Micro-machining of NiTi shape memory alloys through material-specific process optimization

The increased requirement for miniaturized and high-precision components in the biomedical and aerospace industries has spurred research in advanced micromanufacturing technologies. In view of the rising focus on sustainable practices, dry micromachining has become a viable option to lubrication-assisted procedures. Nonetheless, the machining of NiTi Shape Memory Alloys offers problems owing to their poor heat conductivity, noticeable work hardening propensity, and substantial adhesion at elevated temperatures. This research analyzes the dry micromilling efficacy of NiTi alloy and creates an intelligent prediction and optimization framework utilizing Artificial Neural Networks (ANN) and the JAYA algorithm. Thirteen experiments were done, and the ANN model anticipated surface roughness (SR) and burr width (BW) with mean absolute percentage errors of 3.99% and 0.11%, respectively. Parametric analysis found that spindle speed exerted the most significant influence on both reactions, preceded by feed rate and axial depth of cut. The JAYA algorithm was utilized to concurrently minimize surface roughness (SR) and burr width (BW), producing an optimal configuration of 35,000 RPM spindle speed, 0.5 mm/min feed rate, and 300 µm axial depth of cut (ADOC), resulting in a surface roughness of 0.245 µm and a bandwidth of 27.23 µm. The research indicates that persistent dry micromilling, together with intelligent optimization, can considerably increase surface integrity in NiTi components.