Compiled: February 12, 2020 (version 5)

2nd US-Japan Workshop on Data-Driven Fluid Dynamics

March 12-14, 2020 Kobe Meriken Park Oriental Hotel, Kobe, Japan

Sponsored by

Compiled: February 12, 2020 (version 5)

Abstract: Over the past two decades, the fluid dynamics community has enjoyed the advancement in computational, experimental, and theoretical techniques to analyze a variety of fluid flows. Developments in computational and experimental hardware, numerical algorithms, and unsteady measurement techniques have enabled not only detailed analysis of flow physics but also initiated cross-talks amongst the various disciplines of fluid mechanics. With these powerful toolsets now available, the fluid dynamics community has started to examine complex flows with high levels of unsteadiness, nonlinearity, and multi-scale dynamics. However, there still exist limitations on how modern analysis techniques can be applied to specific fluid dynamics problems. Theoretical and computational approaches are often limited to relatively simple flows at low Reynolds numbers, while practical applications require extension to more complex unsteady and turbulent flows. Extending the current state of the art in flow analysis to higher Reynolds number flows requires tackling high-dimensional physics and the associated big-data from numerical simulations or experimental measurements. Some of the recent innovations in data science may hold the key to address these issues. The aim of this workshop is to gather fluid dynamics and data science experts from their respective areas and discuss their ongoing progress and challenges in emerging analysis techniques, including data science, computational & theoretical fluid dynamics, and advanced experimental diagnostic methods, that can be shared with others to facilitate breakthroughs as a community. This event will stimulate discussions and collaborations between members of the research communities to identify key areas that can make the largest impact and to offer a vehicle to further strengthen research collaborations across the Pacific. Following the success of the first “US-Japan Workshop on Bridging Data Science and Fluid Mechanics,” we are holding the second workshop now entitled the “US-Japan Workshop on Data-Driven Fluid Dynamics.” Target Areas: Data-inspired techniques for fluid dynamics, including but not limited to data-driven analysis, modeling, estimation, and control of fluid flows. Sponsors:

We thank the generous support from

Asian Office of Aerospace Research and Development Ebara Corporation Hitachi Ltd. Honda R&D Co., Ltd. Intelligent Light Mathworks Tokyo University of Science (Center for Data Science)

Compiled: February 12, 2020 (version 5)

Organizers

Kunihiko Taira (UCLA)

Kozo Fujii (Tokyo University of Science)

Koji Fukagata (Keio University)

Taku Nonomura (Tohoku University)

Steve Brunton (University of Washington)

Maziar Hemati (University of Minnesota) Attendees

Gregg Abate (AFOSR)

Kei Ambo (Honda R&D)

Byungjin An (Ebara)

Steven Brunton (University of Washington)

Tim Colonius (Caltech)

Scott Dawson (Illinois Institute of Technology)

Jiro Doke (Mathworks)

Kozo Fujii (Tokyo University of Science)

Koji Fukagata (Keio University)

Shunsuke Fukagawa (Ebara)

Kai Fukami (Keio University)

Akira Goto (Ebara)

Hiroshi Gotoda (Tokyo University of Science)

Michael Graham (University of Wisconsin)

Maziar Hemati (University of Minnesota)

Michio Inoue (Mathworks)

Yoshinobu Kawahara (Kyushu Univ/RIKEN)

Sung-Eun Kim (ONR Global)

Petros Koumoutsakos (ETH Zurich/Harvard University) - keynote speaker

Nathan Kutz (University of Washington)

Steve Legenski (Intelligent Light)

Stephen Makinen (Intelligent Light)

Krithika Manohar (Caltech)

Takashi Misaka (AIST)

Takemasa Miyoshi (RIKEN) - keynote speaker

Matthew Munson (ARO)

Takayuki Nagata (Tohoku University)

Aditya Nair (University of Washington)

Kumi Nakai (Tohoku University)

Masamichi Nakamura (Hitachi)

Motohiko Nohmi (Ebara)

Taku Nonomura (Tohoku University)

Fortunato Nucera (Honda R&D)

Compiled: February 12, 2020 (version 5)

Yuya Omichi (JAXA)

Clarence Rowley (Princeton)

Koma Sato (Hitachi)

Peter Schmid (Imperial College London)

Koji Shimoyama (Tohoku University)

Douglas Smith (AFOSR/EOARD)

Kunihiko Taira (UCLA)

Tomoaki Tatsukawa (Tokyo University of Science)

Seiji Tsutsumi (JAXA)

Chi-An Yeh (UCLA)

Compiled: February 12, 2020 (version 5)

DAY 1 (March 12) REGISTRATION 8:00-8:30 OPENGING REMARKS AND LOGISTICS 8:30-8:45 Kunihiko Taira (UCLA) MACHINE LEARNING FOR FLUID FLOWS (Chair: Koji Fukagata) 8:45-9:10 Koji Fukagata (Keio University) Applications of CNN autoencoder-type models to several flow problems 9:10-9:35 Steven Brunton (University of Washington)

Interpretable and Generalizable Machine Learning for Fluid Mechanics 9:35-10:00 Kai Fukami (Keio University)

Image-based flow reconstruction with machine-learning-based super-resolution analysis

10:00-10:20 Coffee break 10:20-10:45 Nathan Kutz (University of Washington)

Discovery of Coordinates and Dominant Balances in Fluid Flows 10:45-11:10 Michael Graham (University of Wisconsin) Deep learning to discover and predict dynamics on an inertial manifold 11:10-11:35 Tomoaki Tatsukawa (Tokyo University of Science)

A Study of Missing Value Interpolation using MLP for Airfoil Surface Pressure Distribution

KEYNOTE PRESENTATION (Chair: Steve Brunton) 11:35-12:15 Petros Koumoutsakos (ETHZ/Harvard University)

Natural and Artificial Intelligence for Flow Control LUNCH 12:15-1:30 PANEL DISCUSSION: MACHINE LEARNING FOR FLUID DYNAMICS

Compiled: February 12, 2020 (version 5)

1:30-2:00 Moderator: Steve Brunton (Washington)

Panelists: Petros Koumoutsakos (ETHZ/Harvard), Koji Fukagata (Keio), Seiji Tsustumi (JAXA)

ANALYSIS OF FLUID FLOWS (Chair: Kunihiko Taira) 2:00-2:25 Scott Dawson (Illinois Institute of Technology)

Predicting modal decompositions without data or numerics 2:25-2:50 Yoshinobu Kawahara (Kyushu University/RIKEN)

Data-driven Koopman analysis for interactions in dynamical systems 2:50-3:15 Seiji Tsutsumi (JAXA)

Modal and Causal Analyses for Thermoacoustic Coupling Mechanism of Combustion Instability

3:15-3:30 Coffee Break 3:30-3:55 Stephen Makinen (Intelligent Light)

Flow Feature Identification with Dynamic Mode Decomposition 3:55-4:20 Chi-An Yeh (UCLA)

Network broadcasting mode of 2D turbulence 4:20-4:45 Kunihiko Taira (UCLA)

Resolvent analysis for large-scale high-Re problems 4:45-5:10 Kozo Fujii (Tokyo University of Science)

Cellular Automaton Air Traffic Flow Simulations toward Efficient and Robust Management

RECEPTION 6:00 At the hotel

Compiled: February 12, 2020 (version 5)

DAY 2 (March 13) REGISTRATION 8:00-8:30 COMPUTING SYSTEMS (Chair: Nathan Kutz) 8:30-8:55 Michio Inoue (MathWorks Japan)

MATLAB ecosystem

8:55-9:15 Kozo Fujii (Tokyo University of Science) How shall we promote AI/DS activities in universities? - An example from TUS/Data Science Center

PANEL DISCUSSION: EDUCATION AND DISSEMINATION 9:15-9:40 Moderator: Nathan Kutz (Washington)

Panelists: Jiro Doke (MathWorks), Steve Brunton (Washington), Kozo Fujii (Tokyo Univ Sci)

9:40-10:00 Coffee Break MODELING FLUID FLOWS (Chair: Taku Nonomura) 10:00-10:25 Clancy Rowley (Princeton University)

Learning low-order models from data 10:25-10:50 Hiroshi Gotoda (Tokyo University of Science)

Nonlinear dynamics of thermoacoustic combustion instability 10:50-11:15 Tim Colonius (Caltech)

Data-driven modeling for RANS-based prediction of turbulence structure and radiated noise

11:15-11:40 Koji Shimoyama (Tohoku University)

Bayesian and Learning-based Uncertainty Quantification in an Aerodynamic Problem

LUNCH BREAK 11:40-1:00

Compiled: February 12, 2020 (version 5)

KEYNOTE PRESENTATION (Chair: Kozo Fujii) 1:00-1:40 Takemasa Miyoshi (RIKEN)

Big Data Assimilation in Numerical Weather Prediction 1:40-2:00 Coffee Break ESTIMATING FLUID FLOWS (Chair: Maziar Hemati) 2:00-2:25 Maziar Hemati (University of Minnesota)

Sensor selection for flow reconstruction and control 2:25-2:50 Peter Schmid (Imperial College London)

Towards data-driven analysis of intermittent systems 2:50-3:15 Taku Nonomura (Tohoku University)

Determinant-based Greedy Sensor Selection Algorithm for Fluid Dynamics 3:15-3:40 Krithika Manohar (Caltech)

Kernel Analog Forecasting for Multiscale Problems 3:40-4:05 Takashi Misaka (AIST)

Image-based Parameter Estimation of Vortex-Induced Vibration Using Neural Network

4:05-4:20 Break PANEL DISCUSSION: ANALYZING, MODELING AND ESTIMATING FLUID FLOWS 4:20-5:00 Moderator: Maziar Hemati (Minnesota)

Panelists: Takemasa Miyoshi (RIKEN), Clancy Rowley (Princeton), Krithika Manohar (Caltech)

Compiled: February 12, 2020 (version 5)

DAY 3 (March 14) REGISTRATION 8:00-8:30 APPLICATIONS OF DATA-DRIVEN FLUID DYNAMICS (Chair: Matthew Munson) 8:30-8:55 Aditya Nair (University of Washington)

Data-driven control strategies for unsteady wake flows 8:55-9:20 Yuya Ohmichi (JAXA)

Dynamic mode decomposition analysis of dominant flow structures around an atmospheric entry capsule

9:20-9:55 Shunsuke Fukagawa & Byungjin An (Ebara Corporation)

Data Science Application in Ebara 9:55-10:10 Coffee Break 10:10-10:35 Matthew Munson (ARO) ARO Investments in Fluid Dynamics 10:35-11:00 Steve Legenski (Intelligent Light)

Robust Data Management for Production-scale CFD Data Science PANEL DISCUSSION: APPLICATIONS OF DATA-DRIVEN FLUID DYNAMICS AND OUTLOOK 11:00-11:40 Moderator: Matthew Munson (ARO)

Panelists: Akira Goto (Ebara), Gregg Abate (AFOSR), Peter Schmid (Imperial) CLOSING REMARKS 11:40-12:00 Kozo Fujii (Tokyo University of Science) EXCURSIONS AND NETWORKING ACTIVITIES 1:30 Depart around 1:30 (details to be provided)

Compiled: February 12, 2020 (version 5)

Compiled: February 12, 2020 (version 5)

Compiled: February 12, 2020 (version 5)

Access from Airports to Conference Venue (Meriken Park Oriental Hotel) From Kobe Airport (UKB)

1) Kobe Airport – Conference Venue (Mekiken Park Oriental Hotel) Approx. 20 min. by taxi (~JPY 4,000).

From Osaka International Airport (Itami/ITM)

1)Osaka International Airport - Sannomiya station Approx. 40 min by Airport Bus. (7:50-21:15, every 15-30min).

Ticket available on site (JPY1,050). 2) Sannomiya station - Conference Venue (Meriken Park Oriental Hotel)

Approx. 5 min. by taxi(~JPY1,000). From Kansai International Airport (KIX)

1) Osaka International Airport - Sannomiya Station Approx. 65 min by Airport Bus. ( http://www.hanshin-bus.co.jp/limo_en/index_k_sannomiya1.html ) Ticket available on site (JPY1,950).

2) Sannomiya station - Conference Venue (Meriken Park Oriental Hotel) Approx. 5 min. by taxi(~JPY1,000).

From Tokyo Narita Airport (NRT)

1) Narita Airport - Tokyo Station Approx. 60 min by Narita Express ( https://www.jreast.co.jp/e/nex/tickets.html )

You could buy a ticket on site all the way from Narita airport to Shin-Kobe (for (1) and (2) totaling JPY16,940).

2) Tokyo Station - Shin Kobe Station Approx. 160 min by Tokaido Shinkansen. It is convenient to buy the tickets of 1) and 2) together at Narita airport as above, but, you can also buy the Shinkansen ticket at Tokyo station.

3) Shin Kobe Station to - Conference Venue (Meriken Park Oriental Hotel) Approx. 15 min. by taxi (~JPY2,000).

From Tokyo Haneda Airport (HND)

1) Haneda Airport - Shinagawa Station Approx. 25 min by Keikyu Railways. Make sure the train stopped at Shinagawa. Ticket available at the station (JPY410).

2) Shinagawa Station – Shin-Kobe Station Approx. 140min by Tokaido Shinkansen. Ticket available at the station (JPY9,290).

3) Shin Kobe Station - Conference Venue (Meriken Park Oriental Hotel) Approx. 15 min. by taxi (~JPY2,000).

Compiled: February 12, 2020 (version 5)

Notes:

Compiled: February 12, 2020 (version 5)

Notes:

Compiled: February 12, 2020 (version 5)

Notes: