In this paper, it has been proposed a viable optimal solution for achieving multi-objective optimization in powder mixed electrical discharge machining (PMEDM) process parameters, using titanium powder. Taguchi–AHP–Deng’s method is used to solve the multi-criteria decision making (MCDM) problem for this technology. The process parameters selected are workpiece material, electrode material, electrode polarity, pulse on time (T_on), pulse off time (T_of), current (I), and powder concentration. The material removal rate (MRR), tool wear rate (TWR), surface roughness (SR), hardness surface (HV), and white layer thickness (WLT) were determined simultaneously. The results showed that the optimal material for the workpiece is SKD11 die steel; the optimal material for the electrode is Gr; and the polarity of the electrode should be positive. The optimal values for the other process parameters were determined to be T_on = 20 μs, T_of = 57 μs, and I = 8 A, and that for the powder concentration was 10 g/l. The quality criteria for the optimal process parameters were determined to be MRR = 59.669 mm³/min, TWR = 14.073 mm³/min, SR = 4.45 μm, HV = 907.64 HV and WLT = 8.56 μm. The results were verified through experiments, which revealed a good accuracy (error ≈ 11.67%). The white-layer formation on the machined surface was reduced using titanium powder-based PMEDM for die steel at the optimal conditions. The results showed that Deng’s similarity method is the most suitable for MCDM in PMEDM using titanium powder.

在本文中，已经提出了一种可行的最佳解决方案，该解决方案可以使用钛粉在粉末混合放电加工（PMEDM）工艺参数中实现多目标优化。Taguchi–AHP–Deng的方法用于解决该技术的多标准决策（MCDM）问题。选择的工艺参数为工件材料，电极材料，电极极性，脉冲开启时间（T _on），脉冲关闭时间（T _of），电流（I）和粉末浓度。同时确定材料去除率（MRR），工具磨损率（TWR），表面粗糙度（SR），硬度表面（HV）和白层厚度（WLT）。结果表明，最适合工件的材料是SKD11模具钢。电极的最佳材料是Gr。电极的极性应为正极。被确定用于其它工艺参数的最优值为T_上 = 20微秒T，_的 = 57微秒，且i = 8 A，并且，对于粉末浓度为10克/升。确定最佳工艺参数的质量标准为：MRR = 59.669 mm ³ / min，TWR = 14.073 mm ³/ min，SR = 4.45μm，HV = 907.64 HV和WLT = 8.56μm。通过实验验证了结果，显示出良好的准确性（误差≈11.67％）。在最佳条件下，使用基于钛粉的PMEDM用于模具钢，可减少加工表面上的白层形成。结果表明，邓氏相似度法最适合使用钛粉的PMEDM中的MCDM。