Dörte C. Sternel

Influence of Time Step Size and Convergence Criteria on Large Eddy Simulations with Implicit Time Discretization

Subject of this work is the influence of numerical parameters on quality and efficiency of Large Eddy Simulations. Variations of the time step size and the convergence criterion are considered. The influence of these parameters on mean values and computational time are presented and discussed. The computations were carried out for the well known test case “Periodic flow over a 2D hill”.

Efficient Numerical Simulation and Optimization of Fluid-Structure Interaction

The paper concerns the efficient numerical simulation and optimization of fluid-structure interaction (FSI) problems. The basis is an implicit partitioned solution approach involving the finite-volume flow solver FASTEST, the finite-element structural solver FEAP, and the coupling interface MpCCI. Special emphasis is given to the grid moving techniques for which algebraic and elliptic approaches are considered. The possibilities for accelerating the computations by the usage of multigrid methods, adaptive underrelaxation, and displacement prediction are discussed. A concept for integrating the FSI solver into an optimization procedure for FSI problems is presented. Numerical results are given to illustrate the capabilities of the approaches considered.

Parallel Code Analysis in HPC User Support.

Over the last decades, the access to high performance computing resources has transitioned from being restricted to only a few specialists to becoming a commodity for people from widely varying contexts. Consequently, this includes the need for a level of in-depth user support which cannot be provided by a sysadmin alone. The Competence Center for High Performance Computing in Hesse aims to establish this service for the Hessian universities. As such, merely troubleshooting user scripts and codes is not enough; instead, the goal is to optimize the usage of computational resources, which needs to include the analysis of user codes and ensuring good performance and scalability. For this task, parallel code analysis tools in user support are becoming more and more commonplace and may indeed be regarded as essential. As a result, HPC user support specialists form a separate user group from application developers and require different sets of features.

High-Performance Computing Techniques for Coupled Fluid, Structure and Acoustics Simulations

A framework for fully coupled numerical simulation of fluid flow, structural deformation, and acoustics is presented. The methodology involves an implicit partitioned approach for fluid-structure interaction, a viscous-acoustic splitting technique for flow acoustics, and corresponding coupling schemes. All components are designed for the use on parallel high-performance computers. Special emphasis is given to the use of geometric multi-grid techniques in order to increase the efficiency of the numerical simulations. Results for several test cases illustrate the capabilities of the approaches considered.

Investigation of the Flow around a Cylinder Plate Configuration with Respect to Aerodynamic Noise Generation Mechanisms

In this work we present a Large Eddy Simulation of an aero-acoustic test case consisting of a plate located in the turbulent wake of a circular cylinder. This configuration is very attractive for the validation of low Mach number aero-acoustic codes and coupling techniques, since a high sound pressure level is present at a very low Mach number and also because its simple geometry. Further it seems to be an interesting test case for future works if besides aero-acoustics also fluid induced vibrations are of interest.

Partitioned Fluid–Structure–Acoustics Interaction on Distributed Data: Numerical Results and Visualization

We present a coupled simulation approach for fluid–structure–acoustic interactions (FSAI) as an example for strongly surface coupled multi-physics problems. In addition to the multi-physics character, FSAI feature multi-scale properties as a further challenge. In our partitioned approach, the problem is split into spatially separated subdomains interacting via coupling surfaces. Within each subdomain, scalable, single-physics solvers are used to solve the respective equation systems. The surface coupling between them is realized with the scalable open-source coupling tool preCICE described in the “Partitioned Fluid–Structure–Acoustics Interaction on Distributed Data: Coupling via preCICE”. We show how this approach enables the use of existing solvers and present the overall scaling behavior for a three-dimensional test case with a bending tower generating acoustic waves. We run this simulation with different solvers demonstrating the performance of various solvers and the flexibility of the partitioned approach with the coupling tool preCICE. An efficient and scalable in-situ visualization reducing the amount of data in place at the simulation processors before sending them over the network or to a file system completes the simulation environment.

Parallel Algorithm for Solution-Adaptive Grid Movement in the Context of Fluid Structure Interaction

We present a new grid movement strategy, tested for a generic fluid-structure interaction (FSI) test case. The test case describes a flat plate with a prescribed rotational movement in a turbulent channel flow. The transient turbulent flow field is calculated with a low-Re RANS model. To account for the deforming fluid domain two different grid movement methods are compared. Using transfinite interpolation with a grid-point distribution fitted to the stationary starting conditions as grid moving method leads to errors for the drag-coefficient. By employing a normalized wall distance adaptive method it is possible to fulfill the near-wall resolution requirements within every time step and, to thereby, get more accurate results. The parallelization is achieved by domain decomposition and is evaluated using a strong scaling experiment.

A coupling algorithm with second-order accuracy for thermal fluid-structure interaction

A second-order accurate coupling algorithm for Thermal Fluid-Structure Interaction (TFSI) is presented. An implicit partitioned solution approach, which combines the advantages of weakly and strongly coupled schemes, is used. The framework consists of a finite-volume solver in Arbitrary-Lagrangian-Eulerian formulation for the flow simulation, a finite-element code for the structural part and a commercial, black-box coupling interface. The thermal coupling is realised by a Schwarz decomposition method. Several elementary configurations are considered, to test the accuracy and to investigate the numerical characteristics of the system. By applying different grid configurations, we show that the results are of second-order accuracy. We also discuss the convergence properties of the method.

Integral Model for Simulating Gas Turbine Combustion Chambers

Relying on the software platform FASTEST different sub-models of various complexities designed in the framework of the Collaborative Research Centre 568 were combined to form integral models. As review paper, this contribution summarizes the developments of a Large Eddy Simulation (LES) based integral model for reliably simulating gas turbine combustion chambers. The model was analyzed, validated and evaluated in terms of fuel flexibility, flow, mixing and combustion modeling and predictability. The parallel efficiency for large simulations with the integral model could be improved by using a load balancing strategy based on graph theory.

Discretization of Mass Forces and its Impact on Drop Break-up in Stirrers

The presence of two-phase flow and complex geometry is the main difficulty during the computation of a flow field in a stirrer. The interface capturing Volume of Fluid method, implemented in the framework of a finite volume Navier-Stokes solver, is used to determine the bubble break-up location and the bubble diameter distribution after the collision with the stirrer plate. Comparisons of numerical and experimental results are given. Moreover, the impact of different discretization schemes of mass forces on non-orthogonal grids (e.g. gravity or surface tension) on simulation results are studied.