Drag reduction of a wing-body configuration via spanwise forcing Andrea Gadda, Jacopo Banchetti, Giulio Romanelli, Maurizio Quadrio Dipartimento di Scienze e Tecnologie Aerospaziali Politecnico di Milano / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Motivation Effectiveness of skin-friction drag reduction techniques in aeronautical applications Drag components of Transport Aircraft in Cruise: Mele et al. , J. of Aircraft, 2016 / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Quadrio, Ricco & Viotti, JFM 09 Potential for large energy savings Low-Re, incompressible flows in simple geometry. [QRV09] What about an airplane? / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Background (Gatti & Quadrio, JFM 16) Waves can be assimilated to drag-reducing roughness Streamwise travelling waves produce a vertical shift ∆ B of the logarithmic portion of the mean velocity profile Drag reduction rate R is linked to ∆ B ∆ B + at non-low Re becomes Reynolds independent U + = 1 κ log ( y + ) + B + ∆ B + / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Case of study AIAA Second Drag Prediction Workshop (DLR-F6) DLR-F6 is a modern transport aircraft, with a transonic design Wing-body configuration RANS Spalart-Allmaras Turbulence model Fully turbulent boundary layer Re = 3 · 10 6 based on reference chord M = 0 . 75 Flight lift coefficient 0.5 / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Present work Forcing is introduced by a modified wall function U + = 1 κ log ( y + ) + B + ∆ B + Coarse mesh available in Drag Prediction Workshop website Forcing applied over the entire aircraft / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
AeroX A GPU-CPU compressible RANS solver Finite volumes Compressible (transonic) Speedup by GPU: AMD 380X FURY X (2015) ∼ 230 USD ∼ 650 USD i7 5930k-6 4 . 3 x 8 . 7 x ∼ 600 USD In the present work: GPU: AMD 380X 2 · 10 6 elements: convergence in ∼ 45 min / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Validation DLR-F6 Polar curve 0 . 8 0 . 6 CL 0 . 4 Experimental Cessna NSU3D 0 . 2 AeroX 0 . 01 0 . 02 0 . 03 0 . 04 0 . 05 CD / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Drag Reduction Friction and Pressure drag decomposition / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Drag Reduction Friction and Pressure drag decomposition Expected Friction drag reduces by 23 % Pressure drag is unchanged Total drag reduces by at most 14 % 30 20 10 DR % 0 Expected − 10 − 4 − 2 0 2 α [ deg ] / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Drag Reduction Friction and Pressure drag decomposition Expected Actual Friction drag reduces by 23 % Pressure drag is unchanged Total drag reduces by at most 14 % 22 % 30 20 10 DR % 0 Expected − 10 Actual − 4 − 2 0 2 α [ deg ] / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Drag Reduction Friction and Pressure drag decomposition Expected Actual Friction drag reduces by 23 % ∼ 23 % Pressure drag is unchanged changed Total drag reduces by at most 14 % 22 % 30 20 10 DR % 0 Friction − 10 Pressure − 4 − 2 0 2 α [ deg ] / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Local Friction Reduction Upper view Local Friction Reduction close to 23 % where the configuration is clean Strong variations on the upper wing surface / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Local Friction Reduction Frontal view Local Friction Reduction unexpectedly high in the front fuselage / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Local Pressure Upper view Fuselage and lower wing surface unchanged Changes on upper wing surface / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Local Pressure Pressure coefficient distribution Secondary effect: Shock delay 0 . 6 Reference Controlled 0 . 4 0 . 2 − CP 0 − 0 . 2 − 0 . 4 0 0 . 2 0 . 4 0 . 6 0 . 8 1 x / c / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Lift Coefficient CL − α curve Secondary effect: Lift increase 1 Reference 0 . 8 Controlled 0 . 6 0 . 4 CL 0 . 2 0 − 0 . 2 − 0 . 4 − 4 − 2 0 2 α [ deg ] / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Drag reduction at constant lift 30 Expected 20 DR % 10 0 0 0 . 2 0 . 4 0 . 6 0 . 8 CL / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Drag reduction at constant lift Drag reduction is always higher than expected Lift increase - Stall begins at higher CL 30 Expected Actual 20 DR % 10 0 0 0 . 2 0 . 4 0 . 6 0 . 8 CL / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Net power saving at constant lift Active techniques require input power Input power is estimated via known trends in channel flow / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Net power saving at constant lift Active techniques require input power Input power is estimated via known trends in channel flow 30 Expected 20 S % 10 0 0 0 . 2 0 . 4 0 . 6 0 . 8 CL / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Net power saving at constant lift Net power saving S higher than 10 % 30 Expected Actual 20 S % 10 0 0 0 . 2 0 . 4 0 . 6 0 . 8 CL / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Comparison with MTC 2016 Despite the differences MTC16 Actual Solver UZEN / FLOWer AeroX Aircraft CRM DLR-F6 5 · 10 6 3 · 10 6 Re M 0 . 85 0 . 75 Turbulence model SST Spalart-Allmaras DR technique Riblets Spanwise forcing Forcing formulation ω at wall Wall function Same qualitative results: Direct effects: R close to the expected value Indirect effects: Shock delay - Lift increase / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Doubts? Do we trust these results? Further investigations needed: Transition? Log Law? / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Doubts? Do we trust these results? Further investigations needed: Transition? Log Law? "One coincidence is just a coincidence. Two coincidences are a clue. Three coincidences are a proof." (A. Christie) / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Thank you for your attention Questions? / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
/ Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
S ( A + ) / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
How does S changes with forcing amplitude? S at low-Re incompressible channel flow rapidly decreases after ∼ A + = 7 25 15 5 S − 5 − 15 0 5 10 15 20 A + / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
How does S changes with forcing amplitude? Shock delay increase with A + 0 . 6 A + − CP 0 . 4 0 . 2 0 . 2 x / c / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
How does S changes with forcing amplitude? S at low-Re incompressible channel flow rapidly decreases after ∼ A + = 7 15 S 5 0 5 10 15 20 A + Actual S slightly increases until A + = 12 / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
∆ LFR & −∇ p ? / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Local Friction Reduction Frontal view Local Friction Reduction unexpectedly high in the front fuselage / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Local Friction Reduction Pressure gradient and Local Friction Reduction ( LFR − LFR expected ) ∝ −∇ p ? / Jacopo Banchetti Drag reduction of a wing-body configuration via spanwise forcing 25
Recommend
More recommend