nadaoka et al., 1989). The period-averaged values are as expected with a thick layer of clockwise (positive) vorticity around the trough level and counter-clockwise (negative) near the bed (lower Fig. 6). Concentrated ensemble-averaged vorticity is also apparent in the surf zone (roller, plunger and bore regions) shown as multiple turbulent coherent structures (Fig. 7).

Figure 6 – Recurring vortical patterns during wave breaking [upper graphs]. Period-averaged vorticity field (coherent turbulent structures) [lower].

Figure 7 – Ensemble-averaged vorticity contours (coherent turbulent structures) at gauges in the incipient breaking region [upper] and inner surf zone [lower].

Contact: cmakris@civil.auth.gr

References

Gómez-Gesteira, M., Rogers, B.D., Dalrymple, R.A., Crespo, A.J.C., Narayanaswamy, M. (2010), User guide for the SPHysics code v2.0.

Makris, C.V., Krestenitis, Y.N., Memos, C.D. (2010), SPHysics code validation against a near-shore wave breaking experiment, Proc. 5th SPHERIC Int. Workshop, pp. 245–252.

Nadaoka, K., Hino, M., Koyano, Y. (1989), Structure of the turbulent flow field under breaking waves in the surf zone, J. Fluid Mech. 204:359–387.

Stansby, P.K., Feng, T. (2005), Kinematics and depth-integrated terms in surf zone waves from laboratory measurement. J. Fluid Mech. 529:279–310.

6

SPHERIC newsletter 15th issue – December 2012

Very short term and short term analysis of violent liquid wave impact with SP Giacomo Viccione & Vittorio Bovolin, Dept. Of Civil Engineering, Ponte Don Melillo St., 84084 Fisciano (IT). Eugenio Pugliese Carratelli, C.U.G.Ri. Centre, Piazza Vittorio Emanuele, 84085 Penta di Fisciano (IT)

Liquid impact is the phenomenon according to which a fluid comes in contact and abruptly interacts with a structure. Among the various contexts concerning fluid-solid interaction, the design of seawall structures is probably one of the major interesting challenge in maritime engineering (see e.g. Neves, 2008). As there still is research in progress, the matter has not reached a complete understanding yet, therefore justifying the use of empirical approaches. By carrying out numerical invest-tigations, the research unit of Fisciano (Italy) is focusing on the very early stages – here called very short term – of the phenomenon, where both compressibility and free surface effects certainly play a key role in the temporal developing of the pressure field, trying to connect this initial phase with the next one – defined as short term –, longer in duration and featuring lower pressures. In this latter phase, liquid compressibility does not play any significant role, as anticipated by Peregrin (2003). Our Lagrangian code (Viccione, 2012a), already tested in a variety of conditions, is based on the WCSPH technique with a diffusive term placed in the continuity equation, as suggested by Molteni and Colagrossi (2008). The variant allows filtering of spurious time oscillation on the pressure field, thus reaching a more stable solution. It has recently been found by the authors (Viccione, 2012b) that during the very early stages of the impact of a free surface liquid flow on a vertical wall, pressures develop spatially like the water hammer case as shown in Figure 1. Moreover, pressure distribution at the wall is non linear initially. As the phenomenon takes place, the free surface starts then to curve near the solid boundary as depicted in Figure 2, featuring the well known “flip through”, described by Lugni (2006) among others. While carrying out this work we also came across the necessity of re-evaluating the classical SPH approach to the compressibility (Viccione, 2012a).

Figure 1 – Fluid-vertical wall interaction: soon after the impact pressure field evolves as the water hammer kind of phenomenon.

Figure 2 – Fluid-vertical wall interaction: after some instants, the free surface curves at the wall.

Contact: gviccion@unisa.it

References

Lugni, C., Brocchini, M., Faltinsen, O.M. (2006), Wave impact loads: The role of the flip-through, Phys. Fluids 18:101.

Molteni, D., Colagrossi A. (2008), A simple procedure to improve the pressure evaluation in hydrodynamic context using the SPH, Comput. Phys. Comm. DOI:10.1016/j.cpc.2008.12.004.

Neves, M.G., Reis, M.T., Losada, I.J., Hu K. (2008), Wave Overtopping of Póvoa de Varzim Breakwater: Physical and Numerical Simulations, J. Waterway, Port, Coastal, Ocean Eng., 134:226.

Peregrine, D.H. (2003), Water wave impact on walls. Ann. Rev. Fluid Mech. 35:23.

Viccione, G., Bovolin, V., Pugliese Carratelli, E. (2012a), Using SPH to compute slamming loads on vertical structures, Proc. 2nd Int. Conf. on Violent Flows pp. 261–266.

Viccione, G., Bovolin, V., Pugliese Carratelli E. (2012b), Simulating fluid-structure interaction with SPH, Proc. Int. Conf. of Numerical Analysis and Applied Mathematics vol. 1479(209), pp. 209–212, DOI: 10.1063/1.4756099.

7

SPHERIC newsletter 15th issue – December 2012

Developing Robust Incompressible SPH schemes in Manchester S. Lind, Manchester Metropolitan University, Manchester, U.K.

A. Skillen, P.K. Stansby, B.D. Rogers, School of Mechanical, Aerospace and Civil Engineering, University of Manchester, Manchester, U.K.

Smoothed Particle Hydrodynamics (SPH) has seen considerable activity in its development, largely because of its potential to simulate highly transient free-surface flows. The fluid is commonly modelled as a weakly compressible fluid using an artificial equation of state to give values for pressures. While this predicts many of the violent free-surface flows well, such as dam break flows, etc., the pressures from weakly compressible formulations are notoriously noisy and the method is dissipative. Some alternative formulations have been proposed in recent years including arbitrary Lagrange Euler formulations and methods based on modifying the equation of mass conservation.

An alternative approach exists to maintain zero velocity divergence, and therefore incompressibility, using the projection method of Chorin (1968). The University of Manchester and the Manchester Metropolitan Uni-versity have been developing Incompressible SPH (ISPH) methods for some time for application to free-surface flows. ISPH can be highly accurate and produce pressure fields that are effectively noise-free. However, instability arises if the particle distributions become highly distorted and produce unphysical behaviour at a free surface.

At Manchester, Xu et al. (2009) suggested a new stabilisation technique by shifting the particles slightly to avoid the instabilities due to particle stretching. This is shown in Figure 1 for the case of Taylor-Green vortices where the particle shifting enables the counter-rotating vortices to decay in good agreement with theory.

Figure 1 – Taylor-Green vortices (Re = 1000) for incompressible SPH (ISPH): without and with particle shifting (Xu et al. 2011).

However, this particle shifting requires modification when applied at the free surface. Lind et al. (2012a) introduced a shifting based on particle concentration gradients similar to Fick’s Law. Figure 2 shows that the new shifting can now reproduce regular propagating waves that do not decay over many wave periods, unlike many weakly compressible SPH schemes. The agree-ment with a high-order semi-analytical solution is very

promising and the pressure field is completely noise free (see Lind et al. 2012a).

Figure 2 – Propagating regular wave: particles coloured according to pressure (Lind et al. 2012a).

The incompressible algorithm is now being developed for multi-phase capability where the gaseous phase is modelled with a compressible SPH formulation (Lind et al. 2012b) as shown for the dam break problem in Figure 3. This allows an incompressible-compressible SPH (ICSPH) scheme to be applied to slam problems taking advantage of the highly accurate and noise-free pressures.

Figure 3 – Multi-phase ICSPH applied to a dam break simulation: (left) water and air particles coloured by pressure, (right) water phase particles only (Lind et al. 2012b) compared with boundary element method (BEM) solution of Colicchio et al. (2003).

With the help of the Science and Technologies Facilities Council (STFC), the code is also being ported to massively parallel architectures using 10,000 cores which will enable simulations to be run for engineering applications.

This work has been funded by the Engineering and Physical Sciences Research Council (EPSRC) under grant EP/H018638/1.

Contact: s.lind@mmu.ac.uk

References

Chorin, A.J. (1968), Numerical solution of the Navier Stokes equations, J. Math. Comput. 22:745–762.

Colicchio, G., Landrini, M., Chaplin, J. (2003), Level-set modelling of the airwater flow generated by a surface

8

piercing body, Proc. 8th Int. Conf. on Numerical Ship Hydrodynamics, Korea, 2003.

Lind, S., Xu, R., Stansby, P.K., Rogers, B.D. (2012a) Incompressible smoothed particle hydrodynamics for free-surface flows: A generalised diffusion-based algorithm for stability and validations for impulsive flows and propagating waves, J. Comput. Phys. 231(4): 1499–1523.

Lind, S., Stansby, P.K., Rogers, B.D. (2012b), A multiphase incompressible-compressible smoothed particle hydrodynamics method, Proc. 7th Int. SPHERIC Workshop, J.J. Monaghan & J. Kajtar editors. pp 324–332. May 2012.

Xu, R., Stansby, P.K., Laurence, D. (2009), Accuracy and stability in incompressible SPH (ISPH) based on the projection method and a new approach, J. Comput. Phys. 228:6703–6725.

Prediction of aircraft structural response during ditching – The SMAES project J. Campbell, School of Engineering, Cranfield University, Cranfield, MK43 0AL. United Kingdom

The focus of the Smart Aircraft in Emergency Situations (SMAES) project is the development of tools to support the design and entry into service of safer aircraft. A set of advanced simulation techniques are being developed to allow more cost effective design methods for ditching and the integration of safety into advanced structural concepts. Of particular interest for future aircraft designs is the effective use of new materials, such as carbon-fibre composites. New design tools are required in order to allow effective use of these materials while maintaining safety. These tools will also support the continuing requirement for designs which provide a greater level of passenger protection.

This paper presents the objectives of SMAES, identifies the key technical challenges and describes the planned demonstration cases. The project covers two main research areas: numerical prediction of the ditching loads, and predictive aircraft models that incorporate non-linear dynamic structural behaviour and are coupled with the hydrodynamic models. The final coupled fluid-structure tools will be demonstrated on metallic, composite and composite-metallic hybrid structures.

Within SMAES, two main approaches to the prediction of ditching loads are being developed. The first approach uses semi-analytical methods, where the modified Wagner method and the Modified Logvinovich Method (MLM) are being extended. The second approach is the use of detailed numerical fluid models that can be coupled with deformable structural models. Under SMAES both meshless (SPH) and Coupled Euler-Lagrange approaches are being used. An important challenge for the prediction of ditching loads is the inclusion of effects resulting from high velocity water flow on the airframe loads. These effects include cavitation, aeration and suction.

Reliable and predictive aircraft models for structural behaviour and rupture under dynamic fluid loads are required. A particular challenge is the need to extract detailed structural response, such as rupture, as well as overall aircraft dynamics from the modelling process. An additional requirement for the structural models is that they must be suitable for coupling with the fluid load models. These activities will be supported by an intensive experimental campaign. A new experimental facility is being commissioned at INSEAN to allow

guided impact tests of structural components at 30–50 m/s into water. These experiments are designed to be representative of the conditions experienced during ditching, and avoid the scaling problems inherent in model tests. In addition the experimental campaign includes a number of impact tests in support of the demonstration cases along with characterisation of specific materials used in the experiments.

The SMAES project began in February 2011, funded under the EU Commission’s Seventh Research Frame-work Programme (grant agreement n° FP7–266172).

Project Consortium

Dassault Aviation INSEAN

Airbus Military CIRA

Alenia Aeronautica Cranfield University

(coordinator)

Airbus Operations University of East Anglia

ESI University of Patras

Altair Engineering Technische Universität

Hamburg-Harburg

DLR Technische Universität Dresden

ONERA

Figure 1 – Coupled finite element – SPH simulation of NACA TN-2929 model ditching experiment. The wing and empennage geometry are not included.

Contact: j.campbell@cranfield.ac.uk