Validation of Unsteady Aerodynamic Models of a Generic UCAV Using Overset Grids

Abstract

This work is part of the NATO STO Task Group AVT-201 Special Sessions being organized by the AIAA Applied Aerodynamics Technical Committee. Reduced-OrderModeling techniques based on response functions are presented and validated for unsteady aerodynamic prediction of a generic unmanned combat air vehicle with trailing edge flaps. The overset grid approach was used to move these control surfaces and simulate the flow fields around the vehicle in static and dynamic conditions and with different control surface deflections. Single meshes were also generated to study the effects of gaps present in the overset mesh and the sting geometry available in the experimental model. The Reynoldsaveraged Navier Stokes predictions were first validated by wind-tunnel static and dynamic case measurements. The flow topology over the configuration was then described for a range of angles of attack and control surface deflections. The response functions with respect to angle of attack and pitch rate were directly calculated from step motion simulations in CFD and a grid motion approach. An aerodynamic model was then created from these functions and used to predict the vehicle aerodynamic responses obtained from pitching oscillations at different amplitudes and frequencies. The results show that predictions from a linear model are only in good agreement with time-accurate simulations of small amplitude motions. A model that has response dependence on the angle of attack match quite well with CFD data of large amplitude motions within the range of data used for model generation, however, small discrepancies can also be seen in pitch moment plots. The results show that model predictions get worsen with including sting geometry in simulations because of the nonlinear hysteretic behavior of a wake formed behind the sting. Nomenclatureroll moment coefficient in wind axes Downloaded by CORNELL UNIVERSITY on July 30, 2015 | http://arc.aiaa.org |

Comparison of Numerical Reduced Order Models of a Generic UCAV Configuration using a New Displacement Grid Method

“…The simulations are performed at 𝑀 = 0.15 and 𝑀 = 0.3. The results obtained with the FMG method are in agreement with Ghoreyshi et al [23,28]. Indeed, the height of the initial peak, the time of the transient regime as well as the asymptotic value reached in the stationary regime are almost identical (Fig.…”

“…In general, in the case of longitudinal aerodynamics, the AOA indicial response is assumed to vary with the AOA and the Mach number whereas the pitch rate indicial response is assumed to vary only with the Mach number for moderate AOAs. However, the accuracy of the model could be improved by including in the database the coupling between the rotation rate and the AOA [14,28]. Besides, indicial responses obtained from a negative unit step variation could be beneőcial to improve the model [14].…”

“…8b and 9b) and for each of them, the height of the peak, the duration of the transient regime and őnal asymptotic value are the same between both methods for the lift and the pitch coefficients. The indicial responses were then compared with the literature and in particular with the simulations performed on the SACCON geometry by Ghoreyshi et al [23,28]. The simulations are performed at 𝑀 = 0.15 and 𝑀 = 0.3.…”

The NATO STO Task Group AVT-201 on ‘Extended Assessment of Stability and Control Prediction Methods for NATO Air Vehicles’

“…15, where the SACCON configuration is flying a half lazy eight maneuver which has been prescribed (see Ref. 18 for details). In this case a series of step changes in angle of attack and sideslip were run using CFD and a model was created using the response to the step changes.…”