Novel artificial neural network model for evaluating hardness of stir zone of submerge friction stir processed Al 6061-T6 plate

Abstract

The Al alloy 6061-T6 was friction stir processed at submerged condition and different tool rotation speeds ω and processing speeds V. The effect of processing parameters on hardness of stir zone was investigated. In order to derive out the relationship between the hardness of stir zone and processing parameters and optimising them, some tests were carried out and a matrix of variation parameters of process was filled and used for training of an artificial neural network (ANN) model. A sensitivity analysis was carried out using the ANN model. It is shown that, among the two process parameters, the processing speed V is more important on stir hardness. In addition, a safe zone can be defined by ANN model in which superior hardness can be achieved.