Surface roughness accuracy prediction in turning of Al7075 by adaptive neuro-fuzzy inference system

Abstract This research proposes the surface roughness inspection by an adaptive Neuro-fuzzy inference system. The adaptive Neuro-fuzzy inference system model developed by input parameters (Speed, Depth of cut, feed rate, and Grayscale value) and an output parameter (surface roughness). The training and testing module, which is used to generate the surface roughness value. The grayscale value derived from the machinability of the Al7075 workpiece. The machined workpiece image as converted as grayscale value, which is feed into one of the inputs of the adaptive Neuro-fuzzy inference system. The vision measurement value was compared with the stylus probe value for predicting the accuracy level of the adaptive neuro-fuzzy inference system. The accuracy was above 98%, which is helpful for inspecting all machined components in the mass production system.