1. Application of deep learning to study aggregative and non-aggregative nanofluid flow within the nozzle of a liquid rocket engine.
- Author
-
Muhammad, Noor, Ahmed, Naveed, Rani, Mehwish, and Mohsin, Bandar Bin
- Subjects
- *
ROCKET engines , *ARTIFICIAL neural networks , *DEEP learning , *AEROSPACE engineering , *BACK propagation , *SPRAY nozzles , *NANOFLUIDICS - Abstract
We have used Bayesian Regularization Back Propagation based deep neural networks to analyze the flow and heat transfer behavior for the flow of a nanofluid (aluminum-oxide in kerosene oil) within a regenerative cooling channel in a rocket engine. The significance of this research lies in its contribution to the advancement of aerospace engineering by addressing the complex flow dynamics of nanofluids in rocket propulsion systems. The aim of this study is to analyze the behavior of a mixture of kerosene oil and aggregated/non-aggregated aluminum oxide nanoparticles, considering thermal radiation and viscous dissipation effects. The research methodology involves transforming primary equations into a dimensionless form using a similarity transformation technique and employing the bvp4c numerical computation method in MATLAB. The ANN-IBR method is trained, tested, and validated using reference datasets obtained from numerical computations to approximate flow solutions under various scenarios with different physical parameters. The conclusion of this study indicates that the proposed ANN-IBR model effectively addresses the flow of kerosene and nanoparticles within the rocket engine, demonstrating its accuracy through comparison with reference outcomes based on mean square error analysis, transition states examination, histograms analysis, and regression analysis. This research provides valuable insights for aerospace engineers to enhance the design efficiency of regenerative cooling systems in liquid rocket engines. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF