Select ADO-DG Case Studies in Aerodynamic Design Optimization

Select ADO-DG Case Studies in Aerodynamic Design Optimization Antony Jameson T. V. Jones Professor of Engineering Dept. Aeronautics & Astronautics Stanford University Stanford, CA 94305-3030, USA John C. Vassberg Boeing Technical Fellow Advanced Concepts Design Center Boeing Commercial Airplanes Long Beach, CA 90846, USA Invited Presentation AIAA Sci-Tech Conference Kissimmee, FL 5 January, 2015 Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 1

OUTLINE NACA0012-ADO INVISCID AIRFOIL ADO-DG Case 1 SYN83 Optimizations FLO82 Cross Analyses GSA Study Case Summary ADO-CRM-WING ADO-DG Cases 4.1-4.3 SYN107 Optimizations Case Summary Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 2

NACA0012-ADO MODEL PROBLEM Optimization Statement Minimize Drag M = 0.85, α = 0, Inviscid Flow Maintain or Exceed Thickness Distribution of Baseline NACA0012-ADO Airfoil Equation Closed Trailing-Edge at x = 1 y A (x) = ± 0.12 0.2 ( 0.2969 x 0.1260x 0.3516x 2 0.2843x 3 0.1036x 4) Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 3

ADO-DG CASE 1: NACA0012-ADO Optimization Methods SYN83 GSA Design Space Survey FLO82 Cross-Analysis Vassberg & Jameson, In Pursuit of Grid Convergence for Two-Dimensional Euler Solutions, AIAA Journal of Aircraft, Vol.47, No.4, pp.1152-1166, July-August, 2010. High-Quality O-Mesh with Aspect-Ratio-1 Cells Family of Meshes (32x32)-to-(2048x2048) Cells Converge Residuals to Machine-Level Zero Richardson Extrapolation Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 4

BASELINE NACA0012-ADO SOLUTION -2.0-1.6-1.2-0.8-0.4 0.4 0.8 1.2 NACA0012-ADO - JCV AR=1 GRID MACH 0.850 ALPHA 000000 CL 00000000 CD 47122638 CM 00000000 GRID 2048x 2048 NCYC 500 RED -10.79 CP MACH: MIN 000604 MAX 1.3691850 CONTOURS 50 FLO82 Solution for NACA0012-ADO Airfoil at M = 0.85, α = 0. Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 5

SYN83 OPTIMIZATION SYN83 C-mesh (768x128) about NACA0012-ADO. Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 6

SYN83 OPTIMIZATION SYN83 Results (C d in counts). Airfoil C d C d Seed NACA0012-ADO 456.34 - Design SYNA 103.71-352.63 Seed SEEDB 101.79-354.55 Design SYNB 79.31-377.03 Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 7

SYN83 OPTIMIZATION -2.0-1.6-1.2-0.8-0.4 0.4 0.8 1.2.. NADOV01 OPTIMUM AIRFOIL MACH 0.850 ALPHA 000000 CL 00000000 CD 11834217 CM 00000000 GRID 2048x 2048 NCYC 500 CP MACH: MIN 000631 MAX 1.6034683 CONTOURS 60 FLO82 Solution for SYNA Airfoil at M = 0.85, α = 0. Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 8

SYN83 OPTIMIZATION -2.0-1.6-1.2-0.8-0.4 0.4 0.8 1.2.. NADOV02 OPTIMUM AIRFOIL MACH 0.850 ALPHA 000000 CL 00000000 CD 08449573 CM 00000000 GRID 2048x 2048 NCYC 500 CP MACH: MIN 000566 MAX 1.5738200 CONTOURS 60 FLO82 Solution for SYNB Airfoil at M = 0.85, α = 0. Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 9

FLO82 O-MESH Close-up view of the NACA0012-ADO 256x256 O-mesh. Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 11

(256 x 256) (512 x 512) 1.4 1.4-2.0 1.2-2.0 1.2-4.0 S S 1.0-4.0 S S 1.0-6.0 0.8-6.0 0.8 Log(Error) -8.0 R 0.6 Nsup Log(Error) -8.0 R 0.6 Nsup -1 0.4-1 0.4-12.0 D R 0.2-12.0 R 0.2-14.0-14.0 D L L L L D D -16.0-0.2-16.0-0.2 10 20 30 40 50 60 10 20 30 40 50 60 Work FLO82 Convergence Histories at M = 0.85, α = 0. Work Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 12

(1024 x 1024) (2048 x 2048) 1.4 1.4-2.0 1.2-2.0 1.2-4.0 S S 1.0-4.0 S S 1.0 R -6.0 0.8-6.0 0.8 Log(Error) -8.0 R 0.6 Nsup Log(Error) -8.0 0.6 Nsup -1 0.4-1 0.4 D -12.0 R 0.2-12.0 R 0.2-14.0 D -14.0 L L L L D D -16.0-0.2-16.0-0.2 10 20 30 40 50 60 10 20 30 40 50 60 Work FLO82 Convergence Histories at M = 0.85, α = 0. Work Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 13

FLO82 CROSS-ANALYSIS FLO82 Drag Assessment (C d in counts). Airfoil N256 N512 N1024 N2048 C d NACA0012-ADO 470.19 479 471.13 471.23 471.27 - SYNA 153.78 123.24 119.03 118.34 118.21-353.06 SEEDB 111.04 98.98 97.68 96.85 95.42-375.85 SYNB 109.25 86.87 84.68 84.50 84.48-386.79 NADOT101 487.50 488.20 488.48 488.58 488.62 17.35 SYNBT101 122.22 99.20 96.75 96.64 96.63-374.64 Note: 1% Increase in Thickness Increases Drag: NACA0012-ADO Airfoil by 3.68% SYNB Design by 14.38%. Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 14

CARRIER-DESTERAC AIRFOIL (G) -2.0-1.6-1.2-0.8-0.4 0.4 0.8 1.2.. CARRIER-DESTARAC BEZ96FP OPTIMUM AIRFOIL MACH 0.850 ALPHA 000000 CL 00000000 CD 03589230 CM 00000000 GRID 2048x 2048 NCYC 500 RED -3.70 XCP 00000000 YCP 00000000 CP MACH: MIN 000687 MAX 1.6421540 CONTOURS 60 FLO82 Solution, Carrier-Desterac Airfoil (G), M = 0.85, α = 0. Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 15

BISSON-NADARAJAH AIRFOIL (S) -2.0-1.6-1.2-0.8-0.4 0.4 0.8 1.2. SIVAFOIL OPTIMUM AIRFOIL MACH 0.850 ALPHA 000000 CL 00000000 CD 04655680 CM 00000000 GRID 2048x 2048 NCYC 500 RED -4.65 CP MACH: MIN 001408 MAX 1.7579063 CONTOURS 70 FLO82 Solution, Bisson-Nadarajah Airfoil (S), M = 0.85, α = 0. Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 16

CARY AIRFOIL (A) -2.0-1.6-1.2-0.8-0.4 0.4 0.8 1.2.. ANDREW CARY OPTIMUM AIRFOIL MACH 0.850 ALPHA 000000 CL 00000000 CD 04014639 CM 00000000 GRID 2048x 2048 NCYC 500 RED -5.99 CP MACH: MIN 000373 MAX 1.6728016 CONTOURS 60 FLO82 Solution, Cary Airfoil (A), M = 0.85, α = 0. Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 17

GSA STUDY GSA Sub-Space Spanned By 3 Airfoils Gerald s, Siva s & Andrew s Optimimum Airfoils Geometric RMS Distances Between GSA Airfoils GS: 258890 GA: 328760 SA: 096097 GSA Triangle Interior Angles G: 12.9886 S: 129.7516 A: 37.2598 Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 18

A-Airfoil S-Airfoil G-Airfoil GSA Triangle & FLO82 Survey Locations. Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 19

GSA FLO82 Results, M = 0.85, α = 0. G WT S WT A WT C d X TRI Y TRI A 0 0 1.00 39.189621 3203 0739 0 0.25 0.75 37.0 3050 0554 0 0 0 37.0 2896 0369 0 0.75 0.25 431373 2743 0185 S 0 1.00 0 45.656136 2589 0000 0.25 0 0.75 43.724493 2403 0554 0.25 0.25 0 39.300977 2249 0369 0.25 0 0.25 40.930957 2095 0185 0.25 0.75 0 44.367019 1942 0000 0 0 0 44.688405 1602 0369 0 0.25 0.25 41.051787 1448 0185 0 0 0 42.651239 1294 0000 0.75 0 0.25 39.288289 0801 0185 0.75 0.25 0 37.520335 0647 0000 0.90 0 0.10 34.978658 0320 0074 * 0.90 0.10 0 33.911086 0259 0000 0.95 0 5 34.777423 0160 0037 0.95 5 0 34.142484 0130 0000 G 1.00 0 0 35.818186 0000 0000 Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 20

GSA FLO82 Solutions, Iso-A Contours, M = 0.85, α = 0. Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 21

GSA FLO82 Solutions, Iso-S Contours, M = 0.85, α = 0. Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 22

FLO82 Solutions, M = 0.85, α = 0. Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 23

FLO82 Solutions, Level View of Valley, M = 0.85, α = 0. Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 24

FLO82 Solutions, Elevated View of Valley, M = 0.85, α = 0. Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 25

ADO-DG CASE-1 SUMMARY Optimizations Yield Pathological Designs Small Geometric Changes Large Solution Deltas Non-Unique Solutions at Design Point Lifting (±) Solutions for Symmetric Airfoils at α = 0 Convergence Issues with Flow & Adjoint Solvers Forcing Symmetric Solutions Helps Shock-Free Design Still Not Found The GSA Airfoils Are Close Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 26

ADO-CRM-WING MODEL PROBLEM ADO-DG Case 4 General Statement Minimize Drag at Fixed Lifting Condition(s) Re = 5.0 10 6 C M 0.17 at C L = 0 & M = 0.85 Maintain or Exceed Internal Volume of Baseline Wing Case 4.1 Single-Point M = 0.85, C L = 0 Case 4.2 Triple-Point C L Sweep M = 0.85, C L = [0.45, 0, 5], WT = [1, 2, 1] Case 4.3 Triple-Point Mach Sweep M = [0.84, 0.85, 0.86], C L = 0, WT = [1, 2, 1] Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 27

ADO-CRM-WING MODEL PROBLEM ADO-CRM-Wing Reference Quantities Sref/2 = 3.407014, Cref = 1.0, b/2 = 3.75820 Xref = 1.2077, Y ref =, Zref = 07669 SYN107 Optimizations Constrained Lifting Condition(s) Active Adjustment of α During Convergence Includes dc D /dc L Terms in Gradient Formulation Added Pitching-Moment Penalty Sufficient to Achieve C M Constraint Constrained Airfoil Area Distribution Over-Constraint on Wing Volume More Representative of Practice Omits Eddy Viscousity Derivatives Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 28

SYN107 OPTIMIZATIONS SYN107 Optimization Require 2.3 Hours Per Design Point, Per 100 Design Cycles 2046 Design Variables; (2x31x33) Cubic B-Spline CPs Initial & Every 10 th Design Cycle 240 Iters for Analysis 160 Iters for dc D /dc L, dc M /dc L & dc L /dα 240 Iters for Adjoint Otherwise 20 Iters for Analysis & Adjoint SYN107 Analyses Require 4.2 Minutes Per Flow Condition (200 Iters) Run as Alpha or Mach Sweeps Grid: (256x64x48) C-Mesh Parallel Execution on 4 Cores Deskside Computer Intel i7-970 CPU at 3.2 GHz; 2011-Q1 Current Computers 5X Faster Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 29

ADO-DG CASE 4: BASELINE WING Design Pt.: M = 0.85, C L = 0, Re = 5.0 10 6 Polar: C L = [0.45, 0, 5] DragRise: M = [0.84, 0.85, 0.86] Comparisons Overlaid Pressure Distributions Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 30

COMPARISON OF CHORDWISE PRESSURE DISTRIBUTIONS ADODG-CRM CASE 4.2 OPTIMIZATION REN = 5.00, MACH = 0.850 - SYMBOL SOURCE Baseline Baseline Baseline ALPHA 1.816 2.196 2.552 CL 0.4497 011 513 CD 1905 2155 2453 CM -0.16701-0.18517-0.20390-43.8% Span 93.8% Span - Solution 1 Upper-Surface Isobars ( Contours at 5 ) - 31.2% Span 81.3% Span - - 18.8% Span 68.8% Span - - COMPPLOT Ver 2.04 6.2% Span 56.2% Span John C. Vassberg 17:48 Sun 14 Dec 14 Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 31

COMPARISON OF CHORDWISE PRESSURE DISTRIBUTIONS ADODG-CRM CASE 4.3 OPTIMIZATION REN = 5.00 - SYMBOL SOURCE Baseline Baseline Baseline MACH 0.840 0.850 0.860 ALPHA 2.299 2.196 2.085 CL 008 011 013 CD 2096 2155 2257 CM -0.18138-0.18517-0.19136-43.8% Span 93.8% Span - Solution 1 Upper-Surface Isobars ( Contours at 5 ) - 31.2% Span 81.3% Span - - 18.8% Span 68.8% Span - - COMPPLOT Ver 2.04 6.2% Span 56.2% Span John C. Vassberg 18:27 Sun 14 Dec 14 Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 32

ADO-DG CASE 4.1: Single-Point M = 0.85, C L = 0, Re = 5.0 10 6 Design Cycle Histories Drag & Pitching Moment Baseline.vs. Final Design Comparisons Overlaid Pressure Distributions Side-by-Side Upper-Surface Isobars Overlaid Spanload Distributions Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 33

216 ADODG-CRM CASE 4.1 DRAG HISTORY SYN107 Optimizations Mach = 0.85, CL =, Ren = 5 million Case 4.1 Opt 214 212 Drag Coefficient, CD 210 208 206 204 202 0 10 20 30 40 50 60 70 80 90 100 Design Cycle Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 34

-0.165 ADODG-CRM CASE 4.1 PITCHING-MOMENT HISTORY SYN107 Optimizations Mach = 0.85, CL =, Ren = 5 million Pitching-Moment Coefficient, CMP -0.170-0.175-0.180-0.185 Case 4.1 Opt -0.190 0 10 20 30 40 50 60 70 80 90 100 Design Cycle Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 35

COMPARISON OF CHORDWISE PRESSURE DISTRIBUTIONS ADODG-CRM CASE 4.1 OPTIMIZATION REN = 5.00, MACH = 0.850 SYMBOL SOURCE Baseline Case 4.1 Opt ALPHA 2.196 2.729 CL 011 0.4983 CD 2155 2055 CM -0.18517-0.16695 - - 43.8% Span 93.8% Span - Solution 1 Upper-Surface Isobars ( Contours at 5 ) - 31.2% Span 81.3% Span - - 18.8% Span 68.8% Span - - COMPPLOT Ver 2.04 6.2% Span 56.2% Span John C. Vassberg 17:34 Sun 14 Dec 14 Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 36

COMPARISON OF UPPER SURFACE CONTOURS ADODG-CRM CASE 4.1 OPTIMIZATION REN = 5.00, MACH = 0.850 ( Contours at 5 ) Solution 1: Baseline ALPHA = 2.20, CL = 011 CD = 2155 Solution 2: Case 4.1 Opt ALPHA = 2.73, CL = 0.4983 CD = 2055 Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 37

COMPARISON OF SPANLOAD DISTRIBUTIONS ADODG-CRM CASE 4.1 OPTIMIZATION REN = 5.00, MACH = 0.850 0.8 SYMBOL SOURCE Baseline Case 4.1 Opt ALPHA 2.196 2.729 CL 011 0.4983 CD 2155 2055 CM -0.18517-0.16695 0.7 C*CL/CREF 0.6 SPANLOAD & SECT CL 0.4 0.3 CL 0.2 0.1 0 10 20 30 40 50 60 70 80 90 100 PERCENT SEMISPAN Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 38

ADO-DG CASE 4.2: Triple-Point C L M = 0.85, C L = [0.45, 0, 5], Re = 5.0 10 6 Design Cycle Histories, All 3 Conditions Drag & Pitching Moment Baseline.vs. Final Design Comparisons Overlaid Pressure Distributions Overlaid Spanload Distributions Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 39

185 ADODG-CRM CASE 4.2a DRAG HISTORY SYN107 Optimizations Mach = 0.85, CL = 0.45, Ren = 5 million ADODG-CRM CASE 4.2b DRAG HISTORY SYN107 Optimizations Mach = 0.85, CL = 0, Ren = 5 million ADODG-CRM CASE 4.2c DRAG HISTORY SYN107 Optimizations Mach = 0.85, CL = 5, Ren = 5 million 191 216 246 Case 4.2a Opt Case 4.2b Opt Case 4.2c Opt 244 190 214 242 Drag Coefficient, CD 189 188 187 Drag Coefficient, CD 212 210 208 Drag Coefficient, CD 240 238 236 234 186 206 232 204 230 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 Design Cycle Design Cycle Design Cycle ADODG-CRM CASE 4.2a PITCHING-MOMENT HISTORY SYN107 Optimizations Mach = 0.85, CL = 0.45, Ren = 5 million ADODG-CRM CASE 4.2b PITCHING-MOMENT HISTORY SYN107 Optimizations Mach = 0.85, CL = 0, Ren = 5 million ADODG-CRM CASE 4.2c PITCHING-MOMENT HISTORY SYN107 Optimizations Mach = 0.85, CL = 5, Ren = 5 million -0.162-0.178-0.194-0.163-0.164-0.180-0.196 Pitching-Moment Coefficient, CMP -0.165-0.166-0.167-0.168-0.169-0.170-0.171 Pitching-Moment Coefficient, CMP -0.182-0.184-0.186-0.188 Pitching-Moment Coefficient, CMP -0.198-0.200-0.202-0.204-0.172 Case 4.2a Opt Case 4.2b Opt Case 4.2c Opt -0.173 0 10 20 30 40 50 60 70 80 90 100-0.190 0 10 20 30 40 50 60 70 80 90 100-0.206 0 10 20 30 40 50 60 70 80 90 100 Design Cycle Design Cycle Design Cycle ADO-DG Case 4.2 SYN107 Design-Cycle Histories. Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 40

COMPARISON OF CHORDWISE PRESSURE DISTRIBUTIONS ADODG-CRM CASE 4.2 OPTIMIZATION REN = 5.00, MACH = 0.850 SYMBOL SOURCE Baseline Case 4.2 Opt ALPHA 1.816 2.146 CL 0.4497 0.4489 CD 1905 1856 CM -0.16701-0.16295 - - 43.8% Span 93.8% Span - Solution 1 Upper-Surface Isobars ( Contours at 5 ) - 31.2% Span 81.3% Span - - 18.8% Span 68.8% Span - - COMPPLOT Ver 2.04 6.2% Span 56.2% Span John C. Vassberg 18:03 Sun 14 Dec 14 Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 41

COMPARISON OF SPANLOAD DISTRIBUTIONS ADODG-CRM CASE 4.2 OPTIMIZATION REN = 5.00, MACH = 0.850 0.7 SYMBOL SOURCE Baseline Case 4.2 Opt ALPHA 1.816 2.146 CL 0.4497 0.4489 CD 1905 1856 CM -0.16701-0.16295 0.6 C*CL/CREF SPANLOAD & SECT CL 0.4 0.3 CL 0.2 0.1 0 10 20 30 40 50 60 70 80 90 100 PERCENT SEMISPAN Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 42

COMPARISON OF CHORDWISE PRESSURE DISTRIBUTIONS ADODG-CRM CASE 4.2 OPTIMIZATION REN = 5.00, MACH = 0.850 SYMBOL SOURCE Baseline Case 4.2 Opt ALPHA 2.196 2.522 CL 011 0.4989 CD 2155 2059 CM -0.18517-0.17821 - - 43.8% Span 93.8% Span - Solution 1 Upper-Surface Isobars ( Contours at 5 ) - 31.2% Span 81.3% Span - - 18.8% Span 68.8% Span - - COMPPLOT Ver 2.04 6.2% Span 56.2% Span John C. Vassberg 18:08 Sun 14 Dec 14 Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 43

COMPARISON OF SPANLOAD DISTRIBUTIONS ADODG-CRM CASE 4.2 OPTIMIZATION REN = 5.00, MACH = 0.850 0.8 SYMBOL SOURCE Baseline Case 4.2 Opt ALPHA 2.196 2.522 CL 011 0.4989 CD 2155 2059 CM -0.18517-0.17821 0.7 C*CL/CREF 0.6 SPANLOAD & SECT CL 0.4 0.3 CL 0.2 0.1 1 2 3 4 5 6 7 8 9 10 11 PERCENT SEMISPAN Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 44

COMPARISON OF CHORDWISE PRESSURE DISTRIBUTIONS ADODG-CRM CASE 4.2 OPTIMIZATION REN = 5.00, MACH = 0.850 SYMBOL SOURCE Baseline Case 4.2 Opt ALPHA 2.552 2.868 CL 513 489 CD 2453 2311 CM -0.20390-0.19507 - - 43.8% Span 93.8% Span - Solution 1 Upper-Surface Isobars ( Contours at 5 ) - 31.2% Span 81.3% Span - - 18.8% Span 68.8% Span - - COMPPLOT Ver 2.04 6.2% Span 56.2% Span John C. Vassberg 18:11 Sun 14 Dec 14 Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 45

COMPARISON OF SPANLOAD DISTRIBUTIONS ADODG-CRM CASE 4.2 OPTIMIZATION REN = 5.00, MACH = 0.850 0.8 SYMBOL SOURCE Baseline Case 4.2 Opt ALPHA 2.552 2.868 CL 513 489 CD 2453 2311 CM -0.20390-0.19507 0.7 C*CL/CREF 0.6 SPANLOAD & SECT CL 0.4 0.3 CL 0.2 0.1 0 10 20 30 40 50 60 70 80 90 100 PERCENT SEMISPAN Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 46

COMPARISON OF CHORDWISE PRESSURE DISTRIBUTIONS ADODG-CRM CASE 4.2 OPTIMIZATION REN = 5.00, MACH = 0.850 - SYMBOL SOURCE Case 4.2a Case 4.2b Case 4.2c ALPHA 2.146 2.522 2.868 CL 0.4489 0.4989 489 CD 1856 2059 2311 CM -0.16295-0.17821-0.19507-43.8% Span 93.8% Span - Solution 1 Upper-Surface Isobars ( Contours at 5 ) - 31.2% Span 81.3% Span - - 18.8% Span 68.8% Span - - COMPPLOT Ver 2.04 6.2% Span 56.2% Span John C. Vassberg 17:41 Sun 14 Dec 14 Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 47

ADO-DG CASE 4.3: Triple-Point M M = [0.84, 0.85, 0.86], C L = 0, Re = 5.0 10 6 Design Cycle Histories, All 3 Conditions Drag & Pitching Moment Baseline.vs. Final Design Comparisons Overlaid Pressure Distributions Overlaid Spanload Distributions Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 48

2045 ADODG-CRM CASE 4.3a DRAG HISTORY SYN107 Optimizations Mach = 0.84, CL = 0, Ren = 5 million ADODG-CRM CASE 4.3b DRAG HISTORY SYN107 Optimizations Mach = 0.85, CL = 0, Ren = 5 million ADODG-CRM CASE 4.3c DRAG HISTORY SYN107 Optimizations Mach = 0.86, CL = 0, Ren = 5 million 2100 216 226 2095 Case 4.3a Opt 215 Case 4.3b Opt 224 Case 4.3c Opt 2090 214 222 2085 213 Drag Coefficient, CD 2080 2075 2070 2065 Drag Coefficient, CD 212 211 210 209 Drag Coefficient, CD 220 218 216 214 2060 208 212 2055 207 2050 206 210 205 208 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 Design Cycle Design Cycle Design Cycle ADODG-CRM CASE 4.3a PITCHING-MOMENT HISTORY SYN107 Optimizations Mach = 0.84, CL = 0, Ren = 5 million ADODG-CRM CASE 4.3b PITCHING-MOMENT HISTORY SYN107 Optimizations Mach = 0.85, CL = 0, Ren = 5 million ADODG-CRM CASE 4.3c PITCHING-MOMENT HISTORY SYN107 Optimizations Mach = 0.86, CL = 0, Ren = 5 million -0.175-0.177-0.182-0.176-0.178-0.183 Pitching-Moment Coefficient, CMP -0.177-0.178-0.179-0.180-0.181-0.182 Pitching-Moment Coefficient, CMP -0.179-0.180-0.181-0.182-0.183-0.184-0.185 Pitching-Moment Coefficient, CMP -0.184-0.185-0.186-0.187-0.188-0.189-0.190-0.183 Case 4.3a Opt -0.186 Case 4.3b Opt -0.191 Case 4.3c Opt -0.184 0 10 20 30 40 50 60 70 80 90 100-0.187 0 10 20 30 40 50 60 70 80 90 100-0.192 0 10 20 30 40 50 60 70 80 90 100 Design Cycle Design Cycle Design Cycle ADO-DG Case 4.3 SYN107 Design-Cycle Histories. Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 49

COMPARISON OF CHORDWISE PRESSURE DISTRIBUTIONS ADODG-CRM CASE 4.3 OPTIMIZATION REN = 5.00, MACH = 0.840 SYMBOL SOURCE Baseline Case 4.3 ALPHA 2.299 2.618 CL 008 0.4990 CD 2096 2051 CM -0.18138-0.17550 - - 43.8% Span 93.8% Span - Solution 1 Upper-Surface Isobars ( Contours at 5 ) - 31.2% Span 81.3% Span - - 18.8% Span 68.8% Span - - COMPPLOT Ver 2.04 6.2% Span 56.2% Span John C. Vassberg 18:17 Sun 14 Dec 14 Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 50

COMPARISON OF SPANLOAD DISTRIBUTIONS ADODG-CRM CASE 4.3 OPTIMIZATION REN = 5.00, MACH = 0.840 0.8 SYMBOL SOURCE Baseline Case 4.3 ALPHA 2.299 2.618 CL 008 0.4990 CD 2096 2051 CM -0.18138-0.17550 0.7 C*CL/CREF 0.6 SPANLOAD & SECT CL 0.4 0.3 CL 0.2 0.1 0 10 20 30 40 50 60 70 80 90 100 PERCENT SEMISPAN Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 51

COMPARISON OF CHORDWISE PRESSURE DISTRIBUTIONS ADODG-CRM CASE 4.3 OPTIMIZATION REN = 5.00, MACH = 0.850 SYMBOL SOURCE Baseline Case 4.3 Opt ALPHA 2.196 2.517 CL 011 0.4989 CD 2155 2062 CM -0.18517-0.17768 - - 43.8% Span 93.8% Span - Solution 1 Upper-Surface Isobars ( Contours at 5 ) - 31.2% Span 81.3% Span - - 18.8% Span 68.8% Span - - COMPPLOT Ver 2.04 6.2% Span 56.2% Span John C. Vassberg 18:20 Sun 14 Dec 14 Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 52

COMPARISON OF SPANLOAD DISTRIBUTIONS ADODG-CRM CASE 4.3 OPTIMIZATION REN = 5.00, MACH = 0.850 0.8 SYMBOL SOURCE Baseline Case 4.3 Opt ALPHA 2.196 2.517 CL 011 0.4989 CD 2155 2062 CM -0.18517-0.17768 0.7 C*CL/CREF 0.6 SPANLOAD & SECT CL 0.4 0.3 CL 0.2 0.1 0 10 20 30 40 50 60 70 80 90 100 PERCENT SEMISPAN Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 53

COMPARISON OF CHORDWISE PRESSURE DISTRIBUTIONS ADODG-CRM CASE 4.3 OPTIMIZATION REN = 5.00, MACH = 0.860 SYMBOL SOURCE Baseline Case 4.3 Opt ALPHA 2.085 2.394 CL 013 0.4990 CD 2257 2090 CM -0.19136-0.18256 - - 43.8% Span 93.8% Span - Solution 1 Upper-Surface Isobars ( Contours at 5 ) - 31.2% Span 81.3% Span - - 18.8% Span 68.8% Span - - COMPPLOT Ver 2.04 6.2% Span 56.2% Span John C. Vassberg 18:22 Sun 14 Dec 14 Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 54

COMPARISON OF SPANLOAD DISTRIBUTIONS ADODG-CRM CASE 4.3 OPTIMIZATION REN = 5.00, MACH = 0.860 0.8 SYMBOL SOURCE Baseline Case 4.3 Opt ALPHA 2.085 2.394 CL 013 0.4990 CD 2257 2090 CM -0.19136-0.18256 0.7 C*CL/CREF 0.6 SPANLOAD & SECT CL 0.4 0.3 CL 0.2 0.1 0 10 20 30 40 50 60 70 80 90 100 PERCENT SEMISPAN Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 55

COMPARISON OF CHORDWISE PRESSURE DISTRIBUTIONS ADODG-CRM CASE 4.3 OPTIMIZATION REN = 5.00 - SYMBOL SOURCE Case 4.3 Opt Case 4.3 Opt Case 4.3 Opt MACH 0.840 0.850 0.860 ALPHA 2.618 2.517 2.394 CL 0.4990 0.4989 0.4990 CD 2051 2062 2090 CM -0.17550-0.17768-0.18256-43.8% Span 93.8% Span - Solution 1 Upper-Surface Isobars ( Contours at 5 ) - 31.2% Span 81.3% Span - - 18.8% Span 68.8% Span - - COMPPLOT Ver 2.04 6.2% Span 56.2% Span John C. Vassberg 18:25 Sun 14 Dec 14 Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 56

ADO-DG CASE 4 RESULTS M = 0.85, C L = 0, Re = 5.0 10 6 Comparison of Baseline & Final Designs Tabulated Data Overlaid Pressure Distributions Drag Polars; Alpha Sweeps Drag Rises; Mach Sweeps Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 57

ADO-DG CASE 4 RESULTS Baseline ADO-CRM-Wing. M C L C D cor C M cor 0.84 0 2092-0.18114 0.85 0.45 1906-0.16710 0.85 0 2149-0.18482 0.85 5 2443-0.20340 0.86 0 2249-0.19087 Case 4.1 Optimum Wing. M C L C D cor C D C M cor 0.85 0 2062-0087 -0.16750 Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 58

ADO-DG CASE 4 RESULTS Case 4.2 Optimum Wing. M C L C D cor C D C M cor 0.85 0.45 1861-0045 -0.16330 0.85 0 2064-0085 -0.17855 0.85 5 2318-0125 -0.19548 Case 4.3 Optimum Wing. M C L C D cor C D C M cor 0.84 0 2056-0036 -0.17578 0.85 0 2066-0083 -0.17801 0.86 0 2096-0153 -0.18294 Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 59

COMPARISON OF CHORDWISE PRESSURE DISTRIBUTIONS ADODG-CRM CASE 4 OPTIMIZATIONS REN = 5.00, MACH = 0.850 - SYMBOL SOURCE Baseline Case 4.1 Opt Case 4.2 Opt Case 4.3 Opt ALPHA 2.196 2.729 2.522 2.517 CL 011 0.4983 0.4989 0.4989 CD 2155 2055 2059 2062 CM -0.18517-0.16695-0.17821-0.17768-43.8% Span 93.8% Span - Solution 1 Upper-Surface Isobars ( Contours at 5 ) - 31.2% Span 81.3% Span - - 18.8% Span 68.8% Span - - COMPPLOT Ver 2.04 6.2% Span 56.2% Span John C. Vassberg 17:15 Sun 14 Dec 14 Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 60

0.65 ADODG-CRM DRAG POLARS SYN107 Optimizations Mach = 0.85, Ren = 5 million 0.60 5 Lift Coefficient, CL 0 0.45 0.40 Baseline 0.35 Case 4.1 Opt Case 4.2 Opt Case 4.3 Opt 0.30 10 11 12 13 14 15 16 Idealized Profile Drag, CD - CL^2/(Pi*AR) Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 61

280 ADODG-CRM DRAG RISES SYN107 Optimizations CL =, Ren = 5 million 270 260 Baseline Case 4.1 Opt Case 4.2 Opt Case 4.3 Opt Drag Coefficient, CD ( counts ) 250 240 230 220 210 200 190 0.80 0.81 0.82 0.83 0.84 0.85 0.86 0.87 0.88 0.89 0.90 Mach Number Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 62

ADO-DG CASE 4 SUMMARY Optimizations Yield Designs with Near Theoretical Minimum Drag Essentially No Shock Drag Near Minimum Induced Drag Near Minimum Profile Drag per Thickness Distribution Case 4.1 Single-Point Design Exhibits Poor Off-Design Performance Case 4.2-4.3 Triple-Point Designs Relinquished < counts at Cruise Condition Multi-Point Optimizations Clearly Improve Off-Design Characteristics Jameson & Vassberg, AIAA Sci-Tech, Kissimmee FL, January 2015 63