Abstract
In this paper, an experimental plan for Taguchi method of experimental design of processing tungsten steel YG15 has been conducted according to Gaussian process regression (GPR) with combinatorial kernel functions. The aim is to develop a proper nonlinear model and seek optimal parameters on materials removal rate (MRR) and 3D surface quality (Sz and Sq) by integrated GPR and non-dominated sorting genetic algorithm-II (NSGA-II). By this method, it has been demonstrated that the method of integrated GPR and NSGA-II is an effective way for multi-objective optimization on 3D micron-scale surface topography in wire electrical discharge machine (WEDM).
ACKNOWLEDGMENTS
The authors, Wuyi Ming and Zhen Zhang, contributed equally to this work.
Notes
Sz, Surface peak-to-peak value (µm); Sq, Surface root-mean-square value (µm); MRR, Material removal rate .
Sz, Surface peak-to-peak value (µm); Sq, Surface root-mean-square value (µm); MRR, Material removal rate (mm2/min).
Sz, Surface peak-to-peak value (µm), Sq, Surface root-mean-square value (µm); MRR, Material removal rate (mm2/min).
Sz, Surface peak-to-peak value (µm); Sq, Surface root-mean-square value (µm); MRR, Material removal rate (mm2/min).
Sz, Surface peak-to-peak value (µm); Sq, Surface root-mean-square value (µm); MRR, Material removal rate (mm2 /min).
Sz, Surface peak-to-peak value (µm); Sq, Surface root-mean-square value (µm); MRR, Material removal rate (mm2/min).
Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/lmmp.