Vasco Varduhn

Multi-resolution Models: Recent Progress in Coupling 3D Geometry to Environmental Numerical Simulation

In this paper, we present recent results on coupling geographic, infrastructure, and building models to multi-resolution numerical simulations. In order to achieve this, a parallel data access framework with interfaces to all parts of the simulation pipeline such as pre-processing, numerical simulation, and post-processing has been developed. The applicability of the approach presented in this work is shown by simulating urban flooding including surface flow of a city, the pipe network interaction, and its consequences to individual buildings. While the real life city model including the drainage system has been provided by the authorities of the city of Munich and comprises an area of about 2 by 2 km with detailed topography and a complete set of approximately 3,000 buildings modelled on LOD 1 of CityGML, IFC-models are the initial starting points for generating octrees on the level of individual buildings. In order to investigate the effects of the drainage system collapsing due to a heavy rain scenario, a fully three-dimensional parallel free surface flow simulation is incorporated with the interaction of the one-dimensional pipe-network flow. The whole simulation is performed on three levels of resolution, each of which is discretised by a fast voxelisation algorithm to generate a computational grid for the CFD simulation.

A Parallel High-Order Fictitious Domain Approach for Biomechanical Applications

The focus of this contribution is on the parallelization of the Finite Cell Method (FCM) applied for biomechanical simulations of human femur bones. The FCM is a high-order fictitious domain method that combines the simplicity of Cartesian grids with the beneficial properties of hierarchical approximation bases of higher order for an increased accuracy and reliablility of the simulation model. A pre-computation scheme for the numerically expensive parts of the finite cell model is presented that shifts a significant part of the analysis update to a setup phase of the simulation, thus increasing the update rate of linear analyses with time-varying geometry properties to a range that even allows user interactive simulations of high quality. Paralellization of both parts, the pre-computation of the model stiffness and the update phase of the simulation is simplified due to a simple and undeformed cell structure of the computation domain. A shared memory parallelized implementation of the method is presented and its performance is tested for a biomedical application of clinical relevance to demonstrate the applicability of the presented method.

A sliding window technique for interactive high-performance computing scenarios

Interactive high-performance computing is doubtlessly beneficial for many computational science and engineering applications whenever simulation results should be visually processed in real time, i.e. during the computation process. Nevertheless, interactive HPC entails a lot of new challenges that have to be solved - one of them addressing the fast and efficient data transfer between a simulation back end and visualisation front end, as several gigabytes of data per second are nothing unusual for a simulation running on some (hundred) thousand cores. Here, a new approach based on a sliding window technique is introduced that copes with any bandwidth limitations and allows users to study both large and small scale effects of the simulation results in an interactive fashion.

A Framework for Parallel Numerical Simulations on Multi-Scale Geometries

In this paper, an approach on performing numerical multi-scale simulations on fine detailed geometries is presented. In particular, the focus lies on the generation of sufficient fine mesh representations, whereas a resolution of dozens of millions of voxels is inevitable in order to sufficiently represent the geometry. Furthermore, the propagation of boundary conditions is investigated by using simulation results on the coarser simulation scale as input boundary conditions on the next finer scale. Finally, the applicability of our approach is shown on a two-phase simulation for flooding scenarios in urban structures running from a city wide scale to a fine detailed in-door scale on feature rich building geometries.

An Immersogeometric Method for the Simulation of Turbulent Flow Around Complex Geometries

In this chapter we summarize a recently proposed immersogeometric method for the simulation of incompressible flow around geometrically complex objects. The method immerses the objects into unfitted tetrahedral finite elements meshes and weakly enforces Dirichlet boundary conditions on the surfaces of the objects. Adaptively refined quadrature rules are used to faithfully capture the flow domain geometry in the discrete problem without modifying the unfitted finite element mesh. A variational multiscale formulation which provides accuracy and robustness in both laminar and turbulent flow conditions is employed. We assess the accuracy of the method by analyzing the flow around an immersed sphere for a wide range of Reynolds numbers. We show that flow quantities of interest are in very good agreement with reference values obtained from standard boundary-fitted approaches. Our results also show that the faithful representation of the geometry in intersected elements is critical for accurate flow analysis. We demonstrate the potential of our proposed method for high-fidelity industrial scale simulations by performing an aerodynamic analysis of a full-scale agricultural tractor.

A Framework for the Interactive Handling of High-Dimensional Simulation Data in Complex Geometries

Flow simulations around building infrastructure models involve large scale complex geometries, which when discretized in adequate detail entail high computational cost. Moreover, tasks such as simulation insight by steering or optimization require many such costly simulations. In this paper, we illustrate the whole pipeline of an integrated solution for interactive computational steering, developed for complex flow simulation scenarios that depend on a moderate number of both geometric and physical parameters. A mesh generator takes building information model input data and outputs a valid cartesian discretization. A sparse-grids-based surrogate model—a less costly substitute for the parameterized simulation—uses precomputed data to deliver approximated simulation results at interactive rates. Furthermore, a distributed multi-display visualization environment shows building infrastructure together with flow data. The focus is set on scalability and intuitive user interaction.