Dr. Arjun Kaithakkal

  • Institut für Strömungsmechanik
    Karlsruher Institut für Technologie
    Kaiserstraße 10, Geb. 10.23
    76131 Karlsruhe
    Tel.: +49 721 608 42368
    Fax: +49 721 608 45147

Project

Machine learning for the design of chemical engineering unit operations (https://chemengml.org/research-projects

Research Interest

  • Flow control  
  • Heat and mass transfer
  • Flow through porous media 
  • Adjoint optimization
  • Immersed boundary method

Journal Publications


2024
Machine learning for rapid discovery of laminar flow channel wall modifications that enhance heat transfer
Koide, Y.; Kaithakkal, A. J.; Schniewind, M.; Ladewig, B. P.; Stroh, A.; Friederich, P.
2024. APL Machine Learning, 2 (1), 016108. doi:10.1063/5.0187783

Conference Contributions


2024
Active learning based surrogate model for drag and heat transfer prediction in internal flows with eddy-promoters
Kaithakkal, A. J.; Koide, Y.; Müller, D.; Friederich, P.; Stroh, A.
2024, September 19. 1st European Fluid Dynamics Conference (EFDC1 2024), Aachen, Germany, September 16–20, 2024
Understanding CNN-Based Shape Optimization for Thermo-Hydraulic Efficiency Using Explainable Deep Learning
Kaithakkal, A. J.; Koide, Y.; Friederich, P.; Stroh, A.
2024, September 9. European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD 2024), Vilnius, Lithuania, September 9–13, 2024
2023
Heat transfer enhancement in laminar channel flow by machine learning guided shape optimization of wall geometry
Kaithakkal, A. J.; Koide, Y.; Schniewind, M.; Friederich, P.; Stroh, A.
2023. 17th International Heat Transfer Conference (IHTC 2023), Cape Town, South Africa, August 14–18, 2023

Supervised thesis projects

  1. Master project by Marion Cormier (2016): "Analyse turbulent flows via Empirical Mode Decomposition"
  2. Bachelor project byRachel See (2016) : "Measure skin-friction accurately in a new wind-tunnel facility"
  3. Master project by Steffen Straub (2015): "Direct numerical simulation of controlled turbulent duct flows"