Die Zerlegung von Beobachtungsgrößen in sogenannte Mittelwerte und Fluktuationen führt zur Parametrisierung des turbulenten Flusses aber auch zu Problemen. Der Erwartungswert der turbulenten Größe ψ ist das Ensemble Mittel über eine große Zahl von Realisierungen, falls ψ normalverteilt ist. Geophysikalische Daten bestehen jedoch aus Zeitreihen und/oder räumlichen Daten. Daher muß jeder vernünftige Mittelungsprozeß von ψ in der Zeit und/oder im Raum durchgeführt werden. Um die Fluktuationen ψ'' von ψ zu trennen, müssen wir den Erwartungswert von ψ bestimmen, d.h. die Fluktuationen in den langen Zeit- und/oder Raumskalen. Für dieses Problem ist der Mittelwert über das Meßintervall eine schlechte Approximation, das gleitende Mittel eine bessere und der numerisch tiefpassgefilterte Wert die bestmögliche Approximation. Eine Fluktuationsmessung (surface flux) im Bereich niedriger Flüsse wurde ausgewertet 1) nach der gewöhnlichen Methode und 2) mit einem numerischen Tiefpass Lanczos-Filter. Mit 2) erhielten wir bessere Ergebnisse. / Decomposition of some observables into so-called mean parts and fluctuations leads to parameterisation of turbulent flow but is also the cause of different problems. The expectation of the turbulent field ψ is, the ensemble mean over a large number of realizations if ψ follows a normal distribution. Geophysical data, however, consist of time- and/or space series. Thus every reasonable averaging process of ψ must be over time and/or space. To separate fluctuations ψ'' from ψ we must estimate the expectation value of ψ, i.e. fluctuations on long time and/or space scales. For this problem the mean over the measuring interval is an inexact approximation, the moving mean is better but the numerically low-pass filtered value probably the best possible approximation. A surface flux measurement in low flux regime is evaluated with 1) a usual procedure and 2) with a numerical low-pass Lanczos-filter. With 2) we obtain better results.

No description available.

Predicting wall heat flux accurately in wall-bounded turbulent flows is critical for a variety of engineering applications, including thermal management systems and energy-efficient designs. Traditional methods, which rely on expensive numerical simulations, are hampered by increasing complexity and extremly high computation cost. Recent advances in deep neural networks (DNNs), however, offer an effective solution by predicting wall heat flux using non-intrusive measurements derived from off-wall quantities. This study introduces a novel approach, the convolution-compacted vision transformer (ViT), which integrates convolutional neural networks (CNNs) and ViT to predict instantaneous fields of wall heat flux accurately based on off-wall quantities including velocity components at three directions and temperature. Our method is applied to an existing database of wall-bounded turbulent flows obtained from direct numerical simulations (DNS). We first conduct an ablation study to examine the effects of incorporating convolution-based modules into ViT architectures and report on the impact of different modules. Subsequently, we utilize fully-convolutional neural networks (FCNs) with various architectures to identify the distinctions between FCN models and the convolution-compacted ViT. Our optimized ViT model surpasses the FCN models in terms of instantaneous field predictions, learning turbulence statistics, and accurately capturing energy spectra. Finally, we undertake a sensitivity analysis using a gradient map to enhance the understanding of the nonlinear relationship established by DNN models, thus augmenting the interpretability of these models. / <p>Presentation online</p>

Predicting wall heat flux accurately in wall-bounded turbulent flows is critical fora variety of engineering applications, including thermal management systems andenergy-efficient designs. Traditional methods, which rely on expensive numericalsimulations, are hampered by increasing complexity and extremly high computationcost. Recent advances in deep neural networks (DNNs), however, offer an effectivesolution by predicting wall heat flux using non-intrusive measurements derivedfrom off-wall quantities. This study introduces a novel approach, the convolution-compacted vision transformer (ViT), which integrates convolutional neural networks(CNNs) and ViT to predict instantaneous fields of wall heat flux accurately based onoff-wall quantities including velocity components at three directions and temperature.Our method is applied to an existing database of wall-bounded turbulent flowsobtained from direct numerical simulations (DNS). We first conduct an ablationstudy to examine the effects of incorporating convolution-based modules into ViTarchitectures and report on the impact of different modules. Subsequently, we utilizefully-convolutional neural networks (FCNs) with various architectures to identify thedistinctions between FCN models and the convolution-compacted ViT. Our optimizedViT model surpasses the FCN models in terms of instantaneous field predictions,learning turbulence statistics, and accurately capturing energy spectra. Finally, weundertake a sensitivity analysis using a gradient map to enhance the understandingof the nonlinear relationship established by DNN models, thus augmenting theinterpretability of these models

This work presents the experimental study of the flow generated in the wakes of three three-dimensional bumps in the Virginia Polytechnic Institute and State University Boundary Layer Wind Tunnel. The three bumps examined are named bump 1, small bump 3, and large bump 3, and are the same test cases studied by Byun et al. (2004) and Ma and Simpson (2004) with a LDV system and a quad-wire hot-wire probe, respectively. Various experimental methods are used in this work: For measuring the mean velocity component in the planes examined, a seven-hole pressure probe is used with the data reduction algorithm developed by Johansen et al. (2001). A sixteen-hole pressure rake is used for boundary layer data on the sidewalls and ceiling of the test section and a Pitot-static probe is used to obtain mean velocity magnitude in the centerline of the test section. Specific techniques are developed to minimize the uncertainties due to the apparatus used, and an uncertainty analysis is used to confirm the efficiency of these techniques. Measurements in the wake of bump 1 reveal a strong streamwise vorticity creating large amounts of high moment fluid entrained close to the wall. In the wake of small bump 3, the amount of high momentum fluid entrained close to the wall is small as well as the streamwise vorticity. The flow in the wake of large bump 3 incorporate the characteristics of the two previous bumps by having a relatively large entrainment of high momentum fluid close to the wall and a low generation of streamwise vorticity. In the wakes of the three bumps, a pair of counter rotating vortices is created. The influence of large bump 3 on the incoming flow-field is found to be significant and induces an increase of the boundary layer thickness. By comparing LDV data and quad-wire hot-wire data with seven-hole probe data in the wakes of the bumps at the same locations, it is shown that uncertainties defined for a quasi-steady, non-turbulent flow-field without velocity gradient are bad indicators of the magnitude of the uncertainties in a more complex flow-field. A theoretical framework is discussed to understand the effects of the velocity gradient and of turbulence on the pressures measured by the seven-hole probe. In this fashion, a model is proposed and validated to explain these effects. It is observed that the main contribution to the uncertainties in seven-hole probe measurements due to the velocity gradient and to the turbulence comes from the velocity gradient. To correct for the effects of the velocity gradient on seven-hole probe measurements in an unknown flow-field, a technique is proposed. Using an estimation of the velocity gradient calculated from the seven-hole probe, the proposed model could be used to re-evaluate non-dimensional pressure coefficients used in the data reduction algorithm therefore correcting for the effects of the velocity gradient on seven-hole probe measurements. / Master of Science

Yes / This work presents new water surface elevation data including evidence of the spatial correlation of water surface waves generated in shallow water flows over a gravel bed without appreciable bed forms. Careful laboratory experiments have shown that these water surface waves are not well-known gravity or capillary waves but are caused by a different physical phenomenon. In the flow conditions studied, the shear present in shallow flows generates flow structures, which rise and impact on the water-air interface. It is shown that the spatial correlation function observed for these water surface waves can be approximated by the following analytical expression W(rho) = e(-rho 2/2 sigma w2)COS(2 pi L-0(-1)rho). The proposed approximation depends on the spatial correlation radius, sigma(w), characteristic spatial period, L-0, and spatial lag, . This approximation holds for all the hydraulic conditions examined in this study. It is shown that L-0 relates to the depth-averaged flow velocity and carries information on the shape of the vertical velocity profile and bed roughness. It is also shown that sigma(w) is related to the hydraulic roughness and the flow Reynolds number.

<p> An experimental study of turbulent flow of air around a 90º bend is reported in this thesis. Four 90º bends of aspect ratio 1, 3, 5 & 10 and radius ratio 1.0 have been tested in the Reynolds number range of 1 × 10⁵ to 5 × 10⁵. The loss in total pressure across the bend (or elbow) is reported for two discharge conditions (1) and the elbow discharging to a plenum chamber through a constant area duct of a length equal to 4 hydraulic diameters; (2) the elbow discharging to the plenum chamber directly. A comparison of the experimental results is made with the curves given in NACA report L4F26 which have been reproduced in the recently published SAE Aero-Space Manual. </p> <p> Various other flow parameters, such as velocity profiles, turbulence levels and pressure distributions are also given in this thesis. </p> / Thesis / Master of Engineering (ME)

Development of civilization faces a challenge of developing the resources of energy demand for the modern life. Extensive use of conventional fuel resources like crude oil and coal rise up a serious problem of increasing CO2 emission. New records levels of CO2 were registered during the early beginning of industrial revolution (http://climate.nasa.gov/evidence). Now a day’s more attention is oriented towards developing of biomass power stations owing to the increasing of conventional fuel prices and due to the potential to be CO2 neutral. One of the essential issues to successfully simulate and design efficient equipment for best utilization of the bio-fuel is to have better understanding of the interaction of bio-particles and the carrier gas. Almost, all two-phase flow system dealing with bio-mass power is turbulent flow. A unifying theory of turbulence does not yet exist. When particles are suspended into such a flow the flow becomes even more complicated and the resulting interactions between the particles and turbulent structures are not fully understood. For non-spherical particles, like most of the bio-mass particles found in cyclone filters and biomass gasification and combustion, the interactions of the particles and the fluid in turbulent flow are extremely complex while theories exists for low Reynolds number flow. The carrier phase turbulence alters the dispersed phase translational and rotational motion and the particles influence the detailed and overall flow of the carrier phase. The presence of the particles may also modify the turbulence of the fluid.To achieve my objective, to study the interaction of bio-particles with the carrier phase, and because of the complexity of the mechanisms related to such flow, it was essential to start to develop the knowledge on the possible mechanisms for the interactions and the importance of each of these interactions, see Paper A. Also, the controlling parameter which may have qualitative and/or quantitative influence of the flow interaction is covered by Paper A. To enable different types of experiments with PIV and LDA, a horizontal rectangular duct was designed and constructed. Design details and test is presented in Paper B. An introductory experimental series was performed in the current set-up using a high spatial resolution PIV system and the results can be found in Paper C.

No description available.

No description available.