Rajeev Jaiman

Professor

B. Tech. (IIT Bombay), M.S., Ph.D. (U of Illinois, Urbana-Champaign, Aerospace Engineering), Senior Member AIAA, Member APS, ASME, SNAME, USACM, NSERC/Seaspan Industrial Research Chair in Intelligent and Green Marine Vessels (IGMVs), SNAME Fellow

Research Interests

• Fluid-structure interaction
• Computational mechanics
• Intelligent ocean systems
• Data-driven modeling and control
• Scientific machine learning
• Multiphase flows

Current Research Work

My current research program is concentrated on a diverse set of topics related to high-fidelity multiphysics and multiphase simulations, fluid-structure interaction (FSI), flow control techniques, model order reduction and data-driven computing.  While the traditional high-fidelity full-order simulations provide valuable physical insight for coupled problems, these full-order models (FOM) are strongly mechanistic, computationally expensive, memory demanding and time-consuming for design space exploration, control, and optimization, even on supercomputing facilities. Another dimension of our research involves the development of efficient data-driven model order reduction (MOR) or dimensionality reduction techniques, which are of practical importance in a broad range of problems in aerospace and marine/offshore engineering. Further highlights of ongoing research themes are as follows:

• Computational Methods and Numerical Analysis: Over the past years, we have developed a broad range of methods and algorithms to address challenges in simulating large-scale multifield, multiphase and multidomain problems. The main difficulty arises from the dilemma of Eulerian-Lagrangian coupling and in the treatment of interface for wide-ranging and differing physical scales and underlying discretizations. Traditional predictive approaches to nonlinear problems pursued by the computational mechanics community have provided many achievements, but have fallen short of the reliability, robustness, and accuracy, in particular for strong inertial coupling of thin-structures with the incompressible fluid flow. In our group, we have solved several numerical issues about fluid-structure interaction and multiphysics modeling. Based on novel techniques, the developed in-house multiphysics solver provides a fully coupled analysis tool for marine/offshore, aerospace and biomechanics applications.
• Software Development and Practical Applications: Our in-house 3D Multiphysics computational framework is novel in a number of ways and is general-purpose to simulate large-scale engineering applications, owing to the underlying principle of a partitioned iterative scheme for coupled partial differential equations. Our variational parallel solver can handle a wide range of boundary conditions and can simulate arbitrary CAD geometries using unstructured meshes for multiphysics and multiphase systems.
• Physics of Fluid-Structure Interaction and Aeroelasticity: Besides the development and the applications of discretization techniques to real-world problems, the developed methods are explored to answer a range of fundamental questions arising from both canonical and practical problems. Owing to the high accuracy and robustness, new methods and numerical solvers enable to simulate a wide range of physical scales and complexity associated with structure-to-fluid mass ratios, Reynolds number, large structural deformation, and proximity interference with the flexibility to use non-matching spatial and temporal discretizations.
• Data-Driven Computing and Machine Learning: Advances in high-performance computing (HPC) have empowered us to perform large-scale simulations for billions of variables in complex coupled multifield, multibody and multiphase systems. These high-fidelity simulations have been providing invaluable physical insight into the development of new designs and devices in offshore engineering. Despite efficient algorithms and powerful supercomputers, the state-of-art CFD and coupled fluid-structure simulations are somewhat inefficient hence less attractive with regard to design optimization, parameter space exploration and the development of control and monitoring strategies for engineering structures. Our recent developments focus on integrating the HPC-based high-fidelity CFD with the emerging field of data science and machine learning.
• New Flow Control Techniques and Devices: The development of high-fidelity tools and the discovery of new physical mechanisms can naturally lead to a host of new designs and control strategies for practical use. As a part of the current research focus, a variety of active and passive flow control techniques have been developed for the complex phenomena of vortex-induced vibration, wall turbulence, droplet-wall interaction and shock-boundary layer interaction, and among others. Various forms of grooves, auxiliary surfaces and patterns, and external excitations are currently explored for controlling these phenomena.
• Efficient Bio-inspired Structures and Feedback Flow Control: It is fair to assume that Nature provides somewhat optimized and well-evolved solutions to many engineering problems. The current challenge is the application of computer simulations to model bio-inspired systems for understanding Nature’s best ideas and then tailor and train (via machine learning algorithms) these designs and processes to solve engineering problems. The primary source of complexity comes from the interdisciplinary and coupled dynamics character of bio-inspired systems. A variety of concepts, ranging from morphing flexible structures, efficient locomotive systems to drag reduction techniques, have the potentials for improving the efficiency and advancement of the existing engineering systems and processes. Currently, we are extending our flexible multibody FSI framework to model bio-inspired flapping motion (e.g., fish or bat-like motion) to maximize aero-/hydrodynamic performance and/or to minimize acoustics and vibrational problems.

Selected Publications

• Gao, R. and Jaiman, R.K. Predicting fluid–structure interaction with graph neural networks, Physics of Fluids, 36, 013622, 2024. Editors’ Pick
• Gao, R., Shayan Heydari, Jaiman, R.K. Towards spatio-temporal prediction of cavitating fluid flow with graph neural networks, Int. J. of Multiphase Flow, 177, 104858, 2024.
• Lak, S., Jaiman, R.K. A numerical study on the oscillatory dynamics of tip vortex cavitation, J. of Fluid Mechanics, 998, A13, 2024. doi:10.1017/jfm.2024.758
• Mallik, W., Jaiman, R.K., Jelovica, J. Deep neural network for learning wave scattering and interference of underwater acoustics, Physics of Fluids, 36, 017137, 2024.
• Venkateshwaran, A., Deo, I.K., Jelovica, J., Jaiman, R.K. A multi-objective optimization framework for reducing the impact of ship noise on marine mammals, Ocean Engineering, 310, Part 2, 118687, 2024.
• Rath, B., Mao, X., Jaiman, R.K. An efficient phase-field framework for contact dynamics between deformable solids in fluid flow, Computer Methods in Applied Mechanics and Engineering, 432, Part A, 117348, 2024.
• Darbhamulla, N.B., Jaiman, R.K. A finite element framework for fluid–structure interaction of turbulent cavitating flows with flexible structures, Computers & Fluids, 277, 106283, 2024.
• Gao, R., Deo, I.K., Jaiman, R.K. A Finite Element-Inspired Hypergraph Neural Network: Application to Fluid Dynamics Simulations, J. of Computational Physics, 504, 112866, 2024.
• Rath, B., Mao, X., and Jaiman R.K. An Interface Preserving and Residual-based Adaptivity for Phase-Field Modeling of Fully Eulerian Fluid-Structure Interaction, J. of Computational Physics, 488, 112188, 2023.
• Li, G. and Jaiman, R.K. “Unsteady aeroelastic characterization of flexible wings at high angles of attack”, AIAA Journal, 61 (11), 5042-5060, 2023.
• Mao, X. and Jaiman R.K. “An interface and geometry preserving phase-field method for fully Eulerian fluid-structure interaction”, J. of Computational Physics, 476, 111903, 2023.
• Kashyap, S. and Jaiman R.K. “Unsteady cavitation dynamics and frequency lock-in of a freely vibrating hydrofoil at high Reynolds number”, Int. J. of Multiphase Flow, 158, 104276, 2023.
• Heydari, S., Patankar, N.A., Hartmann, M.J.Z. and Jaiman, R.K. “Fluid-structure interaction of a flexible cantilever cylinder at low Reynolds numbers”, Physical Review Fluids, 7 (2), 024702, 2022.
• Chizfahm, A. and Jaiman R.K. “Deep Learning for Predicting Frequency Lock-in of a Freely Vibrating Sphere”, Physics of Fluids, 34, 127123, 2022.
• Deo, I.K., Jaiman, R.K., “Predicting waves in fluids with deep neural network”, Physics of Fluids, 34, 067108, 2022.
• Gupta, R., Jaiman, R.K. “Hybrid physics-based deep learning methodology for moving interface and fluid-structure interaction”, Computers and Fluids, 233, 105239, 2022.
• Mallik, W., Jaiman, R.K., Jelovica, J. “Predicting transmission loss in underwater acoustics using convolutional recurrent autoencoder network”, J. of the Acoustical Society of America, 152 (3), 1627-1638, 2022.
• Kashyap, S.R., Jaiman, R.K., “A robust and accurate finite element framework for cavitating flows with fluid-structure interaction”, Computers & Mathematics with Applications, 103, 19-39, 2021.
• Chizfahm, A. and Jaiman R.K. “Data-driven stability analysis and near-wake jet control for the vortex-induced vibration of a sphere”, Physics of Fluids, 33 (4), 044104, 2021.
• Mao, X., Joshi, V. and Jaiman R.K. “A variational interface-preserving Allen-Cahn phase-field formulation for surface-tension effect”, J. of Computational Physics, 433, 110166, 2021.

For a full list of publications, visit my profile on: