Authors: Aydin Salimiasl; Ahmet Özdemir
Addresses: Department of Mechanical Engineering, Payame Noor University, Iran ' Manufacturing Engineering Department, Faculty of Technology, University of Gazi, Ankara, Turkey
Abstract: In this paper, a neural network model (ANN) was created to predict the cutting forces in turning process for a new tool. A dynamometer was used to measure the static and dynamic cutting forces during the machining process. AISI 4140 steel was used as the work piece material due to its common application in machining industry. Cutting force, thrust force and radial force were measured for three combinations of cutting speeds (V), feed rates (f) and cutting depths (d). The tool angles were kept constant throughout the experiments. Full factorial method was used to design the experiments. For establishing the prediction model, a back propagation network (BPN) was developed with two layers and five neurons. Experimental results were compared with the predicted results of the neural network model (NN). The R2 values for training and test data were obtained 0.9992 and 0.9985 respectively.
Keywords: cutting forces; modelling; artificial neural networks; ANNs; estimation; turning; AISI 4140 steel; cutting force; thrust force; radial force; cutting speed; feed rate; cutting depth; tool angle; full factorial design.
International Journal of Mechatronics and Manufacturing Systems, 2016 Vol.9 No.2, pp.160 - 172
Received: 12 Jun 2015
Accepted: 07 Nov 2015
Published online: 26 Apr 2016 *