Search

In this article, a novel multiscale modeling method is proposed for transient computational fluid dynamics (CFD) simulations of the human airways. The developed method is the first attempt to incorporate spatial coupling and temporal coupling into transient human airway simulations, aiming to improve the flexibility and the efficiency of these simulations. In this method, domain decomposition was used to separate the complex airway model into different scaled domains. Each scaled domain could adopt a suitable mesh and timestep, as necessary: the coarse mesh and large timestep were employed in the macro regions to reduce the computational cost, while the fine mesh and small timestep were used in micro regions to maintain the simulation accuracy. The radial point interpolation method was used to couple data between the coarse mesh and the fine mesh. The continuous micro solution–intermittent temporal coupling method was applied to bridge different timesteps. The developed method was benchmarked using a well-studied four-generation symmetric airway model under realistic normal breath conditions. The accuracy and efficiency of the method were verified separately in the inhalation phase and the exhalation phase. Similar airflow behavior to previous studies was observed from the multiscale airway model. The developed multiscale method has the potential to improve the flexibility and efficiency of transient human airway simulations without sacrificing accuracy.

A computational model has been developed which facilitates the simulation of particle suspensions in non-Newtonian fluids. The model employs the discrete element method (DEM) to represent a range of particle geometries. The fluid phase is captured using a non-Newtonian formulation of the lattice Boltzmann method (LBM) and full hydrodynamic coupling of the LBM and DEM is achieved using an immersed moving boundary condition 1, 2. The developed model has the ability to simulate Navier-Stokes hydrodynamics, turbulence, a range of rheological models (e.g. power law) as well as varying fluid viscosities and densities. This paper outlines the extension and application of the non-Newtonian LBM-DEM model to simulate numerical rheometry experiments 3 and channel-flow analyses 4 using particle suspensions in power law fluids. The numerical rheometry procedure employs a three-dimensional periodic shear cell geometry, which is an extension of previous work that studied the behaviour of granular media in a cylindrical Couette rheometer 5. Currently, the periodic shear cell is utilised in a displacement-controlled arrangement. Future work will investigate the feasibility of a stress-controlled arrangement which would facilitate the simulation of drained rheometry experiments under varying consolidation. This will require the development of a porous immersed moving boundary which can apply both normal and shear stresses to the cell geometry without affecting the fluid pressure. The channel-flow analyses utilise a three-dimensional periodic aperture geometry to study the development length of steady-state, pressure driven particle suspensions. By directly capturing the physical phenomena which dominate the interaction of particles suspended in confined geometries, the developed LBM-DEM model can be employed to investigate a range of physical phenomena in the oil, gas and mining industries. Examples include sanding in oil reservoirs, proppant transport in hydraulic fractures, and fin

Presented at the SCS Spring Simulation Multi-Conference – SpringSim 2011, April 4-7, 2011 – Boston, USA Awarded Best Paper in the 19th High Performance Computing Symposium and Best Overall Paper at SpringSim 2011., This paper presents a numerical framework that enables scalable, parallel execution of engineering simulations on multi-core, shared memory architectures. Distribution of the simulations is done by selective hash-tabling of the model domain which spatially decomposes it into a number of orthogonal computational tasks. These tasks, the size of which is critical to optimal cache blocking and consequently performance, are then distributed for execution to multiple threads using the previously presented task management algorithm, H-Dispatch. Two numerical methods, smoothed particle hydrodynamics (SPH) and the lattice Boltzmann method (LBM), are discussed in the present work, although the framework is general enough to be used with any explicit time integration scheme. The implementation of both SPH and the LBM within the parallel framework is outlined, and the performance of each is presented in terms of speed-up and efficiency. On the 24-core server used in this research, near linear scalability was achieved for both numerical methods with utilization efficiencies up to 95%. To close, the framework is employed to simulate fluid flow in a porous rock specimen, which is of broad geophysical significance, particularly in enhanced oil recovery.

In this paper, a programming model is presented which enables scalable parallel performance on multi-core shared memory architectures. The model has been developed for application to a wide range of numerical simulation problems. Such problems involve time stepping or iteration algorithms where synchronization of multiple threads of execution is required. It is shown that traditional approaches to parallelism including message passing and scatter-gather can be improved upon in terms of speed-up and memory management. Using spatial decomposition to create orthogonal computational tasks, a new task management algorithm called H-Dispatch is developed. This algorithm makes efficient use of memory resources by limiting the need for garbage collection and takes optimal advantage of multiple cores by employing a “hungry” pull strategy. The technique is demonstrated on a simple finite difference solver and results are compared to traditional MPI and scatter-gather approaches. The H-Dispatch approach achieves near linear speed-up with results for efficiency of 85% on a 24-core machine. It is noted that the H-Dispatch algorithm is quite general and can be applied to a wide class of computational tasks on heterogeneous architectures involving multi-core and GPGPU hardware., Schlumberger-Doll Research Center, Saudi Aramco

This paper discusses the implementation of particle based numerical methods on multi-core machines. In contrast to cluster computing, where memory is distributed across machines, multi-core machine can share memory across all cores. Here general strategies are developed for spatial management of particles and sub-domains that optimize computation on shared memory machines. In particular, we extend cell hashing so that cells bundle particles into orthogonal tasks that can be safely distributed across cores avoiding the use of "memory locks" while still protecting against race conditions. Adjusting task size provides for optimal load balancing and maximizing cache hits. Additionally, the way in which tasks are mapped to execution threads has a significant influence on the memory footprint and it is shown that minimizing memory usage is one of the most important factors in achieving execution speed and performance on multi-core. A novel algorithm called H-Dispatch is used to pipeline tasks to processing cores. The performance is demonstrated in speed-up and efficiency tests on a smooth particle hydrodynamics (SPH) flow simulator. An efficiency of over 90% is achieved on a 24-core machine.

In this paper, a 3D Smooth Particle Hydrodynamics (SPH) simulator for modeling grain scale fluid flow in porous rock is presented. The versatility of the SPH method has driven its use in increasingly complex areas of flow analysis, including flows related to porous rock for both groundwater and petroleum reservoir research. Previous approaches to such problems using SPH have involved the use of idealized pore geometries (cylinder/sphere packs etc.). In this paper we discuss the characterization of flow in models with geometries acquired from 3D X-ray microtomograph images of actual porous rock. One key advantage of SPH is realized when considering the complexity of multiple fluid phases (e.g., water and oil). By incorporating interfacial physics such as surface tension and wettability, it is possible to model the capillary behavior of multiple fluid phases with accuracy. Simulation results for permeability will be presented and compared to those from experimentation and other numerical methods showing good agreement and validating the method. By accurately reproducing the flow characteristics of actual porous rock samples, this work has made significant progress towards validating SPH for such applications., Saudi Aramco, Schlumberger-Doll Research Center