ANN BASED OPTIMIZATION MODEL FOR DIE CASTING OF ALUMINIUM ALLOY LM6

In the current research work the optimization of process parameters of an aluminium alloy diecasting operation is discussed. The major defect causing quality issue during the manufacturing of a die casted component was identified as porosity. An analysis of variance (ANOVA) is carried out to trace the parameters with significant effects on porosity. The plunger pressure used in the die casting machine the liquid aluminium temperature are identified as significant factors after ANOVA. Afterwards a BP-ANN is modelled and trained with the two significant factors and porosity in order to predict output by optimizing the two input process parameters. Further, the calculated values from a regression analysis are graphically compared with predicted ANN model in order to validate the ANN model.