An adaptive-network based fuzzy inference system for prediction of workpiece surface roughness in end milling

Abstract

An adaptive-network based fuzzy inference system (ANFIS) was used to predict the workpiece surface roughness after the end milling process. Three milling parameters that have a major impact on the surface roughness, including spindle speed, feed rate and depth of cut, were analyzed. Two different membership functions, triangular and trapezoidal, were adopted during the training process of ANFIS in this study in order to compare the prediction accuracy of surface roughness by the two membership functions. The predicted surface roughness values derived from ANFIS were compared with experimental data. The comparison indicates that the adoption of both triangular and trapezoidal membership functions in ANFIS achieved very satisfactory accuracy. When a triangular membership function was adopted, the prediction accuracy of ANFIS reached is as high as 96%.