Computer Communication & Collaboration

Computer Communication & Collaboration

ISSN:2292-1028 (Print)    ISSN:2292-1036(Online)

Vol. 6, Issue 2(2018.5)

Table of Contents

Editorial Board of CCC



1. An Artificial Neural Network Model for the Comprehensive Study of the Solidification Defects During the Continuous Casting of Steel[Download PDF]


Ishita Ghosh(Corresponding author), Nilratan Chakraborty


In this study, the prediction of some of the solidification defects during the continuous casting of steel alloy is put forward by employinga data driven multilayer perceptron (MLP) based neural network model. The inputs to this neural model are the various important processing parameters such as Aluminum percent, carbon drop percent in steel production, iron oxide percent in the sand mold, carbon percent, sulphur percent, fraction solid percent and critical temperature. Efforts have been done to minimize the network training error within few training cycles by optimizing the network training architecture using the Levenberg-Marquardt (LM) training algorithm. The characterization of the behavior of the various defects during the continuous casting of the steel alloy under the influence of various processing parameters is illustrated by the parametric sensitivity analysis. It has been observed that carbon drop percent during steel production and aluminum percent in the steel alloy have significant contribution in the formation of the shrinkage defect in steel alloy castings. The regression fit between the Artificial Neural Network (ANN) predictions and the target (measured) values of the output parameters demonstrates the appreciable concurrence of the results obtained.


Artificial Neural Network, Steelalloy, Shrinkage Volume, Burn on, Penetration Area