A Fuzzy Model for Predicting Surface Roughness in Plasma Arc Cutting of AISI 4140 Steel

Abstract

In the present study, fuzzy logic (a tool in artificial intelligence) was used for the prediction of cutting parameters in plasma arc cutting (PAC) process of AISI 4140 steel. The parameters considered in this study were plasma arc current, cutting speed, and thickness of cut material. Fuzzy rule-based modeling was employed for prediction of surface roughness. These models can be effectively used to estimate the surface roughness. The experimental results were compared with fuzzy logic, and good agreement between them was observed. Analysis of the influence of the individual important machining parameters on the surface roughness have been carried out and presented. Statistically, cutting speed was found the most important factor on surface roughness, while the plasma arc current had the least. The characteristics of machined surfaces and microstructural changes are also studied.