Multi-objective parameter optimization of Ti6Al4V surface integrity for SEAM

This study addresses surface defects such as micro-cracks and phase change layer on the surface of Ti6Al4V workpieces after short electric arc machining (SEAM). The effects of process parameters on surface integrity (surface crack density (SCD), phase change layer thickness (PCLT), and surface roughness (Ra)) are analyzed using Taguchi L32 orthogonal experimental system. SCD and PCLT are estimated from scanning electron microscope (SEM) photographs of the workpiece surface based on the micro-crack area and cross-sectional phase change layer area, respectively, by ImageJ software. The combination of parameters under the optimal index is analyzed by the graph of the variation of the output index with the processing parameters, and analysis of variance (ANOVA) results show that voltage, feeding speed, and milling depth have the greatest effect on surface integrity. Finally, the 16 best sets of solutions satisfying the objective function are predicted using the non-dominated sorting genetic algorithm (NSGA II), which are obtained from the Pareto optimal bound, and the accuracy of the predicted solution sets is verified experimentally with an estimation error of less than 10%.