Speaker: Kaushik Reddy
Date: Thursday, 28th March 2024
Time: 3:00 PM – 3:45 PM (Indian Standard Time – GMT +5:30)
Abstract:
While trucks make up only a small proportion of the total number of vehicles on roads, they contribute to approximately 40% of the total CO2 emissions released by automobiles. The GCM Tractor-Trailer model is a simple geometry of a Heavy Duty Truck (HDT), whose aerodynamic drag contributes up to ~65% of the total energy needed for operation.
Computational Fluid Dynamics (CFD) is an essential tool that can be used to predict the flow over such entities, while offering potential for massive fuel cost savings and aerodynamic optimisation. The aim of this webinar is to introduce and demonstrate the capability of SankhyaSutra Taral (a state-of-the-art Lattice-Boltzmann Method-based CFD platform) towards accurate prediction of heavy-vehicle aerodynamics and validate the resulting simulation data with experiments.
Topics Covered during the talk:
– CFD steps in the simulation using SankhyaSutra Taral
– Post-processing workflow for SankhyaSutra Taral’s web-based GUI
Speaker: Kaushik Reddy
Date: Wednesday, 28th February 2024
Time: 3:00 PM – 3:45 PM (Indian Standard Time – GMT +5:30)
Abstract:
Dive into the world of Computational Fluid Dynamics (CFD) with our upcoming webinar, where we’ll unveil the power of SankhyaSutra Taral, a cutting-edge CFD platform based on the Lattice-Boltzmann Method.
Join us as we showcase how SankhyaSutra Taral empowers engineers to effortlessly capture critical data like Drag and Lift coefficients, enabling seamless comparison with real-world results. Don’t miss this opportunity to explore validation studies for intricate geometries such as the Ahmed Body, a proven gateway to mastering complex flow dynamics in vehicle models.
Topics Covered during the talk:
– Description of Ahmed Body model
– CFD steps in the simulation using SankhyaSutra Taral
– Post-processing workflow for SankhyaSutra Taral’s web-based GUI
Speaker: Athreya Ballapur Jayasimha
Date: Wednesday, 24th January 2024
Time: 3:00 PM – 3:45 PM (Indian Standard Time – GMT +5:30)
Abstract:
Static and dynamic stability characteristics of aircrafts define how they behave during operation. Experimental analysis of dynamic stability requires extracting dynamic stability derivatives from the force and moment coefficients as the aircraft undergoes periodic motion about a body axis.
As a part of the ongoing series on how to leverage LBM simulations using SankhyaSutra Taral to address real life CFD problems, SankhyaSutra Labs has organized this webinar to explain the methodology and workflow for performing stability analyses of aircraft.
We use the Stability And Control Configuration (SACCON) as an example. SACCON is a generic Unmanned Combat Aerial Vehicle (UCAV) model developed by NATO to assess state-of-the-art in CFD methods for the prediction of static and dynamic stability, and control characteristics of military vehicles.
Topics Covered during the talk include:
Speaker: Athreya Ballapur Jayasimha
Date: Monday, 18th December 2023
Time: 2:30 PM – 3:15 PM (Indian Standard Time – GMT +5:30)
Abstract:
In this webinar, a case study on mixing in a cylindrical stirred-tank reactor of diameter 150mm will be performed. A single Rushton-type impeller of diameter 50mm, with 4 impeller blades rotating at 250 rpm, will be simulated using the Single Resolution (SR) grid strategy and Moving Geometry implementation built into SankhyaSutra Taral. Step-by-step commentary on the simulation setup will be given, and basic post-processing to visualise flow field patterns inside the stirred tank will be showcased.
Speaker: Athreya Ballapur Jayasimha
Date: Friday, 27th October 2023
Time: 2:30 PM – 3:15 PM (Indian Standard Time – GMT +5:30)
Abstract: SankhyaSutra Taral is a state-of-the-art Lattice Boltzmann method based CFD solver. During this webinar, we will simulate a NACA0012 airfoil at Reynolds number of 50,000 and AoA of 12 degrees using SankhyaSutra Taral. We will study the complex flow phenomena over the airfoil. Step-by-step commentary on importing geometry file, specifying relevant parameters like periodic boundary conditions, and introducing probes to save simulation data will be provided. An overview of the web-based Graphical User Interface (GUI) with its in-built functionality for pre- and post-processing will also be given, to highlight how SankhyaSutra Taral can be utilized within your CFD workflow.
Speaker: Prof Pinaki Chakraborty, Okinawa Institute of Science & Technology, Japan
Date: Friday, 1st September 2023
Time: 2:30 PM – 3:30 PM (Indian Standard Time – GMT +5:30)
Abstract: A hurricane over ocean functions as a heat engine, its heat source being the moisture from the warm ocean. When a hurricane hits land, the heat source is lost and consequently it decays. This decay is considered to be a non-thermodynamic process. Contrary to this prevailing paradigm, we argue that thermodynamics plays a key role in the evolution of landfalling hurricanes and that the thermodynamic effect is orchestrated by the moisture stored in the hurricane from its journey over ocean prior to landfall. This talk is based on joint research with Lin Li.
Biography:
Speaker: Dhairyashil Ghatage, SankhyaSutra Labs
Date: Thursday, 17th August 2023
Time: 2:30 PM – 3:30 PM (Indian Standard Time – GMT +5:30)
Abstract:
An semiconductor device be it electronic chip or solar cell is fabricated via a series of physico-chemical processes such as etching, photolithography, deposition which take place on a substrate, typically a silicon wafer. The fabrication process takes anywhere between 10 to 30 days. While being in the clean room, the substrate is transported in and out of different chambers each specialised to carry out a particular process. The yield of a fabrication plant depends upon the speed and accuracy with which the physico-chemical processes.
The inherent assumptions in all these steps are a) the substrate is placed in exactly the same manner in each chamber, b) the temperature of the wafer is uniform, c) the reactive gases also have identical concentration anywhere inside the chamber without any gradients. A series of mechanical processes such as wafer descent at given speed, electric chucking, and thermal cooling keep the wafer at the desired position and operating conditions i.e. temperature. Typically a showerhead is used to distribute the reactive agents inside the chamber in a uniform manner.
As the requirements for the high precision fabrication and yield get more stringent, it is essential to understand the processes in more detail and explore the avenues for further optimizations. Computational simulations allow us to study the processes from first principles. However the fabrication processes occur at significantly varied length and time scales, e.g. the thickness of deposits in chemical vapour deposition would be few nanometers whereas the substrate diameter itself is typically 300 mm for silicon wafer. The depositing particles undergo surface diffusion at time scales of sub-micro seconds whereas the chamber gas flow can have characteristic scale of few seconds. Hence, it is important to identify and use appropriate numerical techniques for each process.
In this talk, we shall describe the customised solutions developed by SankhyaSutra Labs to carry out computational simulations for some of the fabrication processes namely gas flow through showerhead, temperature control assembly, wafer descent, wafer cooling, and plasma enhanced chemical vapour deposition.
Speaker: Prof. Krishnaswamy Nandakumar, Louisiana State University
Date: Wednesday, 26th July 2023
Time: 3:00 PM – 4:00 PM (Indian Standard Time – GMT +5:30)
Abstract: The manufacturing technologies of the future for converting chemicals, materials, energy etc will be done in efficient, distributed, modular process equipment where multiphase flows are ubiquitous. Our traditional design approach has been to rely on rules of thumb, pilot-scale development, and testing of process equipment which takes up to 20 years to develop a single technology. The design procedures are often highly empirical, dismissing an engineer’s high degree of freedom at the early stages of design by making ad-hoc design decisions but paying the price during scale-up of processes through expensive pilot-scale experiments. The question I address in this presentation is, “Can Advanced Computational modeling tools come to our rescue in minimizing the need for pilot-scale experiments?” On the fundamental side, advanced algorithms for direct numerical simulation (DNS) and Discrete Element Modelling (DEM) of multiphase flows aid in detailed understanding but for limited size. For dispersed rigid particles, the Navier-Stokes equations are coupled with the rigid body dynamics in a rigorous fashion to track the particle motion in a fluid. These classes of algorithms show great promise in attempting to shed light on multiphase flows from which we can extract statistically meaningful average behavior for use in the design of large-scale engineering equipment.
At LSU we focus on technologies that integrate multiphase flow modeling with process diagnostics, intensification studies, and optimization and control as applied to the process industries. Case studies of industrial relevance will be presented to illustrate the benefits of such an approach.
Biography: Dr. K. Nandakumar is Gordon A and Mary Cain Chair Professor at Louisiana State University. Before this, he was the GASCO Chair Professor at The Petroleum Institute, Abu Dhabi. Formerly he was in the Department of Chemical and Materials Engineering at the University of Alberta, Edmonton, Canada, for nearly 25 years. Dr. Nandakumar received his B. Tech from Madras University in 1973, his M. Sc from the University of Saskatchewan in 1975, and his Ph.D. from Princeton University in 1979. He received the Alexander von Humboldt research fellowship from the German government in 1989-90 and the Albright & Wilson Americas Award from the Canadian Society of Chemical Engineering in 1991 for distinguished contributions to chemical engineering before reaching the age of 40. Dr. Nandakumar was elected a Fellow of the Chemical Institute of Canada in 1991, a Fellow of the Engineering Institute of Canada in 2006, and a Fellow of the Canadian Academy of Engineering in 2007. He has received, from the University of Alberta, the McCalla Professorship (1992), the Killam Annual Professorship (2001) for excellence in research, and the Rutherford Award (2001) for excellence in teaching. He has also received the Excellence in Education award (2002) from APEGGA, the professional engineering association in Alberta. He was Editor-in-Chief of The Canadian Journal of Chemical Engineering from 2005 to 2009. Dr. Nandakumar also received the premier The Canadian Society for Chemical Engineering award, the R.S. Jane Memorial Award 2008. At LSU, he received the 2021 LSU Alumni Association Faculty Excellence Award. During the summer of 2022, he visited TIFR under the VAJRA fellowship of SERB, Govt. of India.
Speaker: Abhishek Chopra, BosonQ Psi
Date: Tuesday, 25th April 2023
Time: 6:30 PM – 7:30 PM (Indian Standard Time – GMT +5:30)
Abstract: Computational fluid dynamics (CFD) plays a crucial role in simulating fluid flow across various fields, including aerospace, automotive, and biomedical engineering. However, classical computing limitations restrict the accuracy, speed, and complexity of CFD simulations. Quantum computing, on the other hand, has the potential to revolutionize CFD by solving complex problems that are currently intractable using classical computing methods. To achieve significant advancements in CFD simulations over the next decade, the CFD Vision 2030 report outlines the key challenges and research areas that need to be addressed.
In this talk, we will discuss the underlying principles of quantum computing and the current state of quantum hardware. We will explore how quantum computing can be applied to CFD simulations, including the use of quantum algorithms such as Harrow-Hassidim-Lloyd (HHL), Variational Quantum Eigensolver (VQE), Differentiable Quantum Circuits (DQC), and others. We will also examine the challenges and opportunities in integrating quantum computing into CFD workflows and the potential impact on various industries, such as aerospace, energy, and healthcare.
Through this talk, our aim is to inspire the role of quantum computing in achieving the goals of the CFD Vision 2030 report through the development of new algorithms, optimization of existing algorithms, and hardware advancements in CFD. This talk will be suitable for anyone interested in the intersection of quantum computing and computational fluid dynamics, from students and researchers to industry professionals. Join us in this exciting and rapidly evolving field to inspire further research and advancements.
Biography: Abhishek Chopra is a dynamic and accomplished entrepreneur and researcher specializing in the fields of computational sciences and Quantum computing. He is the Founder and CEO of BosonQ Psi, where he is driven to achieve the company’s vision and mission. Abhishek draws on his extensive experience and expertise to explore new business opportunities with customers, partners, and investors, propelling the company’s growth. He has played a pivotal role in securing more than $1 Mn in private equity and government funding. He has successfully established strategic partnerships with major tech companies like IBM, Microsoft, Intel, and TCS.
Abhishek holds a master’s in aerospace engineering from Rensselaer Polytechnic Institute (USA). His research centered on developing high-fidelity and parallel Computational Fluid Dynamics (CFD) code for eVTOLs, wind turbines, and rotorcrafts. He graduated with honors (magna cum laude) from Rutgers University, where he earned his bachelor’s degree in aerospace engineering. He has received awards like J.J. Slade Scholar, AIAA GNC Travel Grant, Dean Scholar, and Purdue SURF fellowship. He has published five research papers in major journals and conference proceedings related to Aerospace Engineering.
Abhishek is a highly respected keynote speaker who has been invited to speak at distinguished events like ASSOCHAM’s Quantum Technology Conclave 2021, PHD Chamber of Commerce and Industry’s Artificial Intelligence in Defense, and The Quantum Computing Summit 2022. His steadfast passion for entrepreneurship, computational sciences, and Quantum computing has established him as an esteemed figure in the industry.
Speaker: Prof Ameya Jagtap, Brown University
Date: Friday, 24th January 2023
Time: 4:30 PM – 5:30 PM (Indian Standard Time – GMT +5:30)
Abstract: Traditional approaches for scientific computation have undergone remarkable progress, but they still operate under stringent requirements, such as the need for precise knowledge of underlying physical laws, precise knowledge of boundary and/or initial conditions, and often time-consuming workflows such as mesh generation and long-term simulation. On top of these limitations, high-dimensional problems governed by parameterized PDEs cannot be tackled. Moreover, seamlessly incorporating noisy data is still a challenge for solving inverse problems efficiently. Physics-informed machine learning (PIML) has emerged as a promising alternative for solving the problems mentioned above. In this talk, we will discuss a particular type of PIML method, namely, physics-informed neural networks (PINNs). We review some of the current capabilities and limitations of PINNs and discuss diverse applications where PINN has proved to be very effective compared to traditional approaches. We also discuss the extensions of the current PINN method, such as conservative PINNs (cPINNs) and extended PINNs (XPINNs), for big data and/or large models. To this end, we will also discuss various adaptive activation functions that can accelerate the convergence of deep and physics-informed neural networks.
Biography: Prof Ameya Jagtap is an Assistant Professor of Applied Mathematics (Research) at Brown University, USA. He received his PhD and Master degrees, both, in Aerospace Engineering from the Indian Institute of Science (IISc), India. He then joined the Tata Institute of Fundamental Research – Centre for Applicable Mathematics (TIFR-CAM), India as a postdoctoral research fellow. Later, He moved to Brown University to pursue my postdoctoral research in the division of applied mathematics. A key focus of his research work is to develop data and physics-driven scientific machine learning algorithms applicable to a wide range of problems in computational physics. His expertise lies in the field of Scientific Machine Learning, Deep Learning, Data/Physics-driven deep learning techniques with multi-fidelity data, Uncertainty Quantification/Propagation, Multi-scale & Multi-physics simulations, Computational Continuum Mechanics (Solids, Fluids, and Acoustics), Spectral/Finite Element Methods, WENO/DG schemes, Domain decomposition techniques, etc. He is also interested in the development of novel artificial neural network architectures, which give faster convergence.
Speaker: Dr Vidyadhar Mudkavi, Retired Scientist, CSIR National Aerospace Laboratories
Date: Thursday, 19th January 2023
Time: 3:00 PM – 4:00 PM (Indian Standard Time – GMT +5:30)
Abstract: In this talk we discuss what constitutes modeling in the context of fluid dynamics. While the Navier-Stokes equations are capable of describing most of the flow situation in real life, they are tedious to solve using some of the most powerful computers. On the other hand, in many situations it is not necessary to deal with Navier-Stokes in their most complex forms. Understanding the physical issues underlying the problem at hand can help simplify the equations and solve them with much less effort without losing sight of physics. A few well known examples will be covered to illustrate this. A few practical applications will also be discussed emphasising the fact that modeling is an art.
Biography: Dr Vidyadhar Mudkavi is Distinguished Emeritus Professor at Academy of Scientific and Innovative Research and Honorary Visiting Professor at IIT Dharwad. He got his Bachelors degree in Aeronautical Engineering from IIT Kanpur in 1983. He completed his Masters studies at IISc as Gold Medalist in 1985. He obtained his Doctoral studies from California Institute of Technology in 1991. He has worked extensively in the areas of Modelling and Simulation, Atmospheric Flows, Flight Related Activities and High Performance Computing (HPC), which have resulted in multiple publications in journals and conference proceedings. He joined CSIR National Aerospace Laboratories (NAL) in 1992, where he rose to the rank of Outstanding Scientist. He was head of CTFD division at CSIR-NAL during 2008-2016. As the head of CSIR Fourth Paradigm Institute (CSIR-4PI) during 2016-2021, he played a key role in setting up major HPC facilities for CSIR. He has served on many important committees including Executive Board of National Supercomputing Mission, Technical Advisory Committee for UAV Development for Atmospheric Measurements, and Interdisciplinary Committee for LCA-Tejas Wake Clearance. He is a Life Member of Aeronautical Society of India and Fellow of The Institution of Engineers (India). He has received numerous awards for his contributions including Outstanding performance award and Director’s special award from CSIR-NAL.
Speaker: Prof. Suresh M Deshpande, Retired Professor, Indian Institute of Science
Date: Thursday, 15th December 2022
Time: 2:30 PM – 3:30 PM (Indian Standard Time – GMT +5:30)
Abstract:The talk will focus on rarefied and low-density flows initially, and then move on to more complex flows, e.g., that on gas turbine blade. After presenting studies of shock structure in rarefied flows, I will discuss Monte Carlo Simulation of low-density flows. The work on the Boltzmann equation led me to develop various kinetic numerical methods- such as kinetic flux vector splitting, peculiar velocity based schemes, low dissipation m-KFVS (adjoint based optimisation) and least squares kinetic upwind method (LSKUM). LSKUM moved from the research table to concrete applications of store separation in CSIR-NAL and DRDL, petal separation in DRDL and strongly rotating flows in BARC. The second part presents computing flow on a gas turbine blade using CPU-GPU HPC platforms. The flow is highly complex, involves transition and relaminarisation.
Biography: A Gold medallist in B.E. (Electrical Engineering) from Nagpur University, S M Deshpande obtained his M.E. and Ph.D degrees in Aeronautical Engineering from IISc Bangalore. He joined the same department as a Lecturer in 1969 and retired as Professor in 2004. He was instrumental in initiating Computational Fluid Dynamics (CFD) research in the country. His Kinetic Theory based CFD algorithms continue to influence research in this area to date. Design data for India’s Agni Missile program were computationally obtained at DRDL Hyderabad under his guidance. He established the AR & DB Centre of Excellence for Aerospace CFD at IISc. Dr. A.P.J. Abdul Kalam used to call him CFD Gandhi. After retirement from IISc in 2004, he served as consultant to Engineering Mechanics Unit at JNCASR till the year 2020. He is a recipient of several honours and awards including Satish Dhawan Chair Professorship at IISc (2000-03), Biren Roy Trust Award of AeSI (1991), IISc Alumni Award for Excellence in Research (Engineering, 2001) and DRDO Academic Excellence Award (2002). He is an elected fellow of the Indian Academy of Sciences (1974), Indian National Academy of Engineering (1995), Aeronautical Society of India (1998) and Maharashtra Academy of Sciences (2001).
Speaker: Prof. Sauro Succi, Italian Institute of Technology
Date: Monday, 28th November 2022
Time: 2:30 PM – 3:30 PM (Indian Standard Time – GMT +5:30)
Abstract: Fluid flows are an ubiquitous presence across most natural and industrial phenomena. Life on our planet and our body alike crucially depend on the motion of air and water in complex geometrical environments and so do most industrial processes, based on the flow of mass momentum and energy across different regions of space and time. Even though the equations of motion of fluids are known for two centuries, their actual solution still poses a steep challenge even to the most powerful supercomputers. This is particularly true for turbulent flows in complex geometries, such as the ones encountered in most engineering applications. This is why computational fluid dynamics (CFD) is in ceaseless demand of new and more efficient methods to solve the fluid equations.
In the last decades, a family of methods based on lattice formulations of the Boltzmann kinetic equation have met with major success in advancing the CFD front, thanks, among others, to their outstanding amenability to massively parallel computing. These Lattice Boltzmann (LB) methods have proven capable to cover an amazingly broad spectrum of applications across scales of fluid motion, from astrophysical jets, all the way down to quark gluon plasmas.
Yet, they still expose some weaknesses, primarily the loss of stability in the presence of substantial compressibility and heat exchange at high Reynolds flows. Such weaknesses have been recently addressed by a new generation of LB methods, based on clever variants of the original LB scheme in the direction of enhancing its stability, accuracy and amenability to high-performance computing.
In this talk, after presenting a rapid survey of the historical development of the LB method and its major achievements, I shall provide a cursory account of the latest developments and associated prospects for advancing knowledge in this paramount field of modern science and society, not least the current harsh energy crisis.
Speaker: Prof. Viswanathan Kumaran, Indian Institute of Science, Bangalore, India
Date: Thursday, 3rd November 2022
Time: 2:00 PM – 3:00 PM (Indian Standard Time – GMT +5:30)
Abstract: Lab-on-a-chip devices have held out the promise to revolutionise health- care by carrying out clinical tests in autonomous devices that perform all the functions of a clinical diagnostic laboratory. However, the few tests that have been commercialised (blood glucose meters, pregnancy kits) are based on capillary action, and do not involve sample preparation. Attempts to carry out tests involving sample preparation have so far proved unviable, and there have been spectacular failures, casting doubt on the initial promise. We discuss the technological reasons for the difficulty in miniaturising a diagnostic laboratory, and the fundamental advances required to make this feasible.
There are two technologies used in a diagnostic laboratory, sample preparation and measurement. The technological barriers to both of these will be discussed. The flow at small scales is laminar, and this results in a technological barrier for sample preparation in chip-based devices. In nature, this barrier is overcome by a transition to a turbulent flow at large scales. Our research work on triggering turbulence in small devices at low velocities using soft surfaces, as a means of significantly increasing mixing rates, will be discussed after an introduction to lab-on-a-chip devices and the transition to turbulence.
Quantitative measurements at small scales are the second technological barrier. The example of a blood cell counting device will be discussed, where optical and impedance measurements lack the resolution required for accurate results. The use of deeper analysis based on the physics and flow dynamics will be discussed as a way forward to reach the required levels of resolution and accuracy.
