Multi-response surface characterization and process optimization in wire EDM of NiTi shape memory alloys

Abstract

Advanced materials produced through cutting-edge technologies play a crucial role in fulfilling the evolving demands of both consumers and industry. Consequently, material scientists actively engage in research focused on the production and enhancement of these materials' properties. Among these, nickel-titanium (NiTi) shape memory alloys stand out due to their exceptional characteristics, including high elastic deformation capability, superior strength, and excellent corrosion resistance. However, these same properties also render NiTi alloys difficult to machine using conventional techniques, often leading to significant tool wear and suboptimal surface quality. To address these challenges, wire electrical discharge machining (WEDM) has emerged as a more effective alternative to traditional machining methods for processing NiTi alloys. The primary objective of this study is to minimize material deformation by achieving the lowest possible surface roughness during the WEDM of NiTi shape memory alloys. These alloys are commonly utilized in high-performance sectors such as aerospace and defense, where precision and surface integrity are critical. Another goal is to optimize the machining parameters to enable accurate cutting without necessitating any additional finishing operations. In this study, gray relational analysis (GRA), a prominent multi-criteria decision-making method, was employed to optimize the WEDM process parameters. An experimental design based on the Taguchi L27 (35) orthogonal array was used to systematically investigate the effects of machining parameters. Analysis of variance (ANOVA) was then conducted to quantify the influence of each parameter. The findings indicate that the most significant control factors are: current for kerf width (73.67%), dielectric fluid flow rate for burr height (37.26%), servo voltage for machining time (59.24%), and current for surface roughness (61.16%). Furthermore, two-way interactions between control factors were also found to have a notable impact on machining outcomes. The optimization results were validated through confirmatory experiments, and the high correlation coefficients obtained support the reliability of the developed mathematical models.