Overview of Research on Hypersonic Flow and Computational

  • Slides: 21
Download presentation
Overview of Research on Hypersonic Flow and Computational Fluid Dynamics (CFD) Xiaolin Zhong MAE

Overview of Research on Hypersonic Flow and Computational Fluid Dynamics (CFD) Xiaolin Zhong MAE Department, UCLA MAE RTR Meeting, 4/27/2012 Research Supported by the NASA/AFOSR National Center for Hypersonic Laminar-Turbulent Transition Research, DOE, and NSF 1

Current Research Topics of Zhong Group Numerical Simulation of Hypersonic Boundary Layer Receptivity and

Current Research Topics of Zhong Group Numerical Simulation of Hypersonic Boundary Layer Receptivity and Transition (Supported by NASA/AFOSR National Center for Hypersonic Laminar-Turbulent Transition Research headed by Prof. Saric, TAMU) – Develop high-order numerical methods and computer codes for hypersonic viscous flow simulation with surface roughness and with thermal and chemical nonequilibrium. – Conduct numerical simulation studies receptivity, instability and transition of hypersonic boundary-layer flows with complex flow physics, including chemical reaction, ionization, surface ablation and roughness, etc. Direct numerical Simulation of Very Strong Shock and Turbulence Interaction Using Shock-Fitting Methods (Supported by DOE SCIDAC ) Numerical Study of Wave-Energy Extraction Device with Electroactive Polymer (Supported by NSF with Prof. Q. Pei) DNS of Transitional and Turbulent Hypersonic Boundary Layers (with Profs. Kim and Eldredge, NASA NRA) Inflow : Temporal DNS with Shock Fitting 2

Simulation of a Compression Cone with a Freestream Impulsive Entropy Spot in Mach 6

Simulation of a Compression Cone with a Freestream Impulsive Entropy Spot in Mach 6 Freestream Yuet Huang 3

Objectives • Development of computer code for simulating the interaction of a freestream hotspot

Objectives • Development of computer code for simulating the interaction of a freestream hotspot with the bow shock and the boundary layer in high order of accuracy. • Simulation of the hotspot perturbed shock layer on compression cone under the effect of a freestream hotspot. Schematic diagram of boundary layer receptivity to freestream disturbances • Analysis and investigation on the receptivity and instability growth of hotspot perturbed boundary layer based on the simulation results. • Lastly, comparison of our numerical results with those obtained by Professor Schneider’s group in the Purdue Mach-6 quiet tunnel. Schematic of wind-tunnel laser-spot experiment

Mean Flow for Compression Cone Cases Freestream flow conditions: • M = 6. 0

Mean Flow for Compression Cone Cases Freestream flow conditions: • M = 6. 0 • T = 52. 8 K • P = 610. 8 Pa • Twall= 300 K • Re /L= 1. 026 x 107 m-1 • Perfect gas Mach number contour plot Geometrical parameters: • Rflared = 3. 0 m • Cone half-angle θ = 2. 0 degrees • Cone length L ≈ 0. 45 m • Grid resolution = 3720 x 240 • R = 0. 001 m (Purdue) Numerical Method: • 5 th order finite-difference upwind schemes with local Lax-Friedrichs fluxsplitting scheme Non-dimensional density contour plot

Simulation Movie of Zone 1

Simulation Movie of Zone 1

Simulation Movie of Zone 12

Simulation Movie of Zone 12

Simulation Movie of Zone 18

Simulation Movie of Zone 18

Wall Pressure Perturbation Evolution x = 0. 0925 m x = 0. 0532 m

Wall Pressure Perturbation Evolution x = 0. 0925 m x = 0. 0532 m x = 0. 0483 m x = 0. 0404 m x = 0. 0337 m x = 0. 0278 m x = 0. 0224 m x = 0. 0175 m x = 0. 0135 m Upstream part of the cone

Wall Pressure Perturbation Evolution x = 0. 333 m x = 0. 313 m

Wall Pressure Perturbation Evolution x = 0. 333 m x = 0. 313 m x = 0. 273 m Downstream x = 0. 25 m part of the cone x = 0. 21 m x = 0. 17 m x = 0. 13 m

Normalized Pressure Perturbations Spectrum Nondimensional frequency spectrum of wall pressure perturbation • • •

Normalized Pressure Perturbations Spectrum Nondimensional frequency spectrum of wall pressure perturbation • • • Response coefficient. For a different shape function of freestream entropy perturbations with the same freestream condition and geometry. The amplitudes of all wave components can be obtained by multiplying the nondimensional freestream forcing amplitudes. The spatial development of wall pressure perturbation with 2 nd mode sampling frequencies in spectral domain • The lowest sampling frequency is neutral/growing. • The other four frequencies are growing with their own rates exponentially.

Stabilization and Control of Hypersonic Boundary Layer Instability and Transition by 2 -D Surface

Stabilization and Control of Hypersonic Boundary Layer Instability and Transition by 2 -D Surface Roughness Danny Fong and Le Duan 12

Motivations and Objectives 1. Roughness 2. Bow Shock 3. Synchronization point 4. Boundary Layer

Motivations and Objectives 1. Roughness 2. Bow Shock 3. Synchronization point 4. Boundary Layer 5. Pure Mode S, Mode F or a wall normal velocity pulse Ø This project is motivated by experimental and numerical results which show 2 -D roughness can possibly stabilize a hypersonic boundary layer. Ø Our goal is to utilize numerical simulation to study the role of roughness on the stability of a hypersonic boundary layer, mainly focus on roughness effect on transition delay. Ø In the current project, we consider 2 -D roughness on a hypersonic flat plate at Mach 5. 92. Ø The roughness is modeled as a hump and is treated by a high order cut cell method. In total, 4 different roughness heights and 4 different roughness locations have been considered. Ø Pure mode S (slow acoustic wave), mode F (fast acoustic wave) perturbation at 100 KHz, and wall normal velocity pulse which has a frequency range 1 MHz are imposed into the mean flow separately. Ø The evolution of disturbance on the wall due to different roughness height and locations are studied using FFT. 13

Visualization of the interaction of disturbance and roughness Mode S disturbance at 100 KHz

Visualization of the interaction of disturbance and roughness Mode S disturbance at 100 KHz Pulse disturbance (frequency range 1 MHz) 14

FFT results for 100 KHz slow acoustic disturbance (Mode S) with different roughness height

FFT results for 100 KHz slow acoustic disturbance (Mode S) with different roughness height and location Roughness located upstream of synchronization point Roughness located downstream of synchronization point ØUpstream of synchronization point, roughness amplifies disturbance. The overall amplification of disturbance depends roughness height. ØHowever, when roughness is placed downstream of synchronization point, it damps disturbance with strength related to roughness height. 15

Free stream waves receptivity to nonlinear breakdown over Mach 5. 5 circular cone Jia

Free stream waves receptivity to nonlinear breakdown over Mach 5. 5 circular cone Jia Lei ØThe objective of current study is to develop a Direct Numerical Simulation code to simulate hypersonic boundary layer flow from laminar to nonlinear breakdown in transition over blunt circular cone. ØOur innovative Two-step approach: 1) Build a linear receptivity database for different types of free stream waves with a selected frequency spectrum. 2) Construct the inflow condition for subsequent nonlinear region using receptivity database and conduct the breakdown simulation. Linear Stability Theory (LST) used to identify the range of unstable wave frequencies in the current case. Linear receptivity simulation were conducted to obtain free stream waves database. The database were used as entrance conditions for subsequent nonlinear breakdown simulation Linear growth region Non-linear breakdown region 3 D nonlinear breakdown simulation: Pressure disturbances contour at wall surface

Efficient control of hypersonic boundary-layer transition using surface porous coating Xiaowen Wang Background Ø

Efficient control of hypersonic boundary-layer transition using surface porous coating Xiaowen Wang Background Ø Porous coating significantly stabilizes Mack’s second mode (acoustics) whereas it moderately destabilizes first mode (T-S wave). Ø Porous coating covers either the entire flat plate or the surface around half the cone circumference. Ø No work is reported on the effect of porous coating location on the first-mode destabilization, and how to increase stabilization efficiency of porous coating. Flow conditions Feltmetal coating Regular porous coating: decreasing of porosity leads to even weaker first mode destabilization. At similar porosity, feltmetal coating is stronger in first mode destabilization than regular porous coating. The hypersonic boundary layer is destabilized or stabilized by feltmetal coating when it is upstream or downstream of the synchronization point. Conclusion Ø Put porous coating downstream of the synchronization point Ø Use regular porous coating if one must put porous coating along the whole surface Ø Design new porous coatings to attenuate first-mode destabilization Initial: r = 25 μm Design: r = 77 μm New designed porous coating with larger pore radius is more efficient in boundary-layer stabilization.

Development and validation of new high-order numerical method for hypersonic nonequilibrium flow simulations Xiaowen

Development and validation of new high-order numerical method for hypersonic nonequilibrium flow simulations Xiaowen Wang and Akshay Prakash Objectives Ø Develop and validate new high-order shock-fitting numerical method for hypersonic nonequilibrium flow simulations. Ø Conduct numerical simulation studies on transient growth of re-entry vehicles, including the effects of small/finite surface roughness, the interaction of surface roughness and freestram disturbances. Ø Update and validation of the code with more advanced 5 -species models and the more realistic 11 -species air models are ongoing. ØThe simulation of the steady flow over 9 -degree half-angle cone shows that real gas effects have significant impacts on flow fields. Two temperature model Ø Translation and rotation energy modes are in equilibrium at translation temperature (T) Ø Vibration and electronic energy modes are in equilibrium at vibration temperature (Tv). Two temperatures along the stagnation line Geometry: Cylinder Ri = 1. 00 m Freestream conditions: U 0 = 5000 m/s ρ0 = 1. 0 e-4 kg/m 3 P 0 = 5. 76 pa M 0 = 17. 6 T 0 = 200 K Tw = 500 K Re 0 = 37617. 25 CN 2 = 0. 76 CO 2 = 0. 24 CNO = CN = CO = 0. 0 Species density along the stagnation line

Effects of Graphite Ablation on Hypersonic Boundary Layer Stability Clifton Mortensen Ø Objective: Study

Effects of Graphite Ablation on Hypersonic Boundary Layer Stability Clifton Mortensen Ø Objective: Study effects of graphite ablation on the receptivity process and linear growth of Mack’s first and second modes in a high speed boundary layer. Ø Surface chemistry model including oxidation, recombination of atomic oxygen and sublimation of C 3, C 2, C is validated ØCompare the implemented surface chemistry model to other surface chemistry models in simulations performed by Chen and Milos 11 species ablation code validation to Keenan’s computations. Left: surface mass fractions. Right: surface mass flux per area Contour of CO mass fraction. CO has highest mass concentration of any carbon species. Mass blowing rate comparison between different surface chemistry models.

Direct numerical simulation of strong shock and turbulence interaction using shock-fitting methods Xiaowen Wang

Direct numerical simulation of strong shock and turbulence interaction using shock-fitting methods Xiaowen Wang and Pradeep Rawat Objectives Ø To conduct extensive DNS studies on strong shock and turbulence interactions for perfect gas flow with mean Mach numbers ranging from 2 to 30. Ø To validate our new 3 -D high-order shock-fitting code for nonequilibrium flow. ØTo conduct DNS studies on strong shock and turbulence interactions for non-equilibrium flow. Inflow : Temporal DNS with Shock Fitting Conclusion Variation of various turbulence statistics in simulation of decaying isotropic turbulence ( and ) Ø Increasing shock-strength reduces the shock deformation. Ø The trend of maximum values of streamwise vorticity fluctuations versus shock Mach number reverses at M 1 = 2. 8 from increase to decrease. Ø The trend of maximum values of Reynolds stress R 11 versus shock Mach number reverses at M 1 = 8. 8 from decrease to increase.

Numerical Study of Wave-Energy Extraction Device with Electroactive Polymer Kin Wai Leung ØCollaboration with

Numerical Study of Wave-Energy Extraction Device with Electroactive Polymer Kin Wai Leung ØCollaboration with the Soft Material Stretched: High Capacitance Research Laboratory of Prof. Pei (UCLA) to + + + + - - - - develop technology using electroactive Vb polymer to extract and convert wave energy to electricity. ØImplemented a finite volume method Relaxed: Low Capacitance ++++++++ Images provided with free-surface tracking done by Volume by Paul Brochu of the Soft Material of Fluid method to simulate the dynamic -------Research Lab interaction between energy-containing Vb + Vg incident wave and the energy extracting device. Change from high to low capacitance state raises ØIn preliminary numerical study, the potential of the charge of the film targeted strain of 20 to 30 percent was achieved, suggesting dielectric elastomer may be practical for large-scale power generation. (1) the film is mechanically stretched (2) a bias voltage is applied to the film via an external power supply (3) the power supply is removed and the film is allowed to relax (4) charge is removed from the film and it returns to its initial state Incident Wave Conditions: • First-Order Stokes Wave (Linear) • Wave Height = 0. 5 m • Wave Length = 15 m • Wave Frequency = 2. 0 Hz (calculated from dispersion relation) • Initial Water Depth = 5. 5 m Geometrical parameters: • Diameter, D = 2. 0 m • Thickness, t= 0. 125 m • Submerged depth, h = 0. 5 m A schematic of the numerical wave tank Pressure fluctuation inside the air chamber. Pressure are in Pascal; X and Y are in meters