The IRMA Community
Newsletters
Research IRM
Click a keyword to search titles using our InfoSci-OnDemand powered search:
|
An Application of Deep Neural Network Using GNS for Solving Complex Fluid Dynamics Problems
|
|
Author(s): Pamir Roy (North Eastern Regional Institute of Science and Technology, India), S. K. Tamang (North Eastern Regional Institute of Science and Technology, India), Samar Das (Indian Institute of Technology, Guwahati, India)and Thanigaivelan Rajasekaran (AKT Memorial College of Engineering and Technology, India)
Copyright: 2024
Pages: 22
Source title:
Metaheuristics Algorithm and Optimization of Engineering and Complex Systems
Source Author(s)/Editor(s): Thanigaivelan R. (AKT Memorial College of Engineering and Technology, India), Suchithra M. (SRM Institute of Science and Technology, India), Kaliappan S. (KCG College of Technology, India)and Mothilal T. (KCG College of Technology, India)
DOI: 10.4018/979-8-3693-3314-3.ch001
Purchase
|
Abstract
The present work investigates the possibilities of solving complex fluid dynamics problems using Navier-Stokes equations, through simulation based techniques using deep neural networks in real time and along with provision of a singular architecture that achieves cutting-edge performance while maintaining a very high accuracy and precision at par with ground truth. The study employs Graph Network-based Simulators (GNS) to compute system dynamics. The developed model shows robust behavior in its prediction giving prediction accuracy of around 99%. The model generalizes well from unit-timestep predictions with huge number of particles at training phase, to completely differing starting conditions for timesteps ranging into the thousands and with even more particles at test time. Based on GNS, the model is immune to choices of hyper parameters over differing metrics for evaluation. The proposed model shows that deep learning is effective for solving a large set of complex fluid dynamics related problems in both forwards and backwards in time.
Related Content
|
Manoj Himmatrao Devare, Anita Manoj Devare, Nirali Verma.
© 2025.
24 pages.
|
|
N. Manjunathan, T. Venkata Ramana, A. Rajasekar, D. Vijayakumar, V. Sameswari, S. M. Nandha Gopal, R. Siva Subramanian.
© 2025.
30 pages.
|
|
J. Rajeshkumar, K. Aravindaraj, T. Uma Mageswari, S. Kerthy, R. Premkumar, S. Gayathri, R. Siva Subramanian.
© 2025.
24 pages.
|
|
J. Refonaa, M. Maheswari, D. Poornima, S. L. Jany Shabu, M. Gowri, S. Praveen, R. S. Amshavalli.
© 2025.
30 pages.
|
|
M. Gokuldhev, K. Vijayakumar, M. Mercy Theresa, K. Sudha, S. Nagarajan, R. Prasath, P. J. Beslin Pajila.
© 2025.
26 pages.
|
|
M. Ezhilvendan, Aniket Gangadharrao Patil, S. M. Sassirekha, A. Mathankumar, T. P. Anish, V. Sathya, P. Gajalakshmi.
© 2025.
32 pages.
|
|
D. Ravindran, G. Mariammal, S. Udhayashankar, K. Dhivya, D. Lekha, T. Maheshwaran, V. Sathya.
© 2025.
18 pages.
|
|
|