Sudeepta Mondal
Senior Research Engineer at Raytheon Technologies- Claim this Profile
Click to upgrade to our gold package
for the full feature experience.
Topline Score
Bio
Experience
-
Raytheon Technologies
-
United States
-
Aviation and Aerospace Component Manufacturing
-
700 & Above Employee
-
Senior Research Engineer
-
Jul 2021 - Present
Machine Learning Machine Learning
-
-
-
Argonne National Laboratory
-
United States
-
Research Services
-
700 & Above Employee
-
Postdoctoral Researcher
-
Jun 2020 - Jul 2021
Machine Learning for accelerating multi-physics flow simulations. Machine Learning for accelerating multi-physics flow simulations.
-
-
-
Penn State University
-
United States
-
Higher Education
-
700 & Above Employee
-
Research Assistant
-
Aug 2015 - Jun 2020
I work on the application of Machine Learning for anomaly detection and prediction in mechanical systems. • My primary application area is prediction and control of thermoacoustic instabilities. • Currently working on stochastic variational inference based learning of Hidden Markov Models for different regimes in combustion pressure time series data. • I also work on anomaly detection in fatigue loading of ductile polycrystalline alloys from ultrasonic sensor data. I work on the application of Machine Learning for anomaly detection and prediction in mechanical systems. • My primary application area is prediction and control of thermoacoustic instabilities. • Currently working on stochastic variational inference based learning of Hidden Markov Models for different regimes in combustion pressure time series data. • I also work on anomaly detection in fatigue loading of ductile polycrystalline alloys from ultrasonic sensor data.
-
-
-
Lawrence Livermore National Laboratory
-
United States
-
Research Services
-
700 & Above Employee
-
Data Science Summer Intern
-
May 2019 - Aug 2019
I worked on accelerating combustion simulations in internal combustion engines using machine learning. Achieved 2x speedup in computational time by optimally selecting the fastest solver under an acceptable error threshold. Particularly the work involved developing Bayesian neural networks for probabilistic predictions of wall times and errors for different solvers involved in the problem, and making informed decisions based on uncertainty estimates. I worked on accelerating combustion simulations in internal combustion engines using machine learning. Achieved 2x speedup in computational time by optimally selecting the fastest solver under an acceptable error threshold. Particularly the work involved developing Bayesian neural networks for probabilistic predictions of wall times and errors for different solvers involved in the problem, and making informed decisions based on uncertainty estimates.
-
-
-
United Technologies Research Center
-
United States
-
Research
-
100 - 200 Employee
-
Research Intern in Machine Learning
-
May 2018 - Aug 2018
I worked on interdisciplinary areas of machine learning and multi-physics modeling by collaborating with researchers from Physical Sciences and Thermal-Fluid Sciences: • Performed Multi-fidelity modeling and Bayesian Optimization for solving non-convex optimization problems real-time, that involved handling design and budget constraints. • The work also involved developing an extensive code-base in Python and integrating it with different simulation platforms for online design optimization.
-
-
-
Indian Institute of Science (IISc)
-
India
-
Research
-
700 & Above Employee
-
Summer Intern (Indian Academy of Sciences Fellowship) and Project Assistant
-
May 2014 - Aug 2015
I worked on a novel idea of controlling thermoacoustic instabilities by rotating the swirler in a combustor. • I was involved in the work from initial stages of the setup design and fabrication till the proof of concept demonstration of the idea. • The work has been published in Combustion and Flame and Journal of Propulsion and Power. The concept has been patented as a dynamic control strategy for mitigating thermo-acoustic instability in swirl stabilized lean-premixed turbulent combustors (Pub. No.: WO/2017/060819).
-
-
-
Indian Institute of Technology, Madras
-
India
-
Higher Education
-
700 & Above Employee
-
Summer Intern
-
May 2013 - Jul 2013
I worked on simulating droplet transport in a crossflow system using Quadrature Moment Method based CFD simulations, with practical implications of efficient combustion in liquid-fueled engines. The work has been published in conferences like ILASS, NCICEC, ASME GT-INDIA. I worked on simulating droplet transport in a crossflow system using Quadrature Moment Method based CFD simulations, with practical implications of efficient combustion in liquid-fueled engines. The work has been published in conferences like ILASS, NCICEC, ASME GT-INDIA.
-
-
Education
-
Penn State University
Ph.D., Mechanical Engineering -
Penn State University
Master of Arts - MA, Mathematics -
Jadavpur University
Bachelor of Engineering (B.E.), Mechanical Engineering