Forward and Reverse Mappings in die Casting Process by Neural Network-based Approaches

Abstract

Abstract The primary objective of this research is to develop an effective process model to map process parameters on quality characteristics of a commercial die-cast part of aluminium alloy by utilizing Back Propagation and Genetic Neural Networks. In the neural network based forward mapping, die cast component properties have been expressed as the functions of input parameters, whereas attempts are made to determine an appropriate set of input parameters, to ensure a set of desired properties, in reverse mapping. In the present work, the problems related to both the forward as well as reverse mappings are tackled by using a back-propagation neural network (BPNN) and a genetic-neural network (GA-NN). The forward modeling consists of predicting the responses for a set of given process parameters, while reverse model provides the information process parameters for a set of desired responses. Batch mode of training is provided to both the networks with the help of one thousand data, generated artificially from the regression equations obtained earlier by the authors. The performances of both forward and reverse models based on BPNN have been compared with the help of twenty randomly generated test cases. The results show that, BPNN outperforms the GA-NN approach and that both the NN approaches are able to carry out the reverse mappings effectively.