Open Access Open Access  Restricted Access Subscription or Fee Access

Simulations of a Turbulent Free Jet


(*) Corresponding author


Authors' affiliations


DOI: https://doi.org/10.15866/ireme.v8i5.602

Abstract


Neural networks have gained prominence as a viable means of expanding a finite data set of experimental measurements in recent times to improve the understanding of any complex phenomenon.
In this study, a generalized feed forward neural network is used to reconstruct experimentally obtained turbulence data of a turbulent free jet. The turbulence variables investigated are the axial and radial velocities, normal and shear stress, and higher order statistical moments in addition to turbulence budgets. The study revealed that the optimum Generalized Feed-forward Neural Network should be constructed with one hidden layer having 18 neurons which utilize a hyperbolic tangent activation function. The network is trained using the Levenberg-Marquardt learning algorithm. Validation studies are conducted to assess the merits of the proposed Neural Network. This investigation showed that artificial neural networks are suitable for predicting the characteristics of a turbulent free jet.
Copyright © 2014 Praise Worthy Prize - All rights reserved.

Keywords


Neural Network; Generalized Feed-forward; Turbulent Free Jet; LDV

Full Text:

PDF


References


P. Panigrahi, M. Dwivedi, V. Handelwal, and M. Sen, “Prediction of turbulence statistics behind a square cylinder using neural networks and fuzzy logic,” Journal of Fluids Engineering, ASME, vol. 125, pp. 385–387, March 2003.
http://dx.doi.org/10.1115/1.1537251

A. Spentzos, G. Barakos, K. Badcock, and B. Richards, “Modeling 3D dynamic stall using CFD and neural networks,” CFD Laboratory, University of Glasgow, vol. 44, no. 1071, 2006,
http://dx.doi.org/10.2514/6.2006-1071

M. Milano and P. Koumoutsakos, ”Neural network modeling for near wall turbulent flow,” Journal of Computational Physics, vol. 182, pp.1–26, 2002.
http://dx.doi.org/10.1006/jcph.2002.7146

J. Pruvost, J. Legrand, and P. Legentilhomme, “Three-Dimensional Swirl Flow Velocity-Field Reconstruction Using Neural Network with Radial Basis Functions,” Journal of Fluids Engineering: Transactions of the ASME, vol. 123, December 2001.
http://dx.doi.org/10.1115/1.1412847

Ahmed, S.A., El Kadi, H., Turbulence intensity predictions using neural network analysis, (2012) International Review of Mechanical Engineering (IREME), 6 (3), pp. 541-546.

S. Ahmed and H. El Kadi, “Dump Combustor Swirling Flow Reconstruction Using Neural Network,” 14th International Conference on Applied Mechanics and Mechanical Engineering AMME-14 conference AM-PT 038, Cairo, Egypt, May 25-27, 2010.

S.A. Ahmed, A.S. Nejad, and A.Z. Al–Garni, “A comparative study of the effects of LDV velocity bias in the near field of a turbulent free jet,” Canadian Aeronautics and Space Journal, vol. 41, pp. 179–184, December 1995.

S. Haykin, Neural Networks: A Comprehensive Foundation, 2nd ed. Pearson Education, 1999.

P. Haley and D. Soloway, “Extrapolation Limitations of Multilayer Feedforward Neural Networks,” in International Joint Conference on Neural Networks, vol. 4, Baltimore, USA, pp. 25–30, 1992.
http://dx.doi.org/10.1109/ijcnn.1992.227294

N.N.R. Ranga Suri and D. Deodhare, “Parallel Levenberg-Marquardt based neural network training on Linux clusters–A Case Study,” Centre for Artificial Intelligence & Robotics, Bangalore, India, 2010.


Refbacks

  • There are currently no refbacks.


Please send any question about this web site to info@praiseworthyprize.com
Copyright © 2005-2024 Praise Worthy Prize