PREDICTION OF CUTTING FORCE IN TURNING USING DESIGN OF EXPERIMENTS AND ARTIFICIAL NEURAL NETWORK

G.KONGURAJA; K.MANIKANDAN; K.KARTHICK

Volume : VIII, Issue : I, January - 2019

PREDICTION OF CUTTING FORCE IN TURNING USING DESIGN OF EXPERIMENTS AND ARTIFICIAL NEURAL NETWORK

G. Konguraja, K. Manikandan, K. Karthick

Abstract :

The cutting force has a significant influence on the dimensional accuracy because of tool and work piece deflection in turning. Force modeling in metal cutting is important for a multitude of purposes, including thermal analysis, tool life estimation, chatter prediction, tool condition monitoring, etc. Cutting force plays a vital role in turning operation. Required surface finish and dimensional accuracy is based only on cutting force. Experiments were conducted in all geared head centre lathe on machining C 45 steel specimens using Carbide tipped cutter using Design of Experiments based on central composite method. Results obtained from the experiment were used to predict the cutting force with the help of mathematical model developed using Quality America software. Using the measured force an Artificial Neural Network (ANN) model was developed using MATLAB 7.0 software. ANN model architecture consists of single hidden layer with 5 hidden neurons and trained by using feed forward back propagation algorithm. In addition, the results obtained by experimental procedure and from ANN model were compared to confirm that the developed ANN is more accurate in prediction