Back propagation genetic and recurrent neural network applicationsin modelling and analysis of squeeze casting process

Patel, Manjunath G C and Shettigar, Arun Kumar and Krishna, Prasad and Parappagoudar, Mahesh B. (2017) Back propagation genetic and recurrent neural network applicationsin modelling and analysis of squeeze casting process. Applied Soft Computing, 59. pp. 418-437. ISSN 1568-4946

[img] PDF
2487.pdf - Published Version
Restricted to Registered users only

Download (6MB) | Request a copy

Abstract

Today, in competitive manufacturing environment reducing casting defects with improved mechanicalproperties is of industrial relevance. This led the present work to deal with developing the input-outputrelationship in squeeze casting process utilizing the neural network based forward and reverse mapping.Forward mapping is aimed to predict the casting quality (such as density, hardness and secondary den-drite arm spacing) for the known combination of casting variables (that is, squeeze pressure, pressureduration, die and pouring temperature). Conversely, attempt is also made to determine the appropriateset of casting variables for the required casting quality (that is, reverse mapping). Forward and reversemapping tasks are carried out utilizing back propagation, recurrent and genetic algorithm tuned neuralnetworks. Parameter study has been conducted to adjust and optimize the neural network parametersutilizing the batch mode of training. Since, batch mode of training requires huge data, the training datais generated artificially using response equations. Furthermore, neural network prediction performancesare compared among themselves (reverse mapping) and with those of statistical regression models (for-ward mapping) with the help of test cases. The results shown all developed neural network models inboth forward and reverse mappings are capable of making effective predictions. The results obtained willhelp the foundry personnel to automate and précised control of squeeze casting process.

Item Type: Article
Uncontrolled Keywords: Squeeze casting process; Genetic algorithm neural network (GA-NN); Back-propagation neural network (
Subjects: Engineering > MIT Manipal > Mechatronics
Depositing User: MIT Library
Date Deposited: 13 Jul 2017 08:56
Last Modified: 13 Jul 2017 08:56
URI: http://eprints.manipal.edu/id/eprint/149335

Actions (login required)

View Item View Item