191101 Presentation in UCLA by Kanazaki Masahiro

Transcript

1. Introduction of my researches and intersts Masahiro Kanazaki (Tokyo Metropolitan University/ JAXA) kana@tmu.ac.jp @Kanazaki_M (twitter.com/Kanazaki_M) University of California (Nov. 1st, 2019) - Self-introduction - My (our) researches - SST with forward-swept wing - Slat noise reduction - Mars airplane (airfoil) - Optimization of driving condition for plasma actuator - My current interests
2. Thank you for your warm welcome! 2
3. Self-Introduction(1/3) Masahiro Kanazaki • Born at Sapporo in Hokkaido • 2004. Ph. D, Tohoku university • 2004-2008, Invited researcher, JAXA • 2008-, Associate Professor, Tokyo Metropolitan University • 2016-, Guest researcher, JAXA  Research interests: Aerodynamic design, MDO, Global optimization methods, in aerospace engineering 3
4. Self-Introduction(2/3)  My (our) researches  Design Optimizations  Exhaust manifold design by evolutionary algorithm • Complex geometry  High-lift airfoil design • Kriging model, expected improvement  Design optimizations in wind tunnel testing  Supersonic wing design  BWB design  Airfoil design and data mining for a Mars airplane  Hybrid-rocket engine design  Trajectory optimization of a satellite for active space debris removal  Aerodynamic Evaluations  Intake-airframe calculation by CFD for a SST  Aerodynamic evaluation of staging separation for TSTO  POD/ DMD for a entry capsule 4  My (our) researches  Design Optimizations  Exhaust manifold design by evolutionary algorithm • Complex geometry  High-lift airfoil design • Kriging model, expected improvement  Design optimizations in wind tunnel testing  Supersonic wing design  BWB design  Airfoil design and data mining for a Mars airplane  Hybrid-rocket engine design  Trajectory optimization of a satellite for active space debris removal  Aerodynamic Evaluations  Intake-airframe calculation by CFD for a SST  Aerodynamic evaluation of staging separation for TSTO  POD/ DMD for a entry capsule
5. Self-Introduction(3/3) Why I visit Prof. Taira lab, UCLA. Impressive keynote speech by Prof. Taira at “CFD symposium” in Tokyo Interesting presentation by Dr. Zhang in AIAA Scitech 2019. (Zhang, K., Taira, K., Hayostek, S., Amitay, M., He, W. and Theofilis, V., “Wake Dynamics of Finite Aspect Ratio Wings. Part II: Computational Study,” AIAA 2019-1385, San Diego, 2019.) 5
6. My (our) researches SST with forward swept wing(1/4) Students studying on SST 6 Mr. Yuki Kishi (Master of eng.) Specialty: Supersonic wing design, Global optimization by surrogate modeling Mr. Nao Setoguchi Specialty: Unstructured mesh based CFD, Aero-Flight computation and trajectory optimization
7. My (our) researches SST with forward swept wing(2/4) Started with sonicboom reduction Low-speed and high-angle aerodynamics Unstructured mesh based RANS (Re=107) 7 JAXA’s Baseline
8. My (our) researches SST with forward swept wing(3/5) Different vortex structure Several studies on aerodynamics for backward swept delta wing (ex. Miller, D. S., et al., J Aircraft, 21, 9, pp. 680-686, 1984.) 8
9. My (our) researches SST with forward swept wing(4/4) Flight path optimization (Aerodynamic-Flight simulation based EA) Future plan: Understanding the relationship between planform and aerodynaimics using undteady flow simulation (DES)
10. My (our) researches Noise Reduction of Slat (1/5) Students studying on HLD 10 Mr. Hotaka Kobayashi (Master of eng.) Specialty: RANS, DES, noise analysis, Global optimization by surrogate modeling, High-lift airfoil Now: working for IDAJ Mr. Syuhei Kondo Specialty: RANS, DES, noise analysis, Geometry representation, High-lift airfoil
11. My (our) researches Noise Reduction of Slat* (2/5) RANS and DES for slat with displacing its positions 11 Baseline = JAXA’s experimental model “OTOMO” RANS mesh (10h/case) DES mesh (1month/case ~) *Airfoil element which maximize Clmax
12. My (our) researches Noise Reduction of Slat (3/5) Noise around 0.1-1.0kHz: Case1<2<3=4 Area of recirculation region: Case1<2<3=4 Correlation between the noise and the circulation area 12 Case1 Case3 Case2 Case4
13. My (our) researches Noise Reduction of Slat(4/5) Reattachment distance s||R-TE|| 13 RANS can estimate s||R-TE|| Optimization technique by optimization technique! Case1(DES) Case4 (DES) Case1(RANS) Case4 (RANS)
14. My (our) researches Noise Reduction of Slat(5/5) Design optimization using s||R-TE|| Kriging based EA Future plan: Geometry optimization of LE of mother airfoil Automated judgement of flowfield similarity between RANS and DES • Clustering of flowfield in view of s||R-TE|| and C|lmax| 14 Maximize s||R-TE|| Maximize C|lmax|
15. My (our) researches Mars Airplane (1/6) Students studying on Mars Airplane 15 Mr. Kazufumi Uwatoko Interest : Evolutionary algorithm, Airfoil design, Rotate wing. (Second job: beast master?) Mr. Kai Tomisawa (Master Eng.) Interest: Mars airplane, Unstructured mesh based CFD, Aircraft design Now: working for NTT Mr. Hiroto Kittaka Interest: Mars airplane, Unstructured mesh based CFD, Aero-Flight
16. My (our) researches Mars Airplane (2/6) Planing in JAXA Expectation: Wider range exploration Aerodynamic challenges: 1% less air density, lower Reynolds number and higher Mach number than the Earth atmosphere Mars Airplane Balloon Experiment（MABE） Flight testing at 36km altitude 16
17. My (our) researches Mars Airplane (3/6) Result of MABE-1 flight testing: lost of pitching control We redesign MABE as MABE-2 Aerodynamic evaluation and database by CFD 17
18. My (our) researches Mars Airplane (4/6) Evolutionary Algorithm 18 Rank1 Rank2 Rank3 f1 f2 Optimum direction →Non dominated sorting genetic algorithm (NSGA) by Prof. Deb (MIT) Non dominated sorting for multi-objective ranking
19. My (our) researches Mars Airplane (5/6) Lift and drag maximization design problem Parametric in section (PARSEC) for airfol representation* 19 Optimum direction Reynolds number: 10000 Mach: 0.02 *Kanazaki, M., Sato, T., and Matsushima, K., "Parametric Airfoil Representation Toward Efficient Design Knowledge Discovery under Various Flow Condition," Transaction of The Japan Society for Aeronautical and Space Science, Vol. 12, No. APISAT-2013, pp. a93-a98, April, 2015.
20. My (our) researches Mars Airplane (6/6) Vortex on airfoil Blade planform design Blade element theory + EA Future plan: Aerodynamic evaluation for MABE-2 Topology analysis of vortex pattern around airfoil 20 Uwatoko, K., Kanazaki, M., Nagai, H., Fujita, K. and Oyama, A., “Blade Element Theory Coupled with CFD Applied to Optimal Design of Rotor for Mars Exploration Helicopter, ” AIAA Scitech2020, January 8 (afternoon, APA-26, Aerodynamic Design and Analysis I), 2020.
21. My (our) researches Optimization of driving condition for PA(1/8) Researchers studying on PA 21 Dr. Takashi Matsuno (Lecturer, Tottori University) Specialty: Flow control, Wind tunnel, experiment, Vortex Masahiro Kanazaki
22. My (our) researches Pulse Width Modulation(PWM) PA  Efficient AC supplement for PA  Optimum values of (T1 , T2 ) or (1/T1 , 1/T2 ) are unknown. Requirement to find the optimum AC wave form Flow simulation by CFD: over 10 hours. Real time scale in wind tunnel: 1~ sec. → Optimization during a wind tunnel experiment in real time 22 Optimization of driving condition for PA(2/8)
23. My (our) researches Optimization of driving condition for PA(3/8) Kriging model based EA 23 Jones, R. D., et al., “Efficient Global Optimization of Expensive Black-Box Functions,” Journal of Global Optimization Volume 13, Issue 4, pp 455–492, 1998. Estimated Improvement sample (min fref ) design variable x f(x) Sample Initial model Initial designs Additional designs c Improved model c c
24. My (our) researches Pulse Width Modulation(PWM) PA  Efficient AC supplement for PA  Optimum values of (T1 , T2 ) or (1/T1 , 1/T2 ) are unknown. Requirement to find the optimum AC wave form Flow simulation by CFD: over 10 hours. Real time scale in wind tunnel: 1~ sec. → Optimization during a wind tunnel experiment in real time 24 Optimization of driving condition for PA(4/8)
25. My (our) researches 25 Fully automated optimization based on the wind tunnel evaluation. Design variable (Power supply) Objective function(Aerodynamic force) Optimization of driving condition for PA(5/8)
26. My (our) researches 26  Modulation frequency：  Duty ratio： [%] m p x f T f 1 20 1 1 mod    1 2 100 T T D cycle   Power supply unit provide frequency fp 9kHz and 20/fp as a one unit wave. [Hz]  Objective function  Design variables Minimize CD (Drag coefficient) 2 .0 ≤ xm ≤ 90.0 10.0 ≤ Dcycle ≤ 70.0 Optimization of driving condition for PA(6/5)
27. 27 Result（1/5） Lower xm = Higher jet energy
28. 28 Result（1/5）
29. 29 Result（1/5）
30. 30 Result（1/5）
31. 31 Result（1/5）
32. 32 Result（1/5）
33. 33 Result（1/5）
34. 34 Result（1/5）
35. 35 Result（1/5）
36. 36 Result（1/5）
37. 37 Result（1/5）
38. 38 Result（1/5）
39. 39 Result（1/5） Local minimum Global minimum
40. My (our) researches 40 x m [-] D cycle [%] f mod [Hz] C D DesA 2.0 60.0 400.0 0.2985 DesB 15.0 25.0 53.3 0.3272 DesC 88.0 55.0 9.1 0.4105 DesA DesB DesC  DesA: Separated region was reduced, and the streamline was less deformed from the uniform flow  DesB: Separated region was reduced, the streak of smoke far downstream from the model was blurred Optimization of driving condition for PA(8/8)
41. My current interests Unsteady flow on the forward-swept wing for SST Automated detection of flow structure similarity between time-averaged flow and unsteady flow Topology analysis of flow structure Constrained optimization problem I will be UCLA ~11th, Nov. (Mon), leaving for Tokyo 12th (0.05am) 41
42. References  Kishi, T., Kitazaki, S., Makino, Y., Kanazaki, M., Ariyarit, Y., "Planform Dependency of Optimum Cross-sectional Geometric Distributions for Supersonic Wing," Aerospace Science and Technology, Elsevier, Vol. 90, pp. 181-193, 2019. [DOI: 10.1016/j.ast.2019.03.057]  Setoguchi, N., and Kanazaki, M., “Low-Speed and High Angle of Attack Aerodynamic Characteristics of Supersonic Business Jet with Forward Swept Wing,” AIAA Scitech2020, January 7 (morning, APA-12, High-Speed Aircraft Aerodynamics), 2020.  Kanazaki, M., and Saisyo, R., “Genetic Algorithm Applied to the Time-Series Landing Flight Path and Control Optimization of a Supersonic Transport,” Proceedings of the 2019 3rd International Conference on Intelligent Systems, Metaheuristics & Swarm Intelligence, pp. 19-24, 2019. [DOI: 10.1145/3325773.3325789]  Uwatoko, K., Kanazaki, M., Nagai, H., Fujita, K. and Oyama, A., “Blade Element Theory Coupled with CFD Applied to Optimal Design of Rotor for Mars Exploration Helicopter, ” AIAA Scitech2020, January 8 (afternoon, APA-26, Aerodynamic Design and Analysis I), 2020.  Kanazaki, M., Tanaka, K., Jeong, S., and Yamamoto, K., “Multi-Objective Aerodynamic Exploration of Elements' Setting for High-Lift Airfoil Using Kriging Model,” J Aircraft, Vol. 44, No. 3, pp. 858-864, May 2007. [DOI: 10.2514/1.25422]  Kanazaki, M., Sato, T., and Matsushima, K., "Parametric Airfoil Representation Toward Efficient Design Knowledge Discovery under Various Flow Condition," Transaction of The Japan Society for Aeronautical and Space Science, Vol. 12, No. APISAT-2013, pp. a93-a98, April, 2015. [DOI: 10.2322/tastj.12.a93]  Kanazaki, M., Matsuno, T., Maeda, K., and Kawazoe, H., "Efficient Global Optimization Applied to Wind Tunnel Evaluation Based Optimization for Improvement of Flow Control by Plasma Actuator," Engineering Optimization, Taylor & Francis, Volume 46, Issue 12, pp. 1-17, 2014. (IDOI: 10.1080/0305215X.2014.958733 )) 42