Surface Roughness Prediction Model Using Ann & Anfis

Now a days the general manufacturing problem can be described as the achievement of a predefined product quality with given equipment, cost and time constraints. There is a rapid development in the quality of advanced aero space materials like aluminum and its alloys with improved properties. The difficulties in machining of these materials economically and effectively are limiting their applications. The development of the new cutting tool materials is reaching an optimum level. Some quality characteristics of product such as surface roughness are hard to ensure and play an important factor in determining the quality of the product. Three cutting parameters viz., speed, feed, depth of cut are considered with constant nose radius. Experiments are carried out on aluminum alloy, AA 6351 and machined on Computer Numerical Control Lathe (CL 20 TL5) Turning Machine. Surface roughness of the machined piece was measured by using surface test stylus instrument with diamond tip and the effect of each cutting parameter over surface roughness was studied. Two models have been developed to predict the surface roughness. This paper utilizes two computational methods that is Adaptive-neuro fuzzy inference system (ANFIS), modeling and Artificial neural network (ANN) to predict surface roughness of work piece for variety of cutting conditions in hard turning. These models are developed in order to capture process specific parameters and predict surface roughness.