Solidification Simulations Coupled With Neural Networks for Prediction of Secondary Dendrite Arm Spacings in Sand Castings

D. Hanumantha Rao*, G.R.N. Tagore**, G. Ranga Janardhana***
*Associate professor of Mechanical Engineering, Deccan college of Engineering and technology, Hyderabad.
**Professor & Head, Dept. of Meeh. Engg. at National Institute of Technology, WARANGAL
***Professor of Mechanical Engineering, Jawaharlal Nehru Technological University, Hyderabad.
Periodicity:August - October'2006
DOI : https://doi.org/10.26634/jfet.2.1.970

Abstract

Systematic knowledge accumulation regarding casting process is essential in order to obtain optimal process conditions. A number of factors define the metallurgical structure in castings of which primary importance is secondary dendrite arm spacing. Controlling the rate of solidification can control dendrite structure. The objective of the present paper is to study the influence of casting process parameters like pouring temperature, insulation on riser, chill thickness and chill contact area on the microstructure parameter secondary dendrite arm spacing. A component of material aluminum alloy A356 is used in the present study. Neural networks are widely used in solving complicated mathematical problems whose analytical description is very difficult and computationally demanding. Design of experiments are used for experimental simulations using finite difference method. The developed neural network models are trained using the response obtained from solidification simulation. The predicted secondary dendrite arm spacing values from neural network are found to be in reasonable agreement with the simulation results and the effect of varied input values on output response is also obtained through sensitivity analysis.

Keywords

Solidification, Secondary Dendrite Arm Spacing, Finite Difference Method, Neural Networks and Sensitivity Analysis.

How to Cite this Article?

D. Hanumantha Rao, G.R.N. Tagore and G.Ranga Janardhana (2006). Solidification Simulations Coupled With Neural Networks for Prediction of Secondary Dendrite Arm Spacings in Sand Castings. i-manager’s Journal on Future Engineering and Technology, 2(1), 42-49. https://doi.org/10.26634/jfet.2.1.970

References

[1]. F.Mampaey 'Practical Examples Using the Finite Volume method for Simulating the Casting Process' AFS Transactions 01-075.
[2]. K.S.Lee, J.C.lin " Design of the runner and gating system paprameters for a multi-cavity injection muold using FEM and neural network" Inti. Journal Adv. Manuf. Tech. (2005).
[3]. M.Joseph Davidson et all " A neural network and reposne surface methodology based surface roughness prediction model for EDM"- Journal of Manufacturing technology today, July 2005.
[4]. Mahesh B. Parappagoudar et all 'Green Sand Mould System Modelling Through Design of Experiments' Indian Foundry Journal April 2005
[5]. Jukka Heikkonen and Jouko Lampinen, "Building Industrial Applications with Neural Networks" - Helsinki University of Technology, Laboratory of Computational Engineering Finland.
[6]. Eichlseder W.: 'Influence of Dendrite-Arm-Spacing on Fatigue Life1,4th FEMFAT User Meeting, 7th-9th May 2003, Steyr/Austria
[7]. XU Qingyan, LIU Baicheng 'Numerical Modeling of Dendrite Growth in Al Alloys'- Tsinghu science and technology-Vol. 9, Number5, Oct 2005.
[8].Merton C.Flemings' Solidification Processing'
[9]. John Campbell 'Castings'. Butterworth-Heinemann ltd
[10].Dale H.Besterfield et all 'Total Quality Management' Pearson Education
[11]. Jiju Antony 'Design of experiments for Engineers and Scientists' Butterworth Heinemann.
[12]. J.S.R.Jung, C.T.sun, E.Mizutnai "Neuro-Fuzzy and soft computing" Prentice hall of India.
[13]. User manuals of Virtual Casting.
[14], User manuals of Neuro Solutions.
If you have access to this article please login to view the article or kindly login to purchase the article

Purchase Instant Access

Single Article

North Americas,UK,
Middle East,Europe
India Rest of world
USD EUR INR USD-ROW
Online 15 15

Options for accessing this content:
  • If you would like institutional access to this content, please recommend the title to your librarian.
    Library Recommendation Form
  • If you already have i-manager's user account: Login above and proceed to purchase the article.
  • New Users: Please register, then proceed to purchase the article.