Multi-objective optimization of wire cutting process parameters for Ti-6Al-4V alloy is performed using NSGA-II and EWM-TOPSIS optimized artificial neural network

Due to its exceptional mechanical properties, the Ti-6Al-4V alloy is increasingly being utilized in various aerospace applications. Among the numerous processing methods available, wire electrical discharge machining (WEDM) stands out as one of the most efficient techniques for working with Ti-6Al-4V alloy. However, due to the presence of numerous processing parameters and complex influencing patterns, improper selection of processing parameters can significantly impact both the surface quality and machining efficiency. To address this challenge, a four-factor, three-level full factorial cutting experiment was conducted in this study. The influence trends of parameters such as pulse-on time ( T on ), pulse-off time ( T off ), peak current ( IP ), and open voltage ( OV ) on surface roughness, material removal rate, and kerf width were analyzed based on the experimental results. Artificial neural networks were employed to establish data-driven models, and their accuracy was subsequently validated. Additionally, the NSGA-II was employed for multi-objective optimization of computational results. Moreover, the EWM-TOPSIS method was utilized to effectively rank the solutions obtained from multi-objective optimization. The optimal parameters for the WEDM process were determined as follows: T on of 8.44 μs, T off of 5.12 μs, IP of 13.9 A, and OV of 209.44 V. The proposed method offers a more convenient and applicable approach for the selection of appropriate processing parameters.