Search

The optimal Reynolds-Averaged-Navier Stokes (RANS) turbulence model to be used in a Computational Fluid Dynamics (CFD) simulation varies depending on the application. Conventionally, the model is selected from benchmark tests and experience, but its performance is difficult to predict. For this reason, this study presents a cost-effective CFD validation routine, which uses three-dimensional experimental velocity data obtained in replicas of the specific flow system. Magnetic Resonance Velocimetry (MRV) is used as the measurement technique. Since the objective is only the validation of the turbulence model, the experiment and the simulation are performed with simplified flow conditions, hence, stationary iso-thermal iso-volumetric flow without inertial forces. The routine applies a data matching routine to align the two three-dimensional data sets before they are interpolated on a common grid. Various error metrics are presented, which provide the degree of the CFD modelling error and indicate its source. For demonstration, the validation routine is used to evaluate RANS-CFD results of a three-pass internal cooling system of a high-pressure turbine airfoil used in a small industrial gas turbine. The simulations are performed with the eddy-viscosity based turbulence model k-w SST and the Reynolds-Stress SSG, and BSL-EARSM turbulence models. The results indicate strong local errors in the examined turbulence models. None of the models performed well enough, underlining that every RANS-CFD application needs to be validated.

Magnetic Resonance Velocimetry (MRV), an imaging method based on Magnetic Resonance Imaging (MRI), enables the measurement of flow parameters such as the velocity and the Reynolds Stress Tensor (RST) in complex structures without optical or physical access to the flow field. Several previous studies investigated the application of MRV velocity measurement in technical flows and obtained results that agreed well with reference data. However, only a few studies have investigated RST measurements using MRV beyond medical applications, and even though the qualitative results were promising, further work is required to establish this method. This study demonstrates the application of two-dimensional three-component (2D3C) velocity and six-component (2D6C) RST measurements in the flow field behind the sudden expansion of a scaled replica of the FDA benchmark nozzle. Particle Image Velocimetry (PIV) data accessible from an interlaboratory study was used for comparison. Furthermore, two different orientations of the imaging plane were measured to investigate the effect of the imaging orientation on the results. The measurement uncertainty of the mean axial velocity is 1.2 % related to the bulk velocity. The RST results agree well with the PIV data, but quantitative deviations occur in the areas where the influence of systematic errors was expected. Comparing different imaging orientations demonstrates that the sequence design affects the quantitative results of the measurement.

Computational fluid dynamics (CFD) is the standard tool in the turbomachinery industry to analyze and optimize internal cooling systems of turbine components, but the code applied has to be validated. This paper presents a combined experimental and numerical study on the flow field and heat transfer in a cooling system consisting of a three-pass serpentine with rib turbulators and trailing edge ejection. The cooling geometry is taken from a stator vane currently used in an industrial gas turbine and operates at a coolant inlet Reynolds number of 45,000. As an experimental technique, magnetic resonance velocimetry (MRV) was used to obtain the three-dimensional time-averaged velocity field of the isothermal flow. The measurements were conducted in a large-scale model and resulted in 3.2 million velocity vectors and measurement uncertainty of 6.1% of the bulk inlet velocity. The local wall heat transfer was measured in a separate experiment using thermochromic liquid crystals (TLC). These measurements yielded the distribution of the heat transfer coefficient on both the pressure and the suction side internal walls with a measurement uncertainty of 12%. The experimental data are used as a reference for the numerical study. In total, eight turbulence models are evaluated here, including one-equation, two-equation, algebraic and differential Reynolds stress models, and a scale adaptive simulation. The results show the differences between the velocity fields and the heat transfer coefficient distribution, allowing for the identification of the optimum turbulence model for this particular type of flow.

Computational Fluid Dynamics (CFD) is the standard tool in the turbomachinery industry to analyze and optimize internal cooling systems of turbine components, but the code applied has to be validated. This paper presents a combined experimental and numerical study on the flow field and heat transfer in a cooling system consisting of a three-pass serpentine with rib turbulators and trailing edge ejection. The cooling geometry is taken from a stator vane currently used in an industrial gas turbine and operates at a coolant inlet Reynolds number of 45,000. As an experimental technique, magnetic resonance velocimentry (MRV) was used to obtain the three-dimensional time-averaged velocity field of the isothermal flow. The measurements were conducted in a large-scale model and resulted in 3.2 million velocity vectors and a measurement uncertainty of 6.1% of the bulk inlet velocity. The local wall heat transfer was measured in a separate experiment using thermochromic liquid crystals (TLC). These measurements yielded the distribution of the heat transfer coefficient on both the pressure and the suction side internal walls with a measurement uncertainty of 12%. The experimental data are used as reference for the numerical study. In total, eight turbulence models are evaluated here, including one-equation, two-equation, algebraic and differential Reynolds stress models, and a scale-adaptive simulation. The results show the differences between the velocity fields and the heat transfer coefficient distribution, allowing for the identification of the optimum turbulence model for this particular type of flow.

The optimal turbulence model to be used in a CFD simulation varies depending on the application and the scientific question at hand. It is always a difficult task to predict which model best represents reality. This study focuses on the internal cooling system of a high-pressure turbine airfoil. It presents a CFD validation routine, which includes a data matching process with experimental data as well as global and local error metrics to quantify the differences between the experiment and CFD. The experimental data of the time-averaged three-dimensional velocity field is obtained with Magnetic Resonance Velocimetry (MRV). The airfoil in question is a second stage blade of a small industrial gas turbine, consisting of a three-pass internal cooling serpentine. The three-dimensional three-component MRV data is compared to CFD using the data matching process and the error metrics. The simulations are performed with the eddy-viscosity based turbulence model k-ω SST and the Reynolds-Stress SSG, and BSL-EARSM turbulence models. The results indicate local differences between the examined turbulence models and the MRV results. Referring to the global and local error quantities presented in the study, the BSL turbulence model is identified as the optimal turbulence model for this specific high-pressure turbine airfoil.

Predicting blood flow velocities in patient-specific geometries with Computational Fluid Dynamics (CFD) can provide additional data for diagnosis and treatment planning but the solution can be inaccurate. Therefore, it is crucial to understand the simulation errors and calibrate the numerical model. In-vitro velocity-encoded MRI is a versatile tool to validate CFD. The comparison between CFD and in-vitro MRI velocity data, and the analysis of the simulation error are the objectives of this study. A three-step routine is presented to validate medical CFD data. First, a properly scaled model of the patient-specific geometry is fabricated to achieve high relative resolution in the MRI experiment. Second, the measured flow geometry is matched with the CFD data using one of two algorithms, Coherent Point Drift and Iterative Closest Point. The aligned data sets are then interpolated onto a common grid to enable a point-to-point comparison. Third, the global and local deviations between CFD and MRI velocity data are calculated using different algorithms to reliably estimate the simulation error. The routine is successfully tested with a patient-specific model of a cerebral aneurysm. In conclusion, the methods presented here provide a framework for CFD validation using in-vitro MRI velocity data.

An established therapy for aortic valve stenosis and insufficiency is the transcatheter aortic valve replacement. By means of numerical simulation the valve dynamics can be investigated to improve the valve prostheses performance. This study examines the influence of the hemodynamic properties on the valve dynamics utilizing fluidstructure interaction (FSI) compared with results of finiteelement analysis (FEA). FEA and FSI were conducted using a previously published aortic valve model combined with a new developed model of the aortic root. Boundary conditions for a physiological pressurization were based on measurements of ventricular and aortic pressure from in vitro hydrodynamic studies of a commercially available heart valve prosthesis using a pulse duplicator system. A linear elastic behavior was assumed for leaflet material properties and blood was specified as a homogeneous, Newtonian incompressible fluid. The type of fluid domain discretization can be described with an arbitrary Lagrangian-Eulerian formulation. Comparison of significant points of time and the leaflet opening area were used to investigate the valve opening behavior of both analyses. Numerical results show that total valve opening modelled by FEA is faster compared to FSI by a factor of 5. In conclusion the inertia of the fluid, which surrounds the valve leaflets, has an important influence on leaflet deformation. Therefore, fluid dynamics should not be neglected in numerical analysis of heart valve prostheses.

In-stent thrombosis is a major complication of stent implantations. Unlike pathological occurrences as in-stent restenosis for instance, thrombosis represents an acute event associated with high mortality rates. Experiments show that low wall shear stress promotes undirected endothelial cell coverage of the vessel wall and therefore increases the risk of thrombus formation. Stent design represents a crucial factor influencing the surface areas of low wall shear stress and thus the incidence of acute in-stent thrombosis. In this study, we present an optimization method for stent designs with minimized thrombosis risk. A generic stent design was developed, based on five different stent design parameters. Optimization was conducted based on computational fluid dynamics analysis and the gradient-free Nelder-Mead approach. For each optimization step, a numerical fluid simulation was performed in a vessel with a reference vessel diameter of 2.70 mm with stent-overexpansion ratio of 1.0:1.1. For each numerical fluid simulation a physiological Reynolds number of 250, resulting in a mean velocity of 0.331 m/s at the inlet and a laminar flow as well as stiff vessel walls were assumed. The impact of different stent designs was analyzed based on the wall shear stress distribution. As a basis for the comparison of different stent designs, a dimensionless thrombosis risk number was calculated from the area of low wall shear stress and the overall stented area. The first two optimization steps already provide a decrease of thrombosis risk of approximately 83%. In conclusion, computational fluid dynamic analyses and optimization methods usind the Nelder-Mead approach represent a useful tool for the development of hemodynamically optimized stent designs with minimized thrombosis risk.

In tissue engineering and regenerative medicine mesenchymal stem cells (MSC) are widely used to replace and restore the function of dysfunctional or missing tissue. Recent studies have shown significant enhancements of the in vivo healing process following dentofacial bone augmentation procedures employing stem cell-loaded xenografts. We conducted experimental and numerical investigations in perfusion flow bioreactor-xenograft-systems to identify flow conditions as well as bioreactor design features that allow for homogeneous MSC-distribution in Geistlich Bio- Oss Block xenografts. Pressure gradient - velocity characteristics and flow distributions were investigated experimentally and numerically for two bioreactor designs at steady-state flow conditions with Reynolds numbers (Re) ranging from 0.01 ≤ Re ≤ 0.32. Distilled water at 20°C with a dynamic viscosity of 1.002 mPa∙s and a density of 998 kg/m3 was used. The geometry of the xenograft utilized in three-dimensional computational fluid dynamics (CFD) simulation was obtained by means of micro-computed tomography (μCT) at an isotropic spatial resolution of 9.5 μm. The permeability values calculated from the experimental data are in good accordance with the numerical results. The investigations showed that the increase of the inflow- and outflow-area diameter, as well as the decrease of the volumetric flow rate, result in a decreasing heterogeneity of the flow distribution within the xenograft. The calculated wall shear stress rates in the three-dimensional (3D) scaffold range from 1∙10-12Pa ≤ τ ≤ 0.2 Pa. Experimentally validated CFD simulations introduced in this study provide an applicable tool to assess optimal flow conditions for homogeneous MSC distribution in bioreactor-xenograft-systems.

For the treatment of severe symptomatic aortic valve stenosis, minimally invasive heart valve prostheses have more recently become the lifesaving solution for elderly patients with high operational risk and thus, are often implanted in patients with challenging aortic root configuration. A correct prosthesis deployment and stent adaption to the target region is essential to ensure optimal leaflet performance and long-term prosthesis function. The objective of this study was the development of a suitable in silico setup for structural numerical simulation of a transcatheter aortic valve (TAV) in different cases of clinical relevance. A transcatheter valve prosthesis comprising an unpressurized trileaflet heart valve and an adapted stent configuration was designed. An aortic root (AR) model was developed, based on microcomputed tomography of a native healthy specimen. Using the finite-element analysis (FEA), various loading cases including prosthesis biomechanics with valve opening and closing under physiological pressure ratios throughout a cardiac cycle, prosthesis crimping as well as crimping and release into the developed AR model were simulated. Hyperelastic constitutive law for polymeric leaflet material and superelasticity of shape memory alloys for the self-expanding Nitinol stent structure were implemented into the FEA setup. Calculated performance of the valve including the stent structure demonstrated enhanced leaflet opening and closing as a result of stent deformation and redirected loading. Crimping and subsequent release into the AR model as well as the stent adaption to the target region after expansion proved the suitability of the TAV design for percutaneous application. FEA represented a useful tool for numerical simulation of an entire minimally invasive heart valve prosthesis in relevant clinical scenarios.

In dentofacial surgery, augmentation procedures employing xenografts have become a reliable treatment. Recent studies, however, have shown significant enhance-ments of the in vivo bone tissue augmentation using mesenchymal stem cells loaded into bone grafts. We conducted experimental and numerical investigations in flow perfusion systems to determine flow conditions which allow for homogenous stem cell distribution in BioOss Block (Geistlich Pharma AG, Switzerland) xenografts. Pressure gradient-velocity characteristics and flow distributions were investigated experimentally and numerically at steady state flow conditions with Reynolds numbers (Re) ranging from 0.01 ≤ Re ≤ 0.40. Distilled water at 20°C with a dynamic viscosity of 1.002 mPa.s and a density of 998 kg/m3 was used. The geometry utilized in three-dimensional computa-tional fluid dynamics (CFD) simulation was obtained by means of micro-computed tomography (μCT). Results of CFD analysis are in good accordance with experimental data. The comparison of the pressure gradient-velocity characteris-tics for experimental and numerical data yields a relative error of 3.6%. According to Darcy’s law for creeping fluid flow the experimentally determined permeability is 2.55.10-9 m2. Moreover, numerical flow distribution analysis shows an increasingly heterogenic streamline distribution for increasing Reynolds numbers. Experimentally validated CFD simulations introduced in this study provide a tool to assess optimal flow conditions for a homogenous stem cell distribution in perfusion flow systems.

Computational fluid dynamics (CFD) is a powerful tool to extent knowledge of biomechanical processes in cardiovascular implants. To provide a standardized method the U.S. Food and Drug Administration (FDA) initialized a CFD round robin study. One of the developed benchmark standard models is a generic nozzle geometry, consisting of a cylindrical throat with a conical collector and sudden expansion on either side. Several fluid mechanical data obtained from international institutes by means of CFD and particle image velocimetry (PIV) measurements under different flow regimes (Re = 500, 2000, 3500, 5000 and 6500) are freely available. This database includes only steady state simulations. In this study we performed pulsatile CFD simulations to consider the physiological environment of the coronary vessels. Furthermore, the nozzle geometry was scaled down to coronary dimension (D inlet = 12 mm to 3 mm) while retaining the average Reynolds number Re = 500 constant. The pulsatile character is described by a Womersley number of Wo = 2.065. Our CFD code was previously validated by using FDA’s data for steady state inflow conditions. It could be shown that time averaged wall shear stress and shear stress values agree well with steady state results. We conclude that steady state simulations are valid for hemodynamic analyses if only time averaged values are needed. This could save computational costs of future hemodynamic investigations. In addition, this study expands FDA’s benchmark case by pulsatile inlet condition for further code validation. This could be necessary for the development of new numerical methods as well as for validation of CFD codes used in the approval process of medical devices.

Alteration of the flow characteristics in coronary vessels is correlated with coronary heart disease (CHD). In particular, wall shear stress (WSS) appears to be a hemody-namic key factor in the genesis of CHD. Since computational fluid dynamics (CFD) is a well-known method for the inves-tigation of WSS, it may be a valuable tool for the prediction of CHD. Latest imaging techniques, such as optical coher-ence tomography (OCT) in conjunction with angiography deliver precise 2D data sets of patient-specific vessel geome-try, which can be used for CFD analysis. Current CFD stud-ies utilize patient-specific geometries, but are lacking well defined physiologic inflow conditions. In this study, we present an inflow mapping method for patient-specific arterial vessels, which is capable of consider-ing the influence of bifurcations located proximal of the OCT-data set. At first, the patient-specific vessel was recon-structed. For this purpose the OCT-based vessel cross sec-tions were arranged along an angiographic based vessel pathway. Secondly, we simulated the flow field in a generic bifurcation model by means of CFD. Thereafter the flow field of a side branch was extracted and transferred (mapped) to the inlet of the patient-specific vessel. To evaluate the influence of the physiological inlet the WSS distribution of the same patient-specific vessel was calculated using an axial-symmetric inflow condition. Analy-sis of the simulation data yielded deviations of the WSS distribution in the proximal vessel segment. A bifurcation, located upstream of the relevant vessel segment strongly affects the flow in the OCT-based vessel reconstruction and has a strong influence on the results of the numerical analy-sis. Therefore, it is important to implement not only the pa-tient-specific geometry, but also an inlet boundary condition adapted to the upstream velocity distribution reflecting the actual proximal flow situation of the vessel.

A new test facility is presented, enabling single-phase isothermal fluid mechanics experiments to support nuclear reactor safety studies. The applied measurement technique Magnetic Resonance Velocimetry (MRV) is based on the medical imaging modality Magnetic Resonance Imaging (MRI). The experiments in this laboratory aim at providing the mean velocity field in models of nuclear fuel assemblies using isothermal water. Additionally, turbulence statistics in terms of the Reynolds stress tensor can be acquired. Since MRV is a non-optical measurement technique, there are no requirements on optical properties, and the investigated models can be quickly built with additive manufacturing techniques. Even though MRV offers a relatively low temporal and spatial resolution compared to laser-optical techniques and has some limitations regarding the test section materials, e.g., no metals, the main gain is that MRV provides a full three-dimensional representation of the flow field in short times. Both the model preparation and the measurement time are significantly faster than with laser-optical measurement techniques. The full-field experimental data obtained with MRV is often referred to as 'CFD grade data' as it enables a full-field validation of numerical solutions obtained with Computational Fluid Dynamics (CFD). As a commissioning experiment, measurements inside a 5x5 fuel element model of a pressurized water reactor (PWR) are conducted and compared with CFD results. The measured mean velocity field inside the PWR replica comprised of 2.9 Million velocity vectors was acquired in 79 min. In addition, turbulence measurements in a periodic hill channel demonstrate how MRV can be used to obtain all components of the Reynolds stress tensor. Although the measurements in this proof-of-concept study suffer from some measurement errors, all major error contributions are well known, and validated methods exist to overcome these shortcomings. In conclusion, this study shows how the velocity field in a specific fuel assembly configuration can be obtained highly efficiently using a specialized MRI laboratory.

Coronary diseases are the main reason for death in the western world. Bio-fluid mechanical correlations with arterial diseases are in the focus of our research. To treat occluded vessels, stents are implanted. Stent implantations can be associated with blood flow disruptions leading to restenosis or thrombosis formation. Numerical flow simulation is a promising tool to evaluate the hemodynamic performance of cardiovascular implants, but is resource-intensive in time and computational power. Therefore, a reduction in grid size would be beneficial due to economic exploitation of computational cost. The purpose of this numerical investigation is to substitute the computational domain of a distal stented bifurcation with a stented straight vessel by using the right inlet condition. The deviation of the results of the two different methods to simulate the blood flow situation in a bifurcation is marginal. This inlet can be used for standardised simulations of bifurcations were lesions commonly occur.

Stent implantation is the treatment of choice for cardiovascular diseases. By introduction of biodegradable thick strut stents investigations of thrombosis formation is one focus of research. This study deals with a transient Euler-Lagrange/DEM approach to simulate the flow field, platelet movement and clotting. The recirculation zones prolong particle residence time. As a result, the vicinity of stent struts shown a particularly higher risk for stent thrombosis.

Bioresorbable scaffolds (BRS) promise to be the treatment of choice for stenosed coronary vessels. But higher thrombosis risk found in current clinical studies limits the expectations. Three hemodynamic metrics are introduced to evaluate the thrombosis risk of coronary stents/scaffolds using transient computational fluid dynamics (CFD). The principal phenomena are platelet activation and effective diffusion (platelet shear number, PSN), convective platelet transport (platelet convection number, PCN) and platelet aggregation (platelet aggregation number, PAN) were taken into consideration. In the present study, two different stent designs (thick-strut vs. thin-strut design) positioned in small- and medium-sized vessels (reference vessel diameter, RVD=2.25 mm vs. 2.70 mm) were analyzed. In both vessel models, the thick-strut design induced higher PSN, PCN and PAN values than the thin-strut design (thick-strut vs. thin-strut: PSN=2.92/2.19 and 0.54/0.30; PCN=3.14/1.15 and 2.08/0.43; PAN: 14.76/8.19 and 20.03/10.18 for RVD=2.25 mm and 2.70 mm). PSN and PCN are increased by the reduction of the vessel size (PSN: RVD=2.25 mm vs. 2.70 mm=5.41 and 7.30; PCN: RVD=2.25 mm vs. 2.70 mm=1.51 and 2.67 for thick-strut and thin-strut designs). The results suggest that bulky stents implanted in small caliber vessels may substantially increase the thrombosis risk. Moreover, sensitivity analyses imply that PSN is mostly influenced by vessel size (lesion-related factor), whereas PCN and PAN sensitively respond to strut-thickness (device-related factor).