High-Performance Computing Model for Bio-Fuel Combustion with Artificial Intelligence

Jay P. Gore and Hasti Veeraraghava Raju

School of Mechanical Engineering
Purdue University

Lean blowout (LBO) calculations and statistical analysis for a conventional (A-2) and an alternative bio-jet fuel (C-1) are performed in a realistic gas turbine combustor geometry. The high-performance computing methodology is developed based on large eddy simulation (LES) models for turbulence and detailed chemistry and flamelet based models for combustion. The bio-jet fuel (C-1) exhibits significantly larger CH2O concentrations in the fuel-rich regions compared to the conventional petroleum fuel (A-2) at an identical equivalence ratio. As expected, the temperature of the recirculating hot gases is an important parameter for maintaining a stable flame. If this temperature falls below a certain threshold value for a given fuel, the evaporation rates and heat release rates decrease significantly and cause lean blowout. This study established the minimum recirculating gas temperature needed to maintain a stable flame for the A-2 and C-1 fuels. Artificial Intelligence (AI) models, based on high fidelity LES data, aimed at early identification of the incipient LBO condition. Sensor-based monitoring using a Support Vector Machine (SVM) detected the onset of LBO approximately 20 ms ahead of the event. A convolutional autoencoder was trained for feature extraction from the mass fraction of the OH for all time-steps resulting in significant dimensionality reduction. The extracted features along with ground truth labels are used to train a support vector machine (SVM) model for binary classification. The binary classification indicated an LBO approximately 30 ms ahead of the actual blowout. This and other early results highlight the promise of AI in much needed engine health monitoring.

Jay P. Gore is the Reilly University Chair Professor of Engineering at Purdue University with a primary appointment in the School of Mechanical Engineering and courtesy appointments in the School of Aeronautical and Astronautical Engineering and the School of Chemical Engineering. Jay was one of the youngest faculty members to receive the honor of a named chair at Purdue. He has served as the Associate Dean for Research and Entrepreneurship within the College of Engineering and as the founding Director of the Energy Center in Discovery Park at the Purdue campus level. Jay Gore spent six months as a visiting faculty at Nagoya University and six months as a visiting gas turbine engineer at the Rolls Royce Corporation. Dr. Gore served on the Indiana State Fire Commission for five years. Jay served as a Jefferson Science Fellow in the Office of the Special Representative for Intergovernmental Affairs (S/SRGIA, an office directly reporting to then Secretary of State) and wrote white papers on how developed countries can minimize their CO2 emissions and interact with emerging and developing economies to help the world address the crisis of global warming. In May 2017, Dr. Vishwanath Karad MIT World Peace University of Pune, India appointed Jay as the Principal Transformation Advisor. During his continuing academic career, Jay has advised the dissertation and theses work of 30 PhD students and twice as many MS students. They have published over 150 archival journal papers and twice as many conference papers in: infrared radiation heat transfer and sensing, combustion energy efficiency, radiant burners, renewable energy carbon dioxide recycling, and lean premixed and partially premixed gas turbine combustors. Jay has taught sophomore and junior thermodynamics classes and MS and PhD level combustion classes. Jay’s awards include the Presidential Young Investigator Award, the ASME Best paper in Heat Transfer literature award, the Purdue Innovator Hall of Fame Award, the Purdue College of Engineering teamwork award for founding the Summer Undergraduate Research Fellowships (SURF) program, and most recently the Purdue School of Mechanical Engineering Discovery Award. Jay has been recognized as a distinguished alumnus by College of Engineering, Pune. Jay has served as the Chair of the ASME Heat Transfer Division K11 Committee, a member of the AIAA Propulsion and Combustion committee and as the Chair of the Central State Section of the Combustion Institute (International), and as the Papers Chair for an International Combustion Symposium. Professor Gore is a Fellow of the American Society of Mechanical Engineering (ASME), the American Institute of Aeronautics and Astronautics (AIAA), and the Combustion Institute International.

Wednesday, January 15, 2020
3:30 PM
Zumberge Hall of Science, Room 159 (ZHS 159)

Refreshments will be served at 3:15 pm.

host: Ronney

Probabilistic Learning on Manifolds: The Small Data Challenge

Roger Ghanem

Gordon S. Marshall Professor of Engineering Technology
Sonny Astani Department of Civil and Environmental Engineering
USC
Los Angeles, CA

As the pace of technological innovation and scientific discovery continues to grow, so does the interest in accelerating their integration. We are thus, increasingly, faced with the task of product development without the benefit of hindsight or historical failures. Examples of this evolving paradigm include new materials and novel configurations of complicated systems with complex behavior. This challenge is exacerbated by the growing interactions between technological and socio-economic systems where failure of a technological component can have implications on social trends and public policy, thus highlighting the need to characterize extreme events both for each component and at the system level. The standard paradigm of mapping knowledge into engineered systems where new systems are essentially construed as perturbations of older systems is not equipped for these emerging requirements. Recent approaches under the general heading of Machine Learning (ML) are motivated by the explosion in sensing technologies. Fundamental advances in these ML methods are being realized at the interface of data science and physics constraints.

In this talk I will describe a recent effort within my group along these ML lines. I will focus on one particular approach, the Probabilistic Learning on Manifolds (PMoL), which is relevant under conditions of small data. This approach aims to augment a (small) training dataset with realizations that share with it some key features making these realizations credible surrogates of the original data. These features consist of 1) co-location on a manifold, and 2) statistical consistency. Thus as a first step, we associated a “manifold” with the training set, that we believe represents all the fundamental constraints (such as physics). We rely on diffusion maps constructs to delineate the manifold. Construed as fluctuating within this manifold, the training dataset is statistically more significant. As a second step, we generate samples on the manifold that have the same probability distribution as the training set. To this end, we construct a projected Ito equation whose invariant measure is that of the training set, and whose samples are constrained to the manifold.

I will show how the above ideas are used as building blocks in a scramjet optimization problem and the design of a digital twin for a structural composite.

Roger Ghanem is the Gordon S. Marshall Professor of Engineering Technology and Professor in the Department of Civil and Environmental Engineering at USC. He has a courtesy appointment in AME. Before joining USC in 2005, he had served on the faculty of Johns Hopkins University and SUNY-Buffalo. He obtained his PhD from Rice University working on stochastic mechanics.

Ghanem’s work is in the general area of predictive science with focus on physics-constrained systems where issues of multiscale and multiphysics interactions are increasingly relevant. Ghanem is fellow of AAAS, SIAM, USACM, IACM and EMI. He has served as president of the Engineering Mechanics Institute of ASCE, as Chair of the UQ-SIAG of SIAM, and on the executive council of USACM.

Wednesday, January 22, 2020
3:30 PM
Zumberg Hall of Science, Room 159 (ZHS 159)

Refreshments will be served at 3:15 pm.

host: Egolfopoulos

Hydrofoiling Honeybee

Chris Roh

Research Engineer and Lecturer
Graduate Aerospace Laboratories
California Institute of Technology
Pasadena, CA

Honeybees display a unique bio-locomotion strategy at the air-water interface. When water’s adhesive force traps them on the surface, their wetted wings lose ability to generate aerodynamic thrust. However, they adequately locomote, reaching a speed up to three body lengths·s-1. Honeybees use their wetted wings as hydrofoils for their water surface propulsion. Their locomotion imparts hydrodynamic momentum to the surrounding water in the form of asymmetric waves and a deeper water jet stream, generating approximately 20 μN average thrust. The wing kinematics show that the wing’s stroke plane is skewed, and the wing supinates and pronates during its power and recovery strokes, respectively. The flow under a mechanical model wing mimicking the motion of a bee’s wing further shows that non-zero net horizontal momentum is imparted to the water, demonstrating net thrust. Moreover, a periodic acceleration and deceleration of water is observed, which provides additional forward movement by ‘recoil locomotion’. Scaling analysis of the hydrodynamic forces associated with the wing motion indicates that the wings utilize added mass force (unsteady inertial force associated with the pulling of the water attached to the wing). Hydrofoiling highlights the versatility of their flapping-wing systems that are capable of generating propulsion with fluids whose densities span three orders of magnitude. This discovery inspires a novel aerial-aquatic hybrid vehicle.

Chris Roh received his B.S. in Bioengineering from Cornell University in 2012 and his M.S. and Ph.D degree in Aeronautics from California Institute of Technology (Caltech) in 2013 and 2017. He is currently a Research Engineer and Lecturer at Caltech. From a young age, Chris has been fascinated by the diversity of insects and different stories each insects tell. This deep-rooted passion combined with his new found love for the intricate ways fluid flow has led him to study the hydrodynamics of insects at Caltech under the guidance of Professor Morteza Gharib. In his thesis work, he studied the jet vectoring ability of aquatic dragonfly larva’s tri-leaflet valve, as well as how honeybees swim using their wings. These studies have broad implications, as the tri-leaflet valve of dragonfly larvae helps us imagine a new prosthetic heart valve and honeybee’s swimming demonstrates versatility of flapping wing systems. Chris continues to observe interesting ways insects interact with the fluid surrounding them with engineering applications in mind. He is a recipient of National Science Foundation Graduate Research Fellowship, and Richard B. Chapman Memorial Award.

Wednesday, January 29, 2020
3:30 PM
Zumberg Hall of Science, Room 159 (ZHS 159)

Refreshments will be served at 3:15 pm.

host: Pahlevan

Fluid Mechanics in Clinical Echocardiography

Pavlos P. Vlachos

Professor
School of Mechanical Engineering
and
School of Biomedical Engineering
Purdue University
West Lafayette, IN

Color-Doppler velocity field reconstruction and vortex identification for a baby with hypoplastic left-heart syndrome in-utero.

In this talk we will probe flows in cardiac disease using in-vivo measurements in clinical settings, and we will discuss how traditional experimental fluids mechanics tools can translate into clinical practice.

Flows in the cardiovascular system manifest intrinsic complexity, which is often associated with diseased states. Imaging modalities such as ultrasound/echocardiography and phase-contrast MRI provide unique opportunities and challenges for flow measurements in patients. Currently, the relationship between clinical flow measurements and clinical diagnostic parameters is qualitative, and often is reliant on heuristics and non-physical assumptions.

In this talk we will discuss how to overcome these limitations by integrating medical imaging with experimental fluid mechanics, in order to, ultimately, improve accuracy, robustness, and clinical diagnostic utility of these tools.

Specifically, we will discuss how fluid mechanics can be used in the analysis of echocardiographic imaging for heart failure. We will show an improved approach for clinical implementation of EchoPIV (echocardiographic Particle Image Velocimetry) and a new method for the velocity reconstruction of Color-Doppler flow imaging. Finally, we will present a use-case in the analysis of fetal and neonatal echocardiograms of babies born with single ventricle (hypoplastic left heart syndrome). If time permits, some additional examples of application to 4D flow MRI will be presented.

Pavlos P. Vlachos is a professor in the School of Mechanical Engineering at Purdue. Dr. Vlachos received his Diploma in Mechanical Engineering from the National Technical University of Athens (1995) and his MS (1998) and PhD (2000) in Engineering Science and Mechanics from Virginia Tech. He joined the Department of Mechanical Engineering at Virginia Tech as an assistant professor (2003) and he was promoted to associate (2007), then full professor (2011). In August 2013 he joined the School of Mechanical Engineering at Purdue, and he currently serves as President’s Fellow for Research Development. He is also associate editor for the International Journal of Multi-Phase Flows. His research focuses on experimental fluid mechanics specializing on flow diagnostics using optical or other non-intrusive methods. He has worked on wakes, boundary layers, aerothermodynamics, and fluid structure interaction. Currently his main interests are in biofluids, arterial flows, heart failure, medical imaging, drag delivery; and tissue and tumor micro-environments. He has been funded for over 90 projects and >$40 million in research. With his students and co-workers he has authored >120 journal papers and over 280 conference proceedings. He has received several awards including: 2005 Dean’s Award of Excellence for Outstanding Assistant Professor; MIT 11th Annual T.F. Ogilvie Lectureship Award for Young Investigator in Ocean Engineering and Fluid Mechanics; 2006 NSF CAREER award; 2007 College of Engineering Faculty Fellow; 2007 and 2010 ASME Fluids Engineering Division Moody Award; 2010 Dean’s Award for Excellence in Research; and the 2009 and 2010 outstanding paper award in fluid mechanics from the journal of Measurement Science and Technology. Several papers have been recognized in journal highlights and covers including the cover for PNAS Dec 2015 issue. In 2010 he was named a Jones Faculty Fellow and in 2012 received the Robert E. Hord professorship in mechanical engineering. In 2015 became Purdue University Faculty Fellow. Bio– produces a new paragraph.

Wednesday, January 29, 2020
3:30 PM
Zumberg Hall of Science, Room 159 (ZHS 159)

Refreshments will be served at 3:15 pm.

host: Pahlevan

Fluid Mechanics of Intracranial Aneurysms:
Fundamental Aspects and Application to Clinical Decision-Making

Alberto Aliseda

PACCAR Professor of Mechanical Engineering
Department of Mechanical Engineering
University of Washington
Seattle, WA

The fluid mechanics inside intracranial aneurysms dominate the efficacy of endovascular treatment methods, modulating the mechanical stresses and residence times inside the sac and at the aneurysmal neck. Embolic coils and flow-diverting stents, the two dominant types of endovascular devices for treatment, are designed to slow down flow inside the aneurysmal volume and reduce stresses on the aneurysmal sac, creating an environment that enables successful thrombosis in the aneurysm, which eliminates the risk of rupture.

In-vitro experiments characterize the hemodynamics inside intracranial aneurysms, prior to treatment and post-treatment with flow-diverting stents. We use stereo (2D-3C) and 3D (3D-3C) particle image velocimetry (PIV) to explore the parameter space of aneurysms in a large cohort of patients followed along several years. The flow measurements are interpreted as a combination of two canonical flows: flow in a curved pipe and cavity flow. As such, the parent-vessel Reynolds and Dean numbers are the relevant non-dimensional parameters. Unsteadiness in the cardiac cycle introduces the Womersley number as a third component of flow inertia. Despite inertia dominating the parent-vessel flow, flow-diverting stents significantly reduce the velocity inside the aneurysmal sac, leading to viscous-dominated flow. A critical Dean number is identified that separates two opposite flow behaviors that could help predict treatment success.

I will also discuss a computational investigation of a large population of patients whose aneurysm treatments are followed over time, to determine the mechanism by which endovascular treatment fails to prevent aneurysmal growth. A novel modeling technique that uses high-resolution, synchrotron micro-CT scans to understand the flow inside coiled aneurysm enables homogenization methods for improved porous medium representation of deployed coils or stents, improving the clinical utility of the simulation results.

Alberto Aliseda is the PACCAR Professor of Mechanical Engineering at the University of Washington in Seattle, WA, USA, where he has been in the faculty since 2006. He also holds adjunct (courtesy) appointments in Neurological Surgery and Aeronautics and Astronautics. Prior to the UW, he obtained his PhD and did postdoctoral research at the University of California, San Diego. Before that, he earned a B.S./M.S. in Aerospace Engineering from the Polytechnic University of Madrid in 1998. His current interests focus on biomedical flows, with a special emphasis on the biomechanical basis of cardiovascular disease and the interaction of medical devices with flows in the heart and arteries. He also works on turbulent and multiphase flows, including flows of interest to energy conversion and environmental problems, such as liquid atomization and cloud microphysics. He has been a Visiting Professor at the Universidad Carlos 3 de Madrid, the Ecole Normale Superieure de Lyon and the Laboratoire des Ecoulements Geophysiques et Industriels (LEGI) of the Université Grenoble-Alps

Wednesday, February 19, 2020
3:30 PM
Zumberg Hall of Science, Room 159 (ZHS 159)

Refreshments will be served at 3:15 pm.

host: Pantano

Human Spaceflight–Recent Past, Near Future and Educational Activities at Viterbi

Garrett Reisman

Professor of Astronautical Engineering
Department of Astronautical Engineering
University of Southern California
Los Angeles, CA

Prof. Reisman’s presentation will highlight his personal experiences flying on the Space Shuttle and the International Space Station while serving as a NASA Astronaut from 1998 to 2011. After describing these unique experiences he will discuss his transition to SpaceX and the state of the commercial human spaceflight industry. Finally, the human spaceflight graduate coursework which has recently been established in Viterbi’s ASTE department will be presented.

Garrett Reisman, a NASA veteran who flew on all three Space Shuttles, was selected by NASA as a mission specialist astronaut in 1998. His first mission in 2008 was aboard the Space Shuttle Endeavour which dropped him off for a 95 day stay aboard the International Space Station after which he returned to Earth aboard the Space Shuttle Discovery. His second mission in 2010 was aboard the Space Shuttle Atlantis. During these missions, Garrett performed 3 spacewalks, operated the Space Station Robot Arm and was a flight engineer aboard the Space Shuttle. Not only was he an astronaut, but Garrett was also an aquanaut serving as a crewmember on NEEMO V, living on the bottom of the sea in the Aquarius deep underwater habitat for 2 weeks.

After leaving NASA in early 2011, he joined Elon Musk at SpaceX where he served in multiple capacities most recently as the Director of Space Operations.

Garrett stepped down from his full-time position at SpaceX in May of 2018 and in June 2018 he became a Professor of Astronautical Engineering in the Viterbi School at USC. He also continues to support SpaceX as a Senior Advisor.

Garrett attended the University of Pennsylvania and Caltech where he received his Ph.D in 1997.

Wednesday, February 26, 2020
3:30 PM
Zumberg Hall of Science, Room 159 (ZHS 159)

Refreshments will be served at 3:15 pm.

host: Luhar

Challenges and Opportunities for Accelerating Scientific Discovery with Deep Learning

Greg Ver Steeg

Research Associate Professor
Information Sciences Institute
University of Southern California
Marina Del Rey, CA

The successes of neural networks in computer vision and natural language processing have not easily translated into breakthroughs in other scientific domains. I will discuss some of the principles behind learning representations of data with deep learning and how we have adapted these ideas to study problems like gene expression, neuroimaging, and clinical health records. I will conclude with a speculative discussion about whether these methods can benefit domains that traditionally rely on large-scale numerical simulations like computational fluid dynamics.

Greg Ver Steeg is a Research Lead at ISI and Research Associate Professor in USC’s CS department. He has slowly transitioned from PhD research at Caltech on detecting quantum entanglement to his current work on detecting hidden variables in more diverse domains using information theory and machine learning. Dr. Ver Steeg’s work has been recognized with an AFOSR Young Investigator Award and an Amazon Research Award.

Wednesday, March 4, 2020
3:30 PM
Zumberg Hall of Science, Room 159 (ZHS 159)

Refreshments will be served at 3:15 pm.

host: Oberai

host: Bermejo-Moreno

host: Pahlevan