Report on a research project supported by a DAAD scholarship

Fluid-structure interaction involving free surface flows

Jessica Marcela Mariño Salguero

Institute of Numerical Methods in Mechanical Engineering (FNB) & Graduate School of

Computational Engineering (GSC CE). Technische Universität Darmstadt, Dolivostraße 15,
64293 Darmstadt, Germany
marino@fnb.tu-darmstadt.de, marino@gsc.tu-darmstadt.de, http:/www.fnb.tu-darmstadt.de

1. INTRODUCTION

In most EU Member States, the production of ships and floating structures is the sector
with the highest added value in maritime manufacturing. However, out of 22 756 companies
in the sector in 2012, 14 503 were repair companies [1]. This shows how much marine
structures are negatively affected by the hydraulic phenomenon of water waves, impacts, and
splashing jets. The three phenomena are known as free surface flows or two-phase flow water-
air problems. The interaction between ships and waves can lead to relatively large deformation
and cause high-frequency vibration which damages the structure [2]. Consequently,
understanding the fluid-structure interaction involving free surface flows is a great engineering
challenge to develop more efficient and sufficiently economically effective naval architectures
and marine infrastructures.
Fluid-structure interaction (FSI) involving free surface flows is a complex multiphysics
phenomenon for which analytical solutions are mostly not available and laboratory experiments
are difficult and expensive to conduct. Thus, the numerical simulation of these problems can
be a fast and efficient way to get detailed information about the involved phenomena, which
may serve as the basis for optimization of marine structure design. However, the simulation of
the FSI involving free surface flows is also an engineering challenge since the free surface
motion can be smooth or violent and undergo topological changes. Furthermore, the boundary
conditions on moving boundaries are nonlinear and their position is part of the solution of the
problem.

2. RESEARCH DESCRIPTION

2.1. Overview and the main goal

I work with an implicit partitioned fluid-structure interaction (FSI) approach proposed by
Schäfer and Teschauer [3]. In a partitioned approach, the solvers act as black-boxes with a
given input-output relation at the common boundary. Regularly, the fluid solver receives
velocities as an input and returns forces as an output. The structure solver takes forces as an
input and returns displacements [4]. In this case, the fluid part is solved with the in-house solver
FASTEST, employing a finite volume method on hexahedral block-structured grids. The
structural part is solved with the program FEAP [5] based on the finite element method. The
two solvers are coupled in the Multiphysics platform PRECICE [6]. This approach has been
tested and optimized to obtain efficient simulations of fluid-structure interaction (FSI)
problems with one fluid [7], [8] but has not yet been coupled and tested with multiphase flows.

In the case of a multiphase system, the Volume of Fluid method (VOF) developed by Hirt
and Nichols [9] is used due to the guaranteed mass conservation and the possibility of
simulating complex free surface flows. The VOF method introduces an additional transport
equation for the volume fraction α that captures the position of the interface between the two
fluids. The standard VOF implementation in the program FASTEST uses the High-Resolution
(HR) scheme M-CICSAM [10] to advect the volume fraction in space. The HR scheme is
incorporated in the discretized equation through the Downwind Weighting Factors (DWF)
method of Leonard and Mokhtari [11]. The drawback of this implementation is that it is
strongly dependent on the Courant number (Co) [12]. The M-CICSAM scheme keeps the
sharpness of the interface and is stable for Co less than 0.5, using very small time-steps.

The simulation of the two viscous, incompressible, and immiscible fluids in a moving
domain is based on the one-fluid formulation of the Navier-Stokes equations and the VOF
equation defined in the Arbitrary Lagrangian Eulerian description (ALE), where the convective
term of the equations is multiplied by the relative velocity that is the difference between the
fluid velocity and the grid velocity. The grid velocity is produced by the structural deformation
and is calculated using the Space Conservation Law.

Unfortunately, the solution of a multiphase system with moving grids using the standard
Volume of Fluid solvers has two major drawbacks: lack of stability and large computational
costs. The frequent changes of the control volume size inside of the FSI cycle induce often Co
larger than 1 which deteriorates the stability of the HR schemes used in the discrete VOF
equation [13]. Also, the decrease of the control volume affects the DWF method that produces
unphysical high velocities and a concentration of pressure at the interface [14]. Moreover, the
correct evaluation of the discontinuous body forces at the interface between fluids is also
essential to achieve consistent FSI coupling.

In order to overcome the above-described disadvantage, the major objective of this Ph.D.
project is the development, investigation, and application of new computational methods that
allow a reliable simulation of fluid-structure interactions involving free-surface flows with high
numerical efficiency.

2.2. Research aims

The research aims of the work are:
 Finding a solution for the lack of stability of the VOF equation using HR-schemes and
improving the techniques to evaluate the body forces to avoid the decoupling between
velocity and pressure in the predictor-corrector SIMPLE algorithm.
 Searching for options or developing a novel HR-scheme and techniques to implement
it in the discretization equation in order to work with moving grids.
 Performing systematic investigations of the efficiency and the accuracy of the
corresponding numerical techniques for selected test cases. This will provide insight
into the capabilities of these methods for dealing with complex real-world applications.
  Comparing the results with the literature to demonstrate the capabilities of the
developments.

3. PROGRESS REPORT

3.1 Main Achievements

Currently, the FSI approach has been modified to work with multiphase flows which
implied an extensive literature review of the state of the art of numerical methods employed in
FSI and multiphase flow problems.

Furthermore, I modified and implemented some new tools to solve efficiently multiphase
flows. For example:
 An adaptative time-step to avoid high Courant numbers.
 Three time-discretization schemes (BDF2, Implicit Euler, Crank-Nicolson) to work
together with the VOF and momentum equation.
 The implementation of three HR-schemes: High-Resolution Interface Capturing
(HRIC) introduced in [15], Modified HRIC developed by [16], and STACS described
in [17] which are often used by the multiphase community due to their lower
dependence on Co.
 Reimplementation of the boundary conditions, and DWF method for calculating the
VOF equation.
 The inclusion of the density variations in the time discretization.

Nevertheless, there are still many challenges to developing a consisted coupling between
the Volume of Fluid method and the FSI approach. Mainly because the solving algorithm
shows some limitations in getting stable and accurate results when the difference of the
physical properties between the fluids is very large and the body forces are considered, as well
as, the loss of stability of the HR-scheme if there is a large deformation of the fluid domain.

3.2. Future Developments

The following developments are ongoing:
 Implementing the harmonic interpolation [18] for the face density and viscosity to work
with fluids with a large difference of physical properties.
 Implementing and comparing the Deferred Correction method and the Revised
Normalized Weighting Factors method [19] to change the current used DWF method.
 Implementing a new pressure correction equation and modify the pressure gradient
calculation based on the work of Efremov et al. [20].
 Developing a novel and consistent HR-scheme to work with moving grids.
 Validating the implementation using benchmark test cases.
 Implementing and validating the Immersed Boundary method only for simulating free
surface flows in contact with rigid bodies.

4. TRAINING PROGRAMME

As part of the Ph.D. training to develop a wide range of skills, I have attended the following
regular events:
  Research Colloquium of the Graduate School CE every Monday.
 Monthly FASTEST meeting to speak about the advances and problems in the code
development.

In the year 2017

 4th International Conference on Computational Engineering (ICCE 2017), the event
took place in Darmstadt on 28th to 29th of September 2017.
 7th Graduate School of Excellence Computational Engineering annual retreat in
Seeheim on March 21st and 22nd of 2017.
 The annual meeting of the FNB group on 12th to 17th of March 2017 in Hirschegg.
 Totalview Debugging Workshop at TU Darmstadt on 28th of June 2017.
 Seminar: Scientific Writing in English at TU Darmstadt on 13th and 14th of July 2017.
 Seminar: Conference Presentation at TU-Darmstadt on 8th and 9th of November 2017.
 German course: level A2 (41-11-0042 Paternoster) at TU-Darmstadt summer semester
2017.

In the year 2018

 Seminar: The Poster and the Pitch at TU-Darmstadt on 10th of January 2018.
 The annual meeting of the FNB group on 12th to 16th of March 2018 in Hirschegg.
 Seminar: Effektiv in Teams zusammenarbeiten – Diversity verstehen und konstruktiv
nutzen at TU-Darmstadt on 27th of March 2018.
 Seminar: Getting the Ph.D. done! Time management and Self-management for
Doctoral Candidates at TU-Darmstadt on 22th and 23th of March 2018.
 7th Graduate School of Excellence Computational Engineering annual retreat in
Seeheim on 9th and 10th of April 2018.
 Workshop: Fortran Modernization on 4th and 5th of June 2018
 German course: level B1 (41-11-0062 Bridging Course Turbo) at TU-Darmstadt
winter semester 2018.
 German course: level B2 (41-98-1120 German UNIcert II Certificate) at TU-
Darmstadt summer semester 2018.

In the year 2019

 Seminar: Team communication in interdisciplinary and intercultural contexts at TU-
Darmstadt on 14th and 15th of March 2019.
 The annual meeting of the FNB group on 18th to 22th of March 2019 in Hirschegg.
 German course: level C1 (41-11-0352 Vocabulary Training in German, 41-11-0382
Listening - Comprehension - Discussing I, 41-11-0472 Writing and Discussing II) at
TU-Darmstadt winter semester 2019.

In the next months, I have planned to attend the following events.

 7th Graduate School of Excellence Computational Engineering annual retreat in
Seeheim on 9th and 10th of April 2019
 10th International Conference on Multiphase Flow (ICMF 2019) in Rio de Janeiro on
18th to 24th of May 2019.
 German course: level C1 (41-11-1411-ku German for Technical Professions I, 41-11-
1322-ku German Grammar III) at TU-Darmstadt summer semester 2019.
4. PUBLICATION
The next paper was sent to the ICMF 2019 conference:
J. Mariño and M. Schäfer, “Investigation of discretization methods for simulating multiphase flows
with moving grids,” 2019, pp. 1–10.

5. CONCLUSIONS
This document described a short summary of the activities during the first and second period
in my doctoral studies, which essentially has been an in-depth literature review of the topic and
modifications of the standard multiphase algorithm to make the coupling with an FSI approach
possible. Although the coupling has been developed, much more research is necessary to
improve the efficiency and the accuracy of the proposed method.

5. REFERENCES.
[1] Eurostat Statistics Explained, “Maritime economy statistics - coastal regions and sectoral
perspective.” p. ISSN 2443--8219, 2015.
[2] Z. Hu, W. Tang, H. Xue, X. Zhang, and K. Wang, “Numerical study of rogue wave
overtopping with a fully-coupled fluid-structure interaction model,” Ocean Eng., vol. 137, pp.
48–58, Jun. 2017.
[3] M. Schäfer and I. Teschauer, “Numerical simulation of coupled fluid–solid problems,”
Comput. Methods Appl. Mech. Eng., vol. 190, no. 28, pp. 3645–3667, Mar. 2001.
[4] M. Mehl, B. Uekermann, H. Bijl, D. Blom, B. Gatzhammer, and A. van Zuijlen, “Parallel
coupling numerics for partitioned fluid–structure interaction simulations,” Comput. Math. with
Appl., vol. 71, no. 4, pp. 869–891, Feb. 2016.
[5] R. L. Taylor, “FEAP-A finite element analysis program.” 2000.
[6] H.-J. Bungartz et al., “preCICE – A fully parallel library for multi-physics surface coupling,”
Comput. Fluids, vol. 141, pp. 250–258, Dec. 2016.
[7] S. Yigit, M. Heck, M. Schäfer, and D. Sternel, “An enhanced implicit partitioned ALE
approach for fluid-structure interaction simulations,” 2007.
[8] M. Schäfer, D. C. Sternel, G. Becker, and P. Pironkov, “Efficient Numerical Simulation and
Optimization of Fluid-Structure Interaction,” Fluid Struct. Interact. II, pp. 131–158, 2011.
[9] C. . Hirt and B. . Nichols, “Volume of fluid (VOF) method for the dynamics of free
boundaries,” J. Comput. Phys., vol. 39, no. 1, pp. 201–225, Jan. 1981.
[10] T. Wacławczyk, Ö. Caner Gemici, and M. Schäfer, “Novel high-resolution scheme for
interface capturing in multi-phase flow,” 2007.
[11] B. P. Leonard and S. Mokhtari, “Beyond first-order upwinding: The ultra-sharp alternative for
non-oscillatory steady-state simulation of convection,” Int. J. Numer. Methods Eng., vol. 30,
no. 4, pp. 729–766, Sep. 1990.
[12] M. Hoekstra, G. Vaz, B. Abeil, and T. Bunnik, “Free-surface flow modelling with interface
capturing techniques,” MARINE2007, vol. 2, pp. 1–4, 2007.
[13] J. Meyer, H. Renzsch, K. Graf, and T. Slawing, “Advanced CDF-Simulations of free-surface
flows around modern sailing yachts using a newly developed OpenFOAM solver,” 2016.
[14] M. S. Darwish and F. Moukalled, “The Normalized Weighting Factor Method: A Novel
Technique for accelerating the convergence of high-resolution convective schemes.,” Numer.
Heat Transf. Part B Fundam., vol. 30, no. 2, pp. 217–237, Sep. 1996.
[15] S. Muzaferija, “A two-fluid Navier-Stokes solver to simulate water entry,” in Proceedings of
22nd symposium on naval architecture, 1999, 1999, pp. 638–651.
[16] I. R. Park, K. S. Kim, J. Kim, and S. H. Van, “A volume-of-fluid method for incompressible
free surface flows,” Int. J. Numer. Methods Fluids, vol. 61, no. 12, pp. 1331–1362, Dec. 2009.
[17] M. Darwish and F. Moukalled, “Convective Schemes for Capturing Interfaces of Free-Surface
Flows on Unstructured Grids,” Numer. Heat Transf. Part B Fundam., vol. 49, no. 1, pp. 19–42,
Jan. 2006.
[18] D. B. Kothe, “Perspective on Eulerian Finite Volume Methods for Incompressible Interfacial
Flows,” in Free Surface Flows, vol. M, Vienna: Springer Vienna, 1998, pp. 267–331.
[19] T. Chourushi, “Computationally inexpensive and revised normalized weighting factor method
for segregated solvers,” Int. J. Comput. Math., vol. 95, no. 8, pp. 1622–1653, Aug. 2018.
[20] V. R. Efremov et al., “Method for taking into account gravity in free-surface flow simulation,”
Comput. Math. Math. Phys., vol. 57, no. 10, pp. 1720–1733, Oct. 2017.