Shape Memory Alloys (SMAs) are unique class of modern material with various functional properties such as pseudo-elasticity, biocompatibility, high specific strength, high corrosion resistance and high anti-fatigue property. The machining of these SMAs is difficult by conventional machining processes due to strain-hardening effect which changes the properties of materials and it is found that non-conventional machining processes are more suitable to machine them. In present investigations, the experiments were performed on wire electrical discharge machining (WEDM) to study the interaction effects of the process parameter on surface characteristics of Ni49Ti51SMA’s by brass tool electrode. Peak current, pulse on time, pulse off time, wire tension and wire feed rate were taken as input parameters and their effect were analysed on material removal rate. Artificial neural network was adopted to develop and to train the experimental data using back-propagation neural network (BPNN) approach. The response surface methodology (RSM) was adopted to develop the second order mathematical based quadratic models. The recast layer formation and surface of machined materials were also analysed by the SEM characterization. It was noticed that the machined surface contains the surface cracks and uneven distribution of crater on the surface.
展开▼