Machinability analysis in wire-EDM of cryogenically treated Ti6Al4V alloy and multi-objective optimization using MOAVOA and MOGA
Authors :- M.K., Dikshit, V.K. Pathak
Publication :- Scientific Reports, (Springer-Nature) 2025.
The performance of Ti6Al4V alloy in engineering and biomedical applications is often limited by its poor wear and abrasion resistance, especially when compared to conventional materials such as CoCr-based alloys and stainless steels. In biomedical implants, such as hip and knee joints, this limitation results in a typical service life of only 10–15 years. Cryogenic treatment has emerged as a potential method to enhance the wear resistance of Ti6Al4V. However, machining this alloy using conventional methods remains challenging due to its low thermal conductivity, high cutting forces, and rapid tool wear, which lead to excessive heat generation and potential tool failure. Furthermore, its high electrical resistivity reduces its machinability using electrical discharge machining. This study investigates the machinability of cryogenically treated and untreated Ti6Al4V alloys using wire electrical discharge machining. A rotary central composite design, based on response surface methodology, was employed to develop quadratic models for material removal rate (MRR) and surface roughness (Ra), with discharge current (I), wire speed, and duty cycle (DC) as input parameters. Multi-objective optimization was carried out using a genetic algorithm (MOGA) and the African vultures optimization algorithm (MOAVOA) to simultaneously maximize MRR and minimize Ra. Results indicate that discharge current had the highest influence on MRR of cryogenically treated samples (MRR_CT) with a percentage contribution (PC) of 58.04%, followed by duty cycle at 20.28%. For surface roughness of cryogenically treated samples (Ra_CT), DC and I were the dominant linear terms with PCs of 11.91% and 7.83%, respectively, while current showed the highest influence in the square term with a PC of 34.44%. Optimization results demonstrated that MOAVOA outperformed MOGA in convergence speed and solution diversity, yielding a broader and more effective set of Pareto-optimal solutions.