Effect of texture on acoustic emission produced by slip and twinning in AZ31B magnesium alloy—part II: clustering and neural network analysis

Abstract

The mechanical behaviour of extruded AZ31B magnesium alloy with two different textures was studied. Acoustic emission (AE) was used to record the signals to detect the various deformation processes occurring during the mechanical test. Hierarchical clustering and Kohonen's self-organising neural network map were used to process the data from AE and to develop a relationship in terms of the deformation modes. The signals produced from the two different deformation mechanisms (i.e. slip and twinning) were successfully identified. The developed method could be efficiently used for the classification of AE data.