جستجوی مقالات مرتبط با کلیدواژه « Turbulent boundary layer » در نشریات گروه « مکانیک »

The study of turbulent boundary layer trailing-edge noise is a fundamental issue in design and production of flying vehicles with minimum noise. Over the past decades, to reduce trailing-edge noise, various passive methods have been proposed, and the most recent strategy is the use of noise-reducing finlets. In the present study, to investigate the effect of noise reduction finlets on the turbulent flow field in the vicinity of the model trailing edge, a flat-plate model equipped with unsteady surface pressure transducers has been designed and built. Then, by installing a set of finlets with different lateral spacing on it, their effects on the flow field downstream of the finlets have been experimentally studied. Results show that the flow behavior downstream of the finlets is strongly affected by the spacing between the finlets. While the use of coarse finlets leads to a reduction in the mean velocity, turbulence intensity, and energy contents of low-frequency turbulent structures in the near-wall regions downstream of finlets, the flow behavior downstream of fine finlets are somewhat similar to the flow behind a backward-step. For fine finlets, the energy contents of turbulent structures in the near-wall regions are significantly reduced over the whole frequency range. he results of cross-correlation and coherence between surface pressure fluctuations and velocity field also showed that while the most important mechanism affecting surface pressure fluctuations for the coarse finlets is the exit low-momentum flow, for the fine finlets, vortical structures formed by flow separation have the most impact.

In this study, the effect of rectangular finlets on the trailing edge noise of a flat plate in an incompressible turbulent boundary layer with Mach number 0.06 has been studied numerically. The finlets are modeled upstream of the trailing edge of the flat plate with two spanwise spacing of 1.5 and 0.9 mm. Large eddy simulation (LES) approach with Lund inflow generation model in open source code of OpenFOAM has been used to simulate the turbulent boundary layer flow. Probe utility has been used for pressure and velocity data acquisitions. The finlets have increased the PSD of pressure fluctuations in the low to mid frequencies and decreased it at high frequencies between and downstream of the finlets. Also, the spanwise length scale of the pressure fluctuations increases but the eddy convection velocity is reduced and their effects has intensified by the reduction of the spanwise spacing. The mean velocity decreases but the turbulence intensity increases downstream of the finlets. The reduction of the spanwise spacing of the finlets results in the formation of a stronger shear layer and the extension of the shear sheltering area downstream of the finlets. According to the results of the far-field noise prediction with Curle analogy, finlets with the spacing of 1.5 mm have slightly reduced the noise about 0.3dB but the finlets with the spacing of 0.9 mm have increased the trailing edge noise of the flat plate, about 1dB.

In the present study, the efficiency of the finlet as a means of passive trailing-edge noise control has been experimentally investigated. Surface pressure spectra, spanwise length scale, and eddy convection velocity in the trailing-edge region are important parameters in determining far-field trailing-edge noise. In the present study to measure the above parameters, a flat-plate model equipped with unsteady surface pressure transducers has been designed and built. Results have shown that the flow behavior downstream of the finlets is strongly affected by the spacing between the finlets. The use of finlets with coarse spacing leads to a reduction in the surface pressure spectrum at mid to high frequencies and an increase in the spanwise length scale at low to mid frequencies. On the other hand, for the finlets with fine spacing, while the surface pressure spectrum has been further reduced at high frequencies, there has been an undesirable increase at low to mid frequencies. Moreover, fine finlets can significantly reduce the coherence and eddy convection velocity at mid to high frequencies. Finally, the Amiet-Roger model has been used to evaluate the changes in far-field trailing-edge noise and the results have shown the effectiveness of finlets in the mid and especially high frequency range.

In the present study, one of the most important mechanisms of aerodynamic noise generation is investigated numerically. Large-eddy simulation approach used to solve the unsteady flow equations of turbulent boundary layer with Mach number 0.06 over a flat plate of length 30-cm. Lund's inflow boundary model used to reduce the computational cost. To calculate the parameters affecting trailing edge noise (include surface pressure spectra, spanwise length scale of the surface pressure fluctuations and eddy convection velocity), data acquisition of SPFs values in different points over the flat plate surface done using the probe tool in OpenFOAM software. Finally, the farfield noise predicted using analytical Amiet-Roger model. The results showed that the numerical solution method used in this study, while having reasonable computational cost, has a good ability to predict the effective parameters on the trailing edge noise. In addition, studying the spectral parameters affecting the turbulent boundary layer trailing edge noise showed that prediction and direct estimation of these parameters, while are using in far-field noise prediction models, provide accurate and proper information on physics of the flow and dimensions and life time of turbulent boundary layer vortex structures.

In this research, a standard propeller was designed for a surface vessel, then using a computer program, 3D propeller geometry based on two- dimensional data generated and griding and flow field around the propeller was solved based on the computational fluid dynamics method. In this regard, we used a hybrid grid and periodic condition based on a spline Cutting. A steady model and three unsteady models were used in the numerical modeling. The results were validated and uncertainty grid discretization error was calculated. The results were in good agreement with the results of open water, an estimated separation point and propeller wake were investigated. In order to modeling the transition of the boundary layer, two models of gamma and γ-Reθ transition were used. The Q criterion was used to detection vortices and wakes. The separation of flow and transition of the boundary layer from laminar to turbulent in high advanced coefficients in the unsteady models was different of the steady model. The effects of adding two standard duct with 19A and 37A with open propeller were compared that 19A duct had better hydrodynamic properties.

In this paper, the pressure and temperature distributions over the ramp of an Aerospike nozzle in different pressure ratios are presented. Governing equations of the turbulent flow have been solved using the finite volume method and different turbulence models. By comparing the obtained results with available experimental data, SST k-ω model has been selected for numerical simulations. In this study, the length of the recirculation zone, the temperature, and pressure variations within the recirculation zone and locations of the separation and reattachment points have been investigated in detail. Results show that the increase in the pressure ratio can move separation, reattachment points, and recirculation zone to downstream of the flow and also increase the length of the recirculation zone simultaneously. The maximum and minimum temperatures did not change but the maximum and minimum pressures reduced. Also, the results show that the reattachment point has the maximum temperature of the flow and this maximum temperature point can be adopted as the reattachment point.

Requirement of reducing undesirable noise from airplanes, wind turbines blades and fans resulted in the identification of noise sources of such equipment in many research works. Turbulent boundary layer trailing edge (T.E.) noise is one of the main sources of aerodynamic noise and extensive studies have been conducted on it during the past decades. Previous studies have shown that surface pressure spectra, the spanwise length scale of the SPFs and eddy convection velocity in the T.E. region are crucial quantities in determination of the far-field trailing edge noise. In the present study, for measuring these parameters and hence predicting the far-field trailing edge noise, a flat-plate model is designed and built. The flat plate is equipped with several streamwise and spanwise miniature condenser microphones, FG-23329-P07, acting as surface pressure transducers. All the experiments were carried out in a subsonic wind tunnel for three free-stream velocities: 10, 15 and 20 m/s. The spanwise length scale and eddy convection velocity are calculated by simultaneously measuring of unsteady surface pressure respectively in both streamwise and spanwise directions and using cross spectral density. The experimental results, including the surface pressure spectra, longitudinal and lateral coherence and eddy convection velocity provide many information about the flow field structure in the turbulent boundary layer. The results also show that the best collapses in the surface pressure spectra at low frequency and middle to high frequencies can be obtained by using outer and inner layer scales respectively. Furthermore, the longitudinal and lateral coherences can provide adequate information about the lifespan (or, inversely, the decay) of the eddies and their physical size. Finally, the trailing edge noise from flat plate has been predicted by using the analytical Amit-Roger model and results show the effectiveness of this model for prediction of far-field turbulent boundary layer trailing edge noise.

In this study, the impacts of three different inflow conditions on turbulent boundary layer characteristics over a flat plate was numerically investigated. The investigated boundary conditions are uniform velocity profile at inlet with turbulent generation using strip, precursor-based simulation model (TVMF) and rescaling/recycling model purposed by Lund, which are applicable in large-eddy simulation (LES). The simulations are done using LES with dynamic smagorinsky coefficient in OpenFOAM software. Validation of the numerical method was done using direct numerical data and analytical solution. Comparison of the skin friction coefficient with analytical equation shows that the maximum deviation in simulations with precursor-based, Lund and strip model are 4.3%, 4.5%, 19.2% accordingly. The numerical values and development trend of intergral thicknesses, velocity components and the other characteristics obtained through simulations using precursor-based and Lund inlet models, are in good agreement with direct numerical simulation (DNS) results while the results obtained by using the strip model are more diverse from the DNS results. The results of this study indicate that using precursor-based and Lund inflow generation models in LES produces more realistic results accompanied with reduced computational cost.

Surface pressure fluctuations beneath turbulent boundary layer on a flat plate have complex physical behavior and due to its importance in acoustic noise generation, extensive studies have been devoted to predicting or measuring the surface pressure behavior. In the present study to investigate the surface pressure fluctuations under zero pressure gradient, a flat plate with a chord length of 580 mm has been used. All experiments were carried out in a subsonic wind tunnel and at three free-stream velocities: 10, 15 and 20 m/s. In order to measure unsteady pressure fluctuations, a condenser microphone is used as a pressure transducer. Moreover, various parameters of turbulent boundary layer are measured to provide the input variables of semi-empirical models. A single constant temperature hot-wire anemometer has been used for boundary layer measurement. Surface pressure spectra has been measured at various velocities and their collapse on a single curve by normalizing with different variables of turbulence boundary layer is studied. The results show that the best collapses in low and middle frequencies can be obtained by using mixed variables. However, in high frequency range the pressure spectra collapses when it is normalized by inner layer scales. Finally, after ensuring the accuracy of surface pressure spectra results, the efficiency of semi-empirical models for predicting turbulent boundary layer wall pressure spectra is evaluated. The results show the effectiveness of the Goody’s semi-empirical model for prediction of surface pressure spectra by using turbulent boundary layer parameters.

Experimental evaluation of overall convection heat transfer coefficient of a rectangular bar in the vicinity of a flat plate is investigated. A quad of rectangular shape and later a quad with the cause of optimal heat transfer are placed at the near and the inside of a turbulent boundary layer over a flat plate. The overall convection heat transfer coefficient of the flat plate are measured and compared to the case similar to a single flat plate. A low speed wind tunnel is employed to maintain main flow field at the requested speed and special electrical circuit is prepared to provide heat and measure heat losses from the flat plate. Conclusion is made that when the obstacle closes to the flat plate، the total convective heat transfer coefficient increases to a maximum and then reduces again by moving towards the plate. Distance from the flat in which the maximum heat transfer coefficient occurs is reported.