AIAA Drag Prediction Workshop Research Articles

DPW-7: Steady and Unsteady Computations of the Common Research Model at Different Reynolds Numbers

This article presents the numerical computations performed at ONERA for the Seventh AIAA Drag Prediction Workshop. By introducing Reynolds numbers up to 30 million closer to the flight conditions, greater lift levels beyond the design point, and time-accurate simulations, this new session has allowed the previous studies to be extended. The Common Research Model aircraft configuration has been considered in its academic wing-body version and calculated in this work with point-matched structured grids. The ONERA Cassiopee software as well as the elsA solver and the FFDπ far-field drag code have been used. The grid convergence study has shown larger pressure drag variations than what was obtained at the cruise lift coefficient, but increasing the Reynolds number seems to reduce this trend. Then, the angle-of-attack sweep study with the lift, drag, and moment polars has given the opportunity to assess different numerical settings such as the Spalart–Allmaras and shear stress transport turbulence models with the quadratic constitutive relation approach (QCR-2000) and to discuss the comparison between computational fluid dynamics results and wind-tunnel data. Concerning the Reynolds number increase, it has appeared that the main part of drag reduction comes from the friction () and viscous pressure drag () components. The prediction of pitching moment increments due to Reynolds number variations still needs to be significantly improved. Finally, for an angle of attack above 4.00 deg, by the use of unsteady Reynolds-averaged Navier–Stokes computations, an unsteady buffet phenomenon has been observed and analyzed.

TAS Code, FaSTAR, and Cflow Results for the Sixth Drag Prediction Workshop

Three mixed-element unstructured Reynolds-averaged Navier–Stokes solvers, the TAS code, FaSTAR, and Cflow, were applied to the Sixth AIAA Drag Prediction Workshop test cases to investigate the accuracy of the solvers with committee-provided grids and custom participant-generated grids by MEGG3D, BOXFUN, and the Cflow built-in grid generator. Solutions were obtained for the verification study using the two-dimensional NACA 0012 airfoil (test case 1), the NASA common research model nacelle–pylon drag incremental study (test case 2), and the common research model wing/body static aeroelastic effect study (test case 3). In test case 1, the three solvers used the same turbulence model, the Spalart–Allmaras one-equation turbulence model without the trip term for transition and without the ft2 function, and their grid convergence trends for lift, drag, and pitching moment coefficients were similar to that of FUN3D, CFL3D, and TAU. In test case 2, the TAS code with the NASA GeoLab grids and with MEGG3D grids, FaSTAR with the NASA GeoLab grids, and Cflow with its own grids predicted nacelle–pylon drag increment well. In test case 3, the three solvers predicted force and moment fairly comparable with experiment if support system interference effects were considered.

DPW-6: Drag Analyses and Increments Using Different Geometries of the Common Research Model Airliner

This paper describes the CFD studies carried out at ONERA in the framework of the Sixth AIAA Drag Prediction Workshop. Different configurations of the well-known Common Research Model have been used to perform drag analyses and increment assessments. The structured overset grids provided by The Boeing Company to the DPW community have been preprocessed with the in-house software Cassiopée: CGNS conversion, cell blanking, and overlapping issues have been handled before running these grids with the RANS solver elsA and the far-field code ffd72. Contrary to the original CRM configuration used in DPW-5, the wing shape is now based on wind-tunnel aeroelastic measurements: through a grid convergence process, a four-count drag increment between new and former wing–body geometries has been quantified, which is very consistent with previous ONERA results. Then, a CRM configuration including a throughflow nacelle–pylon installation has been computed: the corresponding drag increment of about 22 counts is in very close agreement with the NASA experimental data. Finally, horizontal and vertical tails have been added to the wing–body–nacelle–pylon geometry so that the aerodynamic performance of such a complete cruise configuration can be assessed. Numerical considerations involving the SA-QCR2000 turbulence model are highlighted.

Normalization of Wind-Tunnel Data for NASA Common Research Model

A wind-tunnel test of an 80%-scale copy of the NASA Common Research Model was performed by the Japan Aerospace Exploration Agency using its transonic wind tunnel. The wind-tunnel model was fabricated by the Japan Aerospace Exploration Agency in consultation with NASA Langley Research Center and AIAA Drag Prediction Workshop committee members. The static aerodynamic forces and surface pressure distributions on the wing of the Japan Aerospace Exploration Agency model were measured at a relatively low Reynolds number of due to tunnel capability limitations, where boundary-layer transition was simulated using optimized roughness. Measured data were compared with those of wind-tunnel tests of the Common Research Model obtained from NASA Langley Research Center’s National Transonic Facility as well as computational fluid dynamics predictions, both at a Reynolds number of . The comparison among these datasets required data normalization to the designed shape aligned at a reference condition, because the wing deformation of the wind-tunnel models deformed due to the dynamic pressure and the Reynolds number was different between the Japan Aerospace Exploration Agency and the National Transonic Facility tests. In this paper, the normalization procedures, in addition to basic wind-tunnel corrections, are proposed and the corrected results are examined.

NSU3D Results for the Fourth AIAA Drag Prediction Workshop

Simulation results for the Common Research Model transport aircraft performed for the Fourth Drag Prediction Workshop using the unstructured computational fluid dynamics code NSU3D are presented. The solution algorithms employed in NSU3D for this study are described along with examples of convergence history. The Common Research Model aircraft configuration used for the workshop consists of the wing, body, and horizontal tail with no vertical tail or engine nacelles. The computational meshes for this study were provided by NASA Langley Research Center and were prepared with the VGrid package. A complete lift-drag polar is produced in NSU3D on a baseline mesh along with additional runs on refined and coarsened meshes to assess sensitivity to grid resolution. The effect of horizontal tail angle on the drag polar is investigated and the trimmed drag polar calculated. A range of Mach number values are evaluated, and the drag polar and drag rise curves are presented. The effect of Reynolds number is assessed by comparing computed results at Reynolds numbers of 5 million and 20 million.

Summary of the Fourth AIAA Computational Fluid Dynamics Drag Prediction Workshop

Results from the Fourth AIAA Drag Prediction Workshop are summarized. The workshop focused on the prediction of both absolute and differential drag levels for wing–body and wing–body/horizontal-tail configurations of the NASA Common Research Model, which is representative of transonic transport aircraft. Numerical calculations are performed using industry-relevant test cases that include lift-specific flight conditions, trimmed drag polars, downwash variations, drag rises, and Reynolds-number effects. Drag, lift, and pitching moment predictions from numerous Reynolds-averaged Navier–Stokes computational fluid dynamics methods are presented. Solutions are performed on structured, unstructured, and hybrid grid systems. The structured-grid sets include point-matched multiblock meshes and overset grid systems. The unstructured and hybrid grid sets comprise tetrahedral, pyramid, prismatic, and hexahedral elements. Effort is made to provide a high-quality and parametrically consistent family of grids for each grid type about each configuration under study. The wing–body/horizontal families comprise coarse, medium, and fine grids; an optional extrafine grid augments several of the grid families. These mesh sequences are used to determine asymptotic grid-convergence characteristics of the solution sets and to estimate grid-converged absolute drag levels of the wing–body/horizontal configuration using Richardson extrapolation.

Fifth Drag Prediction Workshop: Computational Fluid Dynamics Studies Carried Out at ONERA

This article presents the computational studies that have been performed at ONERA–The French Aerospace Lab in the context of the 5th AIAA Drag Prediction Workshop. A grid-convergence process and a buffet-onset study have been carried out using a Reynolds-averaged Navier–Stokes solver and far-field methods. In this work, the multiblock grids of the Common Research Model provided by the 5th AIAA Drag Prediction Workshop Committee are employed, and a new meshing strategy involving overset techniques and Cartesian grids is tested. All of these meshes are computed with the ONERA elsA code using the Spalart–Allmaras turbulence model and far-field drag analyses are performed with the ONERA ffd72 software. The grid-convergence study led with the common multiblock grids gives an extrapolated drag value close to 250 drag counts, which corresponds to the numerical median of all 5th AIAA Drag Prediction Workshop participants. The meshing approach using Cartesian grids also provides promising results. Comparisons with NASA wind-tunnel tests show reasonable discrepancies between computational fluid dynamics predictions and experimental drag measurements. The buffet-onset study completed with the medium common multiblock grid is aimed at investigating an angle-of-attack sweep at cruise Mach number. Global and local analyses combining near-field and far-field approaches are presented.

Application of the FUN3D Solver to the 4th AIAA Drag Prediction Workshop

FUN3D Navier–Stokes solutions were computed for the 4th AIAA Drag Prediction Workshop grid-convergence study, downwash study, and Reynolds-number study on a set of node-based mixed-element grids. All of the baseline tetrahedral grids were generated with the VGRID (developmental) advancing-layer and advancing-front grid-generation software package following the gridding guidelines developed for the workshop. With maximum grid sizes exceeding 100 million nodes, the grid-convergence study was particularly challenging for the node-based unstructured grid generators and flow solvers. At the time of the workshop, the super-fine grid with 105 million nodes and 600 million tetrahedral elements was the largest grid known to have been generated using VGRID. FUN3D Version 11.0 has a completely new pre- and postprocessing paradigm that has been incorporated directly into the solver and functions entirely in a parallel, distributed-memory environment. This feature allowed for practical preprocessing and solution times on the largest unstructured-grid size requested for the workshop. For the constant-lift grid-convergence case, the convergence of total drag is approximately second-order on the finest three grids. The variation in total drag between the finest two grids is only two counts. At the finest grid levels, only small variations in wing and tail pressure distributions are seen with grid refinement. Similarly, a small wing side-of-body separation also shows little variation at the finest grid levels. Overall, the FUN3D results compare well with the structured-grid code CFL3D. For the grid-convergence case, the FUN3D total and component forces/moments are within one standard deviation of the workshop core solution medians and are very close to the median values especially at the finest grid levels. The FUN3D downwash study and Reynolds-number study results also compare well with the range of results shown in the workshop presentations.

Fifth Drag Prediction Workshop: ONERA Investigations with Experimental Wing Twist and Laminarity

This paper is aimed at presenting the computational studies that have been performed at ONERA–The French Aerospace Lab in 2013 in the framework of the 5th AIAA Drag Prediction Workshop. As data concerning the Common Research Model configuration were collected from the NASA Langley Research Center and Ames Research Center wind tunnels, a significant discrepancy appeared between the experimental and computational pitching moment evaluations. Investigations carried out by NASA in 2012 showed that the experimental model and the numerical geometry were slightly different. Indeed, at the design point, the wing twist of the experiments was stronger than the one used for the computations. Therefore, new computational-fluid-dynamics studies using the Common Research Model wing–body configuration with the corrected twist have been completed at ONERA. The common multiblock grids provided by the Drag Prediction Workshop Committee have been adequately modified before being computed with the ONERA elsA Reynolds-averaged Navier–Stokes solver and the ONERA ffd72 software. This paper exhibits the results of the new grid-convergence process and buffet study (cases 1 and 2 of the Fifth Drag Prediction Workshop). The impacts of the wing-twist modification on drag and moment productions are investigated. The grid-convergence study led with the as-tested shape in fully turbulent conditions gives a drag value close to 254 counts, which is four counts higher than the drag produced by the original numerical geometry. Moreover, to further match the wind-tunnel conditions, computations including a laminar zone up to 10% of the wing chord have been performed. These calculations with experimental wing twist and laminar/turbulent transition significantly improve the agreement between the numerical and experimental data.

Far-Field Drag Decomposition Applied to the Drag Prediction Workshop 5 Cases

A far-field drag prediction and decomposition method has been applied to the results of the AIAA Drag Prediction Workshop 5 held in Louisiana during the summer of 2012. The method has two principal advantages: it allows the removal of spurious drag inherent to computational fluid dynamics solutions, and it allows the decomposition of drag into viscous, wave, and induced physical drag components. This research shows that accurate drag coefficients can be predicted on coarse grids when the spurious drag is extracted with the far-field method and that these results are closer to experimental values than drag coefficients computed on finer meshes when spurious drag is not extracted. The research also investigated the reasons behind the lift and drag losses found by some participants in the workshop. It is shown that the lift loss is caused by the boundary-layer separation at the wing root, inducing a reduction of 20% of the shock wave drag and a significant change in the wing loading. The initiation of buffet is also analyzed. The study shows that mesh refinement is critical to capture the physical effects of the flow, such as its separation, and provides an explanation of the discrepancies in results observed at the Drag Prediction Workshop 5.

Computations from the Fourth Drag Prediction Workshop Using the Edge Solver

Results carried out for the 4th AIAA Drag Prediction Workshop with the flow solver Edge are summarized. Simulations have been carried out for wing–body–tail configurations with three horizontal tail incidences and for one tail-off configuration. The computations consist of a grid-refinement study and a downwash study with polar calculations to explore the trimmed condition and delta effects. An investigation of the sensitivity to the grid and turbulence model is carried out from calculations on two sets of unstructured grids with three turbulence models, including the explicit algebraic Reynolds stress model, the SST model, and the Spalart–Allmaras model. The grid-refinement study, for which the flow is attached, shows rather small differences due to different grids and models. The major difference is obtained from the turbulence models where about 10 drag counts between the highest and lowest drag values are predicted, respectively, by the explicit algebraic Reynolds stress model and the Spalart–Allmaras model. The difference between the two sets of grids is only about two drag counts. The polar calculations in the downwash study also show rather small differences in which the same tendency for the drag prediction is observed as in the grid-refinement study. The largest differences between the solutions are due to different turbulence models and observed at the highest angles of attack, where the flow starts to separate after the shock on the wing.

Numerical Simulation of Flow Field around DLR_F4 Using Meshless Method

In order to investigate the calculation ability of meshless method and calculation precision for lift and drag force coefficient based on N-S equations with S-A turbulence model and Euler equations, numerical simulation of flow field around the DLR-F4 wing-body, supplied by the AIAA Drag Prediction Workshop, is employed using meshless method. Furthermore, the calculated results are compared with experiment results and Waller’s simulation results achieved by MGAERO software. It’s found that the present simulation results without consideration of fluid viscosity don’t agree well with Waller’s and experimental results. The present simulation results with consideration of fluid viscosity were found to be in good agreement with experimental results. These results indicate that meshless method coupled with S-A turbulence model could predict the natural flow characteristic around the DLR-F4 wing-body configuration well.

Anisotropic -Adaptation Framework for Functional Prediction

This paper presents a method for concurrent mesh and polynomial-order adaptation with the objective of direct minimization of output error using a selection process for choosing the optimal refinement option from a discrete set of choices that includes directional spatial resolution and approximation order increment. The scheme is geared toward compressible viscous aerodynamic flows, in which various solution features make certain refinement options more efficient than others. No attempt is made, however, to measure the solution anisotropy or smoothness directly or to incorporate it into the scheme. Rather, mesh anisotropy and approximation order distribution arise naturally from the optimization of a merit function that incorporates both an output sensitivity and a measure of solution cost on the new mesh. The method is applied to output-based adaptive simulations of laminar and Reynolds-averaged compressible Navier-Stokes equations on body-fitted meshes in two and three dimensions. Two-dimensional results show significant reductions in the degrees of freedom and computational time to achieve output convergence when discrete choice optimization is used compared to uniform or refinement. Three-dimensional results show that the presented method is an affordable way of achieving output convergence on difficult cases such as the third AIAA Drag Prediction Workshop W1 configuration.

Comparative Study of Three-Dimensional Wing Drag Minimization by Different Optimization Techniques

The main goal of this paper is to document a comparative study of different computational-fluid-dynamics-based optimization techniques applied to the solution of a three-dimensional wing drag minimization problem. To achieve this objective, three optimization tools were used: SYN107 (Intelligent Aerodynamics International), MDOPT (The Boeing Company), and OPTIMAS (Israel Aerospace Industries). The first tool employs gradient-based search techniques using the continuous adjoint equation, the second one is a response-surface method, and the last one uses a floating-point genetic algorithm as its search engine. As the starting geometry, the public domain DPW-W1 wing (a test case for the Third AIAA Drag Prediction Workshop) was used. The comparisons included herein are provided in three stages: cross analysis of the initial geometry by the computational fluid dynamics tools employed in the optimizations, optimization of the initial geometry to minimum drag, and cross analysis of optimal shapes achieved by the optimization tools using all computational fluid dynamics tools employed. The cross analysis also includes results from an independent computational fluid dynamics method that was not used in any of the optimization efforts. These results help quantify the level of variation that is inherent in, and can be expected from, application of the current state-of-the-art aerodynamic optimization methods. The present work may be regarded as a move toward the construction of reliable test cases for an aerodynamic shape optimization problem. Another goal of this collaborative investigation is to collect lessons learned from this pilot project to help develop a model for an aerodynamic optimization workshop.

Transonic drag prediction on a DLR-F6 transport configuration using unstructured grid solvers

A second international AIAA Drag Prediction Workshop (DPW-II) was organized and held in Orlando Florida on June 21–22, 2003. The primary purpose was to investigate the code-to-code uncertainty, address the sensitivity of the drag prediction to grid size and quantify the uncertainty in predicting nacelle/pylon drag increments at a transonic cruise condition. This paper presents an in-depth analysis of the DPW-II computational results from three state-of-the-art unstructured grid Navier-Stokes flow solvers exercised on similar families of tetrahedral grids. The flow solvers are USM3D – a tetrahedral cell-centered upwind solver, FUN3D – a tetrahedral node-centered upwind solver, and NSU3D – a general element node-centered central-differenced solver. Overall, grid refinement did not consistently improve the correlation with experimental data for either the wing/body or the wing/body/nacelle pylon configuration. Although, the range in total drag for the wing/body fine grids was only 5 counts, a code-to-code comparison of surface pressures and surface restricted streamlines indicated that the three solvers were not all converging to the same flow solutions– different shock locations and separation patterns were evident. Similarly, the wing/body/nacelle/pylon solutions did not appear to be converging to the same flow solutions. Although the absolute values of total drag predicted by two of the solvers for the medium and fine grids did not compare well with the experiment, the incremental drag predictions were within ±3 counts of the experimental data. Although, the sources of code-to-code variation in force and moment predictions for the three unstructured grid codes have not yet been identified, the current study reinforces the necessity of applying multiple codes to the same application to assess uncertainty.

Drag Prediction for the DLR-F6 Wing-Body Configuration Using the Edge Solver

Numerical investigations are reported on the DLR-F6 wing-body configuration with and without fairing. The configurations have been adopted as test cases for the Third AIAA Drag Prediction Workshop. The addition of the fairing is to eliminate the flow separation bubble in the junction between the wing trailing edge and the fuselage. The computations have been carried out using two groups of unstructured grids with different sizes. In addition to the effect of incidences, studies of grid convergence have also been performed. The computational fluid dynamics solver Edge is used for the investigation. The calculations confirm that the flow separation can be removed in the wing-fuselage junction with the fairing. For this configuration, the results obtained with the two groups of grid are very similar. Without fairing, however, one group of grids has pronounced a lower lift and produced a more extended trailing-edge separation. Because no experimental data are available for the flow condition as swerved in Drag Prediction Workshop-3, additional calculations have been carried out with the clean wing-body configuration at the Drag Prediction Workshop-2 Reynolds number to validate the numerical results against available experimental data. Very good agreement is obtained, in particular, for the global forces and moments. The calculation indicates that, as compared with experimental data, the grid which predicts a relatively large separation region provides improved predictions.

Third Drag Prediction Workshop Results Using the NSU3D Unstructured Mesh Solver

Results from the third AIAA drag prediction workshop on using the unstructured mesh Reynolds-averaged Navier-Stokes solver NSU3D are presented. Computations include a grid convergence study on a transonic wing-body and wing-body-fairing configuration at a fixed C L condition using grids up to 41 million points, as well as an incidence sweep (drag polar) at fixed Mach and Reynolds numbers. A second set of results on a pair of closely related wing geometries is also described, including a grid convergence study at a fixed incidence and an incidence sweep (drag polar) for both wing geometries. For all cases, approximate second-order-accurate grid convergence characteristics are demonstrated, with overall accuracy and efficiency comparable with other structured, overset, and unstructured workshop calculations. However, it is found that differing grid-converged results may be inferred based on different families of self-similar coarse and fine-grid sequences, particularly for values such as absolute drag at a fixed incidence. More consistent grid convergence for idealized drag values (omitting induced drag) is observed, thus validating the procedure of performing grid convergence studies at a fixed C L . These grid convergence issues are attributed to the large range of disparate scales that must be resolved in aerodynamic flows, and they point to the need for further advances in quantifying and resolving discretization errors for such problems.

Abridged Summary of the Third AIAA Computational Fluid Dynamics Drag Prediction Workshop

Results from the Third AIAA Drag Prediction Workshop (DPW-III) are summarized. The workshop focused on the prediction of both absolute and differential drag levels for wing-body and wing-alone configurations that are representative of transonic transport aircraft The baseline DLR-F6 wing-body geometry, previously used in DPW-II, is also augmented with a side-of-body fairing to help reduce the complexity of the flow physics in the wing-body juncture region. In addition, two new wing-alone geometries have been developed for DPW-III. Numerical calculations are performed using industry-relevant test cases that include lift-specific and fixed-alpha flight conditions, as well as full drag polars. Drag, lift, and pitching-moment predictions from numerous Reynolds-averaged Navier-Stokes computational fluid dynamics methods are presented, focused on fully turbulent flows. Solutions are performed on structured, unstructured, and hybrid grid systems. The structured grid sets include point-matched multiblock meshes and overset grid systems. The unstructured and hybrid grid sets are composed of tetrahedral, pyramid, and prismatic elements. Effort was made to provide a high-quality and parametrically consistent family of grids for each grid type about each configuration under study. The wing-body families are composed of a coarse, medium, and fine grid, whereas the wing-alone families also include an extra-fine mesh. These mesh sequences are used to help determine how the provided flow solutions fare with respect to asymptotic grid convergence, and are used to estimate an absolute drag for each configuration.

Drag Prediction of Engine-Airframe Interference Effects with CFX-5

The commercial computational fluid dynamics (CFD) code CFX-5 of ANSYS, Inc., has been used to compute the engine installation drag for the German Aerospace Center (DLR) F6 aircraft configuration as part of the Second AIAA Drag Prediction Workshop. The computations were performed with the standard hexahedral meshes provided by ICEM.CFD to the workshop. The full drag polar for the DLR-F6 configuration has been computed. For all cases, good agreement between the experiment and the predictions were obtained for lift, drag, and pitching moment coefficients. All simulations where based on the shear stress transport turbulence model, and additional computations have indicated that turbulence modeling issues are largely responsible for the overprediction of the lift curve slope that was observed by many of the workshop participants.

Data Summary from Second AIAA Computational Fluid Dynamics Drag Prediction Workshop

Results from the Second AIAA Drag PredictionWorkshop are summarized. The workshop focused on absolute and configuration delta drag prediction of the DLR, German Aerospace Research Center F6 geometry, which is representative of transport aircraft designed for transonic flight. Both wing–body and wing–body–nacelle–pylon configurations are considered. Comparisons are made using industry relevant test cases that include single-point conditions, drag polars, and drag-rise curves. Drag, lift, and pitching moment predictions from several different Reynolds averagedNavier–Stokes computational fluid dynamics codes are presented and compared to experimental data. Solutions on multiblock structured, unstructured, and overset structured grids using a variety of turbulence models are considered. Results of a grid-refinement study and a comparison of tripped transition vs fully turbulent boundary-layer computations are reported.