Experimental Investigation and Hybrid Metaheuristic Optimization Using ANN-MOJAYA on Corner Accuracy During WEDM for Ti-3Al-2.5V
摘要
Geometrical accuracy is an impactful parameter that pertains to the machined components utilized in recent equipment. The WEDM is well-known and demanding for geometrical accuracy, but it is not always met for difficult-to-machine materials. The geometrical deviation in the machined part can significantly impact the efficiency and lifespan of the finished parts. So, this experimental investigation consequently focused on identifying the optimum parameter setting to achieve the required quality of WEDMed Ti-3Al-2.5V end product. The machining performance was selected as cutting speed (Cs), surface roughness (SA), and corner geometrical deviation (CGD) during corner cutting during WEDM using copper-core-double layer ZnCu50 coating (“Topas plus X”) wire. Here, the design of the experiment has been chosen based on the Taguchi philosophy by varying the machining parameter (i.e., Spark-on Time (STon), Wire-speed (Sw), Servo-Voltage (Sv), Discharge Current (DC) and Wire Tension (WT)) to analyze the machining outcomes. The main effect plot and analysis of variance (ANOVA) framework were to analyze machining characteristics. Moreover, a hybrid parametric optimization is selected with Ortho-ANN-coupled with MOJaya algorithm to predict the necessary machining parameters setting to improve the machinability of Ti-3Al-2.5V. The most significant parameter observed is DC for Cs and CGD, which is determined to be 36.51% and 42.65%, respectively. However, the interaction term of DC and WT exhibited the highest contribution of 39.16% to the SA. Finally, the confirmation has been conducted and found the maximum absolute percentage errors are 9.2%, 12.5%, and 10.6% for Cs, SA, and CGD, respectively. This proposed optimum approach provides a promising solution for machining performance and significantly enhances the economic efficiency of production.