Software for thermal and flow fields analysis in the supersonic/hypersonic boundary layers.
Verifications and uncertainty qualification.
ANNOUNCEMENT OF OPPORTUNITY AO/1-7169/12/NL/CBi
1st CALL FOR OUTLINE PROPOSALS UNDER THE ROMANIAN INDUSTRY INCENTIVE SCHEME
ESTEC Contract no. 4000109853/13/NL/SC
Software for thermal and flow fields analysis in the supersonic/hypersonic boundary layers. Verifications and uncertainty qualification.
The main objectives of the present contract refers to the development of the engineering verified software package for thermal and flow fields calculations in supersonic/hypersonic laminar and turbulent boundary layers.
Advances in computer simulation have lead researchers to interpret predictions of increasingly more complicated physical phenomena. However, the complexity of recent simulations makes their reliability difficult to assess and one faces the danger of drawing false conclusions from inaccurate predictions. The present proposal is dedicated to the development of methods and techniques for reliable computational predictions through the verification of numerical computations, the validation of the underlying physical models. We have chosen reentry vehicles as the overarching target application for our efforts. A vehicle entering the atmosphere from space experiences an extreme thermal environment due to the very large velocities involved. The extremely high kinetic energy of the on-coming air (up to about 100MW/m2) is suddenly converted into thermal energy as the air passes through the bow shock. This energy conversion raises the temperature of the air to as high as 20,000K, resulting in dissociation and possibly ionization of the air molecules. When this super-heated air flows over the vehicle, it results in very high heat transfer to the surface, requiring a thermal protection system to ensure the survival of the vehicle. To predict flow and thermal field quantities, models of hypersonic flow, aerothermochemistry, thermal radiation and turbulence are needed.
A space capsule reentering the atmosphere, with relevant physical phenomena
The development of methods and computational tools for prediction, validation and uncertainty quantification is best pursued in the context of a complex multi-physics application, since this ensures that many of the challenges inherent in real applications will be addressed. Furthermore we have an opportunity to make meaningful advances in engineering by pursuing the application of validation and eventually, uncertainty quantification.
The mathematical, numerical and experimental investigation of boundary layer flows in supersonic/hypersonic regimes is still a challenging and open problem. The present trend in this kind of engineering calculations is the use of sophisticated physical-chemical models and numerical methods implemented in complex both commercial and in-house codes. In our opinion, the current level of maturity of the technology (TLR) concerning existing numerical codes for supersonic/hypersonic flows is at least TLR6.
Nevertheless, usually, the quantitative and, sometimes the qualitative data obtained from numerical predictions are difficult to provide objective confidence level. In addition, the use of these programs involves considerable computational resources with significant costs. We can indentify two types of major sources of uncertainty in practical computations of supersonic/hypersonic flows: physical/mathematical model and numerical techniques. Concerning the physical/mathematical model uncertainty, the main aspects refer to: modeling the continuum-discrete transition (presently the fluid is assumed either continuum or discrete on sub-domains of the flow field depending on the local Knudsen number, neglecting a possible smooth transition region), the source terms in governing equations (chemical mass source in species concentration transport equation, radiation terms in energy equation, level of thermodynamic non-equilibrium, vibrational relaxation models for ionized species etc), laminar-turbulent transition modeling, turbulence modeling, chemistry modeling. From numerical point of view the key aspects involved in the accuracy of the numerical data are: discretization order (high order schemes up to six order are commonly use in in-house codes), shock capturing techniques and the boundary/initial conditions implemented (e.g. for DNS solvers the problem of generating of initial conditions is critical and open one). Consequently, we can appreciate the current level of maturity of validated and uncertainty qualified numerical codes for supersonic/hypersonic flows below TLR2.
Because we propose research and development activities which lead to a deeper validated software modules, we appreciate to improve the maturity level from TLR 1 to TLR 3.
The package will contain a set of computer codes dedicated to laminar or turbulent boundary layer calculations in supersonic / hypersonic regimes, each module based on a specific mathematical model. Models covered will correspond to a hierarchy: from the “simple” to the “complex”. The modules will be interfaced, each level approximation is designed to provide, in addition to a series of reference or ranges for the physical quantities involved, the boundary and initial conditions for the higher level. The software package will contain the following modules (interfaced computer programs):
- Code (CD_level_0) for inviscid fluid in supersonic/hypersonic 2D (plane and axisymmetric) and 3D flows (Euler equations);
- Code (CD_level_1.1) supersonic/hypersonic boundary layer; laminar/turbulent flow; RANS models, constant turbulent Prandtl number, no chemistry;
- Code (CD_level_1.2) supersonic/hypersonic boundary layer; laminar/turbulent flow; RANS models, variable turbulent Prandtl number, no chemistry);
- Code (CD_level_1.3) supersonic/hypersonic boundary layer; chemical reacting radiative flow; laminar/turbulent flow; RANS models, constant and variable turbulent Prandtl number;
- Code (CD_level_2) for supersonic/hypersonic boundary layer; chemical reacting radiative flow; laminar/turbulent flow; LES models
In order to verify and qualify the numerical code (and each module separately) we will also impose the following requirements:
- Validation procedures using Method of Manufactured Solutions (MMS);
- Sensitivity Analysis (SA);
- Uncertainty qualification (UQ);
- Experimental campaign in the trisonic wind tunnel of INCAS. Measurements of static pressure distribution and boundary layer velocity profile (micro traversing anemometer) for the first canonical case – dihedral flat plate; Schlieren and oil-paint flow visualisation;
- The second case approaches a complex flow field configuration, having a re-entry vehicle similar to IXV; Static (steady and unsteady) pressure measurement in a small number of key-points; Thermal and Schlieren visualisation; Oil-paint flow visualisation.
The present contract will offer an efficient and rapid collection of engineering tools to predict the various flow and thermal parameters at an imposed level of confidence for supersonic/hypersonic boundary layers. The main advantages of this approach are:
- High computational efficiency with an increase compared with standard CFD formulations;
- High confidence since verification via MMS and experimental data;
- Versatility (fast response to new input data, fast production and implementation of physically correct initial and boundary conditions);
- New key parameter definition (variable turbulent Prandtl number);
- Simplicity due to the use a cascade of simpler and easy to control codes;
- Cost reduction.
Application area: Design of hypersonic vehicles, reentry, thermal protection (e.g. PRIDE program), uncertainty qualification of CFD codes.
The design of hypersonic vehicles requires accurate prediction of the surface properties. These quantities are typically the heat flux, pressure and shear stress. During its trajectory through an atmosphere, a hypersonic vehicle will experience vastly different flow regimes due to the variation of atmospheric density with altitude. In addition, the high temperatures encountered due to the high velocities cause dissociation and ionization of the atmospheric gases. Reproduction of these varied flow conditions in ground-based laboratory facilities is both expensive and technically challenging. Hence, there is an extremely important role for computational models in the development of hypersonic vehicles.
Prime contractor: University POLITEHNICA of Bucharest – Research Center for Aeronautics and Space
- Prof. Sterian DANAILA, Ph.D., MEng.
- Prof. Corneliu BERBENTE, Ph.D,., MEng.
- Assoc. Prof. Marius STOIA-DJESKA, Ph.D., MEng.
- Assoc. Prof. Dragos ISVORANU, Ph.D., MEng., MSc.
- Assoc. Prof. Teodor-Viorel CHELARU, Ph.D., MEng.
- Assoc. Prof. Laurentiu MORARU, Ph.D., MEng.
- Lecturer Brebenel Marius, Ph.D., MEng.
- Lecturer Alina BOGOI, Ph.D., MEng.
Sub-Contractor: National Institute for Aerospace Research “Elie Carafoli” INCAS
- Catalin Nae, Ph.D
- Mihai Victor Pricop, MEng.
- Adrian CHELARU, MEng.
- Cornel STOICA, MEng.
Restricted access: Click here.
University POLITEHNICA of Bucharest – Research Center for Aeronautics and Space
Address: No. 1, Gh. Polizu Street, Bucharest, CP 0011061 Romania
Technical Officer: Sterian DANAILA, phone: (+40) 21 402 39 67, fax: (+40)214023967, email: email@example.com
Contractual Officer: Marius BREBENEL, phone: (+40) 21 402 39 67, fax: (+40) 21 318 100, e-mail: firstname.lastname@example.org
Advanced Solutions for Modeling the Laminar-Turbulent Transition
Analysis Studies of the Laminar Movement Stability
Justification: As mentioned above, if the solution of the NS system is perturbed, equations may be deducted (Stability Equations), which describe the time and space evolution of the perturbations. The most complex stability models currently are those in the PSE category (”Parabolic Stability Equation”). The basic simplification hypothesis consists of admitting the perturbation as a quasi-periodic phenomenon, shaped as a rapid variation wave in space and time and with a slow variation amplitude. This allows obtaining a differential model for amplitude by applying a Fourier transform. Moreover, a carefully thought choice of the waveform and the shape function (the amplitude) yield a parabolic model, the numerical solution moves from downstream to upstream with a low computing effort as compared to DNS/LES. The solution accuracy is remarkable because the laminar instability (some mentioned above) satisfy the PSE model hypotheses, especially in its non-linear variant.
Analysis and Implementation of the DNS and LES Methods.
Justification: The only mathematical model capable to capture the laminar instability and the formation of turbulent structures without any simplifying hypothesis is the Navier-Stokes model. That is why we set our target to develop a numerical method to directly integrate the Navier-Stokes equations (DNS) and LES respectively.
Importance to the field: Although attempted has been made internationally for over a decade in this line of research (see 10.1 presentation), to the best of our knowledge, there is nothing similar in the country. However, the experience gathered by the team of this proposal in the theoretical and applied CFD development, demonstrated by papers and books published nationally and internationally, is a basis for such an accomplishment.
Development of Turbulence Models applicable in Transition.
Justification: Both in the problem of turbulent boundary layer, and in general in RANS models, there is the closure problem (by a turbulence model). More recently, a series of non-linear turbulence models aim at a more accurate simulation of the effects of transition to replace the traditional way based on correlations and on the intermittent factor concept. For this reason, we consider significant to include as a project objective the statistical approach of the turbulence description and of the transition (RANS).
Importance to the field: The above mentioned deficiency generates very large errors in determining of aerodynamic characteristics of aerodynamic shaped bodies, because in such situations there is progressive transition from laminary to transitory and then to turbulent. Usually, the turbulence models used in engineering practice (and implemented in the CFD software codes) such as k-epsilon, k-omega, SST models etc. are well adapted for developed turbulence, but lack capacity to capture transient regime correctly.
Impact: Turbulence Models will be implemented in the form of UDFs (User Defined Functions) working in the commercial CFD software Fluent.
- Prof. dr. ing. Sterian DANAILA – Director
- Prof. dr. ing. Corneliu BERBENTE
- Conf. dr. ing. Marius STOIA-DJESKA
- As. ing. Constantin LENEVTIU
- As. ing. Claudiu VADEAN – out of the project in 2010
- Drd. ing. Mihai NICULESCU