Zonal Flow Solver (ZFS): a highly efficient multi-physics simulation framework

Lintermann, Andreas; Meinke, Matthias; Schröder, Wolfgang

doi:10.1080/10618562.2020.1742328

Cited by 25 publications

(9 citation statements)

References 88 publications

(114 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The computations can be performed efficiently in parallel, it is straightforward to parallelize the code, and boundary conditions can be easily applied in contrast to, e.g., cut-cell methods. Furthermore, there is no need to solve a pressure Poisson equation for quasi-incompressible flow as the pressure is an explicit result of the lattice-BGK algorithm [13].…”

Section: Step (V) Of the Automated Pipeline: Simulation Methodsmentioning

confidence: 99%

“…Previous studies underline that fluid mechanical properties play a key role in analyzing nasal respiration. Lintermann et al [12] use a lattice-Boltzmann (LB) method integrated into the simulation framework multi-physics Aerodynamisches Institut Aachen (m-AIA) (former name: Zonal Flow Solver (ZFS)) [13] to simulate the flow in the nasal cavity. They analyze the flow to classify nasal cavities into ability groups and thereby support a priori surgical intervention decision processes.…”

Section: Numerical Analysis Of Respiratory Flowsmentioning

confidence: 99%

“…The generation of the computational mesh employs the massively parallel grid generator of m-AIA [13]. In parallel, it reads the 3D airway model and stores its triangle in an alternating digital tree (ADT) [42] for accelerated triangle searches and retrieval.…”

Section: Step (Iv) Of the Automated Pipeline: Mesh Generationmentioning

confidence: 99%

See 2 more Smart Citations

A machine-learning-based method for automatizing lattice-Boltzmann simulations of respiratory flows

et al. 2022

Self Cite

View full text Add to dashboard Cite

Many simulation workflows require to prepare the data for the simulation manually. This is time consuming and leads to a massive bottleneck when a large number of numerical simulations is requested. This bottleneck can be overcome by an automated data processing pipeline. Such a novel pipeline is developed for a medical use case from rhinology, where computer tomography recordings are used as input and flow simulation data define the results. Convolutional neural networks are applied to segment the upper airways and to detect and prepare the in- and outflow regions for accurate boundary condition prescription in the simulation. The automated process is tested on three cases which have not been used to train the networks. The accuracy of the pipeline is evaluated by comparing the network-generated output surfaces to those obtained from a semi-automated procedure performed by a medical professional. Except for minor deviations at interfaces between ethmoidal sinuses, the network-generated surface is sufficiently accurate. To further analyze the accuracy of the automated pipeline, flow simulations are conducted with a thermal lattice-Boltzmann method for both cases on a high-performace computing system. The comparison of the results of the respiratory flow simulations yield averaged errors of less than 1% for the pressure loss between the in- and outlets, and for the outlet temperature. Thus, the pipeline is shown to work accurately and the geometrical deviations at the ethmoidal sinuses to be negligible.

show abstract

Section: Step (V) Of the Automated Pipeline: Simulation Methodsmentioning

confidence: 99%

Section: Numerical Analysis Of Respiratory Flowsmentioning

confidence: 99%

See 1 more Smart Citation

A machine-learning-based method for automatizing lattice-Boltzmann simulations of respiratory flows

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…A more general description of the solver can be found in (Lintermann et al, 2020), where more details of the solver structure are described along with performance measures.…”

Section: Methodsmentioning

confidence: 99%

Large-Eddy Simulations of Rim Seal Flow in a One-Stage Axial Turbine

Hösgen

Meinke

Schröder

2020

J. Glob. Power Propuls. Soc.

Self Cite

View full text Add to dashboard Cite

The flow field in a one-stage axial flow turbine with 30 stator and 62 rotor blades including the wheel space is investigated by large-eddy simulation (LES). The Navier-Stokes equations are solved using a massively parallel finite-volume solver based on a Cartesian mesh with immersed boundaries. The strict conservation of mass, momentum, and energy is ensured by an efficient cut-cell/level-set ansatz, where a separate level-set solver describes the motion of the rotor. Both solvers use individual subsets of a shared Cartesian mesh, which they can adapt independently. The focus of the analysis is on the flow field inside the rotor stator cavity between the stator and rotor disks. Two cooling gas mass flow rates are investigated for the same rim seal geometry. First, the time averaged flow field for both simulations is compared, followed by a detailed investigation of the unsteady flow field. The results for the cooling effectiveness are compared to experimental data. Both cases show good agreement with experimental data. It is shown that for the lower cooling gas mass flux several of the wheel space’s acoustic waves are excited. This is not observed for the higher cooling gas mass flux. The excited waves lead to stable, i.e., bounded, fluctuations inside the wheel space and result in a significantly higher hot gas ingestion.

show abstract

“…The pressure can be computed using the ideal gas law by p = c 2 s ρ = (1/3)ρ. The LB method has been chosen for several reasons [18]: (i) the computations can be performed efficiently in parallel, (ii) it is straightforward to parallelize the code, (iii) boundary conditions can easily be applied in contrast to, e.g., cut-cell methods, and (iv) there is no need to solve a pressure Poisson-equation for quasi-incompressible flow as the pressure and hence the acoustic field is an explicit result of the lattice-BGK algorithm. Furthermore, the LB method can be applied for low to high Knudsen numbers Kn.…”

Section: Lattice-boltzmann Methodsmentioning

confidence: 99%

Prediction of Acoustic Fields Using a Lattice-Boltzmann Method and Deep Learning

Rüttgers

Koh

Jitsev

et al. 2020

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Using traditional computational fluid dynamics and aeroacoustics methods, the accurate simulation of aeroacoustic sources requires high compute resources to resolve all necessary physical phenomena. In contrast, once trained, artificial neural networks such as deep encoder-decoder convolutional networks allow to predict aeroacoustics at lower cost and, depending on the quality of the employed network, also at high accuracy. The architecture for such a neural network is developed to predict the sound pressure level in a 2D square domain. It is trained by numerical results from up to 20,000 GPU-based lattice-Boltzmann simulations that include randomly distributed rectangular and circular objects, and monopole sources. Types of boundary conditions, the monopole locations, and cell distances for objects and monopoles serve as input to the network. Parameters are studied to tune the predictions and to increase their accuracy. The complexity of the setup is successively increased along three cases and the impact of the number of feature maps, the type of loss function, and the number of training data on the prediction accuracy is investigated. An optimal choice of the parameters leads to network-predicted results that are in good agreement with the simulated findings. This is corroborated by negligible differences of the sound pressure level between the simulated and the network-predicted results along characteristic lines and by small mean errors.

show abstract

Zonal Flow Solver (ZFS): a highly efficient multi-physics simulation framework

Cited by 25 publications

References 88 publications

A machine-learning-based method for automatizing lattice-Boltzmann simulations of respiratory flows

A machine-learning-based method for automatizing lattice-Boltzmann simulations of respiratory flows

Large-Eddy Simulations of Rim Seal Flow in a One-Stage Axial Turbine

Prediction of Acoustic Fields Using a Lattice-Boltzmann Method and Deep Learning

Contact Info

Product

Resources

About