Estimating surface roughness for different EDM processing parameters on Inconel 718 using GEP and ANN

Abstract

Nickel-based alloys have good mechanical properties along with excellent corrosion and oxidation resistances. Inconel 718, which is in the nickel-based alloy group, has excellent fatigue strength, corrosion resistance, high temperature resistance, and also features such as high strength, toughness and deformation hardening. The biggest disadvantage of Ni alloys is their poor processability. Especially in the machining method, expensive tools should be used. In addition, high surface roughness values ​​obtained as a result of machining in such materials negatively affect the fatigue resistance. On the other hand, difficult-to-machine materials can be manufactured cheaply and easily by methods known as non-traditional production. The most widely used of these methods is the Electrical Discharge Machining (EDM) due to its precision and machining capabilities. The Inconel 718 material used in this study was etched by EDM method using different process parameters. The effects of finishing process parameters on surface roughness after scouring were determined. In the study, the probable surface roughness value estimation model was created by using the process parameters and the surface roughness values ​​obtained as a result of these parameters using the artificial intelligence methods ANN and GEP. For this study, it was concluded that the ANN method is more suitable than the GEP method.