Pinaki Pal preview image

Pinaki Pal

Principal Research Scientist

Biography

Dr. Pinaki Pal is a Principal Research Scientist in the Advanced Propulsion and Power Department within Argonne’s Transportation and Power Systems (TAPS) division. He also has a joint appointment with the School of Mechanical Engineering at Purdue University. Dr. Pal’s research expertise and areas of interest span computational fluid dynamics (CFD), scientific machine learning and foundation modeling (SciML/SciFM), turbulent multiphase combustion, boundary layer flows, heat transfer, alternative fuels (hydrogen, ammonia, biofuels), extreme combustion events, high-order numerical methods, and HPC geared towards advancing modeling, design and optimization of propulsion/power systems (gas turbine engines, rotating detonation engines, piston engines) and manufacturing processes.

Dr. Pal leads multiple research efforts on developing advanced SciML algorithms and reduced-order modeling techniques to augment/accelerate CFD simulations of fluid flows and turbulent combustion, as well as engineering design optimization. His group pioneered the development of a first-of-its-kind physics-informed neural ordinary differential equations (NODEs) approach for accelerating stiff chemical kinetic computations (known as ChemNODE/Phy-ChemNODE) in reacting flow simulations by 10-1000x. On the other hand, novel active ML-driven design optimization algorithms, ML-GA and ActivO, co-developed by Pinaki have been widely leveraged by industry to accelerate virtual product design at low cost. Dr. Pal and his team won the 2021 R&D 100 Award for ML-GA and were recognized as 2023 R&D 100 Finalist for ActivO. Moreover, ML-GA was commercially licensed in 2021 by a U.S. startup company, Parallel Works Inc., and integrated within their commercial high-performance computing platform as part of a U.S. DOE Technology Commercialization Fund (TCF) project, thereby offering a powerful one-of-its-kind artificial intelligence-driven design optimization technology across a wide range of U.S. industries, such as automotive, aerospace, oil and gas, building energy and materials, etc. This technology demonstration and commercialization effort won the 2022 Federal Laboratory Consortium (FLC) National Award for Excellence in Technology Transfer. 

Currently, Dr. Pal is leading a multi-lab team under the DOE Genesis Mission, known as COMB-FLOW-AI, with the ultimate goal of dramatically accelerating discovery science and engineering for novel fuel combustion technologies. This effort is focused on developing advanced scientific AI foundation models and agentic AI frameworks to enable fast, predictive digital twins for multiscale flow and combustion systems and orchestrate end-to-end design and optimization workflows. Other ongoing SciML research led by Dr. Pal includes data-driven subgrid wall models for large-eddy simulations of turbulent boundary layer flows, generative AI for novel fuel design, deep learning approaches for uncertainty quantification (UQ) and multi-fidelity surrogate modeling, and multi-scale graph neural network (GNN) and graph transformer frameworks for surrogate modeling of fluid flows.

Dr. Pal also leads the R&D effort at Argonne pertaining to computational modeling and design of full-scale rotating detonation engines (RDEs), which are promising candidates for stationary power generation and hypersonic propulsion. His work pioneered the application of chemical explosive mode analysis (CEMA) technique for combustion mode detection and quantification of non-ideal deflagrative losses in RDEs. On the other hand, he is leading turbulent combustion modeling efforts for predictive CFD simulations of hydrogen/ammonia-fueled stationary gas turbines, with a particular focus on extreme events (flame flashback, lean blow-out) and NOx emissions. Dr. Pal received the Impact Argonne Award in 2019, 2021-2023, and 2025 for his computational research on RDEs, gas turbine engines, SciML, and hypersonic propulsion. In 2022-2023, he was selected to participate in the Argonne Launchpad Program which is designed to provide motivated early- and mid-career researchers with enhanced training and mentoring for developing multimillion-dollar sponsored research programs. As part of the Launchpad Program, he led Argonne’s lab-wide effort to develop a research program centered around advancing end-use applications of alternative fuels (hydrogen, ammonia) in advanced gas turbines and rotating detonation engines for the stationary power generation sector. 

Lastly, Dr. Pal developed a turbulent combustion modeling framework to predict knocking combustion in reciprocating engines and extended the state-of-the-art engine CFD models by incorporating fuel composition effects on turbulent flame propagation. He has led multiple numerical studies utilizing this combustion modeling approach in a first-of-its-kind virtual cooperative fuel research (CFR) engine configuration to investigate the impact of gasoline-biofuel blending on knock-limited performance of boosted spark-ignition (SI) combustion and autoignition propensity in advanced compression ignition (ACI) mode. These research works won the ASME International Combustion Engine Division Best Paper Award (in Fuels category) in 2020 and 2021. Another research paper co-authored by Dr. Pal, published in Physics of Fluids (in 2020) and focusing on high-fidelity numerical modeling and proper orthogonal decomposition (POD) analysis of noise, vibration, and harshness (NVH) in compression-ignition engines was selected as an Editor’s Pick by the journal on account of its novelty and noteworthy contribution. 

Education

  • PhD, Mechanical Engineering, University of Michigan - Ann Arbor, USA (2015)
  • MS, Mechanical Engineering, University of Michigan - Ann Arbor, USA (2012)
  • Bachelor of Technology (B. Tech), Mechanical Engineering, Indian Institute of Technology (IIT) Kharagpur, India (2011)

Active Research Areas of Interest

  1. Probabilistic AI foundation models and multi-agent AI frameworks for Combustion and Fluid Flows
  2. High-fidelity CFD modeling of combustion dynamics in rotating detonation engines (RDEs) and RDE-turbomachinery systems for power generation and propulsion applications
  3. Physics-informed neural ODEs (Phy-ChemNODE) for accelerating stiff chemistry computations in multiscale reacting flow simulations
  4. Multiscale graph neural network (GNN) and graph transformer based SciML frameworks to accelerate high-fidelity CFD simulations of fluid flows
  5. Turbulent combustion modeling for predictive CFD simulations of extreme events (such as, flame flashback, lean blowout, etc.) and NOx emissions in hydrogen/ammonia-fueled gas turbine engines
  6. High-order CFD simulations of gas turbines (combustors & film cooling systems) and RDEs using Nek5000/NekRS spectral element solver
  7. Physics-guided data-driven near-wall turbulence modeling for gas turbine film cooling flows  
  8. Generative deep learning for inverse molecular design of fuels
  9. Active machine learning algorithms for rapid design optimization of energy systems and manufacturing processes
  10. CFD-driven numerical investigation of fuel-engine interactions; co-optimization of advanced reciprocating engines and fuels
  11. Causal analysis of combustion instabilities in propulsion systems using high-fidelity CFD simulations and SciML
  12. Integration of non-equilibrium wall models and multi-regime combustion models into high-order spectral-element CFD solver, Nek5000, for multi-mode engine simulations

Google Scholar: https://​schol​ar​.google​.com/​c​i​t​a​t​i​o​n​s​?​u​s​e​r​=​5​z​p​s​l​S​k​A​A​A​A​J​&​hl=en

Awards & Honors

  • Invitee to the US National Academy of Engineering (NAE) Frontiers of Engineering (FOE) Symposium (2026)
  • SAE John Johnson Paper Award for Outstanding Research in Diesel Engines (2026)
  • AIAA Associate Fellow (Class of 2026)
  • Early Career Award, ASME Internal Combustion Engine Division (ICED) (2025)
  • HPCwire Readers’ Choice Award, Best Use of High Performance Data Analytics & AI (2025)
  • Impact Argonne Award for Enhancement of Argonne’s Reputation, Argonne National Laboratory (2025)
  • HPCwire Readers’ Choice Award, Best Use of HPC in Industry (2024)
  • HPCwire Readers’ Choice Award, Best Use of High Performance Data Analytics & AI (2024)
  • ASME Internal Combustion Engine Division (ICED) Award – Valued services in advancing engineering profession as a session organizer (Modeling and Simulation) (2024)
  • HPCwire Readers’ and Editors’ Choice Awards, Best HPC Collaboration (2023)
  • R&D 100 Award Finalist, ActivO (2023)
  • Impact Argonne Award for Innovation, Argonne National Laboratory (2023)
  • Best Paper Nomination (Pressure Gain Combustion Track), AIAA (2022)
  • HPCwire Readers’ Choice Award, Best Use of HPC in Industry (2022)
  • Cohort 7 Selectee of Launchpad Program, Argonne National Laboratory (2022-2023)
  • Impact Argonne Award for Program Development, Argonne National Laboratory (2022)
  • Federal Laboratory Consortium (FLC) National Award for Excellence in Technology Transfer (2022)
  • HPCwire Readers’ Choice Award, Best Use of High Performance Data Analytics & AI (2021)
  • R&D 100 Award Winner, ML-GA (2021)
  • Best Paper Award (Fuels Track), ASME Internal Combustion Engine Division (2021)
  • Federal Laboratory Consortium (FLC) Excellence in Technology Transfer Award, Midwest Region (2021)
  • Impact Argonne Award for Innovation, Argonne National Laboratory (2021)
  • Best Paper Award (Fuels Track), ASME Internal Combustion Engine Division (2020)
  • Editor’s Pick, Physics of Fluids Journal (2020)
  • HPCwire Editors’ Choice Award, Best Use of HPC in Automotive (2019)
  • Pacesetter Award for Program Development, Argonne National Laboratory (2019)
  • 1st place in ​‘Fluid Dynamics, Acoustics, and Thermal Science’ technical session, Engineering Graduate Symposium (EGS), University of Michigan (2015)
  • Mechanical Engineering Graduate Fellowship, University of Michigan (2011-2012)
  • Institute Silver Medal, Rank 1 in the Department of Mechanical Engineering (B. Tech), Indian Institute of Technology (IIT) Kharagpur (2011)

Patent & Invention Disclosure

  • O. Owoyele and P. Pal, ​“Active optimization approach for rapid and efficient design space exploration using ensemble machine learning (ActivO)”, Patent awarded on Jan 7, 2025. 

Professional Activities