A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Use of artificial neural networks to predict the properties of replicated open-cell aluminum alloy foam via processing parameters of melt squeezing procedure
Highlights An artificial neural network (ANN) methodology was employed in this study. In foams of similar porosities, the one includes larger cell has higher strength. Foams with finer cell size enjoy higher energy absorption capacity, regardless of their porosity.
Abstract Open-cell aluminum foams based on the cast A356 alloy were produced by the melt squeezing replication process using NaCl particles as space-holder material. The porosity and compressive mechanical properties of the produced foams, i.e. Young’s modulus, yield strength and energy absorption capacity, were measured under uniaxial compression condition. An artificial neural network (ANN) methodology was then employed to relate the foaming process parameters to the porosity and the compressive mechanical properties. The behaviors of the mechanical properties were predicted against cell size variation at various porosities, and the optimum condition for maximum strength–density ratio was determined. The prediction accuracy of the optimized ANN model was also compared with that of Gibson–Ashby equations. It was shown that there was a good agreement between the experimental and ANN predicted results with mean square error of 0.038; also, the developed ANN model with the optimal architecture of 2-3-6-4 enjoyed more accuracy in prediction of foam’s compressive properties in comparison with the Gibson–Ashby equations.
Use of artificial neural networks to predict the properties of replicated open-cell aluminum alloy foam via processing parameters of melt squeezing procedure
Highlights An artificial neural network (ANN) methodology was employed in this study. In foams of similar porosities, the one includes larger cell has higher strength. Foams with finer cell size enjoy higher energy absorption capacity, regardless of their porosity.
Abstract Open-cell aluminum foams based on the cast A356 alloy were produced by the melt squeezing replication process using NaCl particles as space-holder material. The porosity and compressive mechanical properties of the produced foams, i.e. Young’s modulus, yield strength and energy absorption capacity, were measured under uniaxial compression condition. An artificial neural network (ANN) methodology was then employed to relate the foaming process parameters to the porosity and the compressive mechanical properties. The behaviors of the mechanical properties were predicted against cell size variation at various porosities, and the optimum condition for maximum strength–density ratio was determined. The prediction accuracy of the optimized ANN model was also compared with that of Gibson–Ashby equations. It was shown that there was a good agreement between the experimental and ANN predicted results with mean square error of 0.038; also, the developed ANN model with the optimal architecture of 2-3-6-4 enjoyed more accuracy in prediction of foam’s compressive properties in comparison with the Gibson–Ashby equations.
Use of artificial neural networks to predict the properties of replicated open-cell aluminum alloy foam via processing parameters of melt squeezing procedure
Jamshidi-Alashti, Ramin (author) / Mohammadi Zahrani, Mohsen (author) / Niroumand, Behzad (author)
2013-05-10
10 pages
Article (Journal)
Electronic Resource
English
| British Library Online Contents | 2013
Producing replicated open-cell aluminum foams by a novel method of melt squeezing procedure
| British Library Online Contents | 2012
Semisolid melt squeezing procedure for production of open-cell Al-Si foams
| British Library Online Contents | 2014
Semisolid melt squeezing procedure for production of open-cell Al-Si foams
| British Library Online Contents | 2014