Abstract—The cutting force of the aluminum workpiece was forecasted using the Artificial Neural Networks (ANNs) methodology in this study. Two ANN structures, one with a single hidden layer and the other with a double hidden layer, were constructed using MATLAB codes. The Levenberg- Marquardt back-propagation technique served as the training algorithm, employing a sigmoidal transfer function in the hidden layer and a purline transfer function in the output layer. The performance of the ANN models was assessed using Mean Squared Error (MSE) and coefficient of determination (R2). The experimental findings revealed that the cutting speed, feed rate, and depth of cut significantly influenced the cutting force. The optimal number of neurons in both single and double hidden layers was determined to be 6. The validation stage achieved the best performance with an MSE of approximately 0.002747 for a single layer and 0.00144 for double hidden layers, both at epoch 5. In conclusion, both ANN structures demonstrated the capability to predict cutting force, with a preference for the double hidden layer structure.

Keywords—depth of cut, feed, cutting force, artificial neural network 

Cite: Dawood S. Mahjoob, Ahmad A. Khalaf, and Muammel M. Hanon, "Forecasting Cutting Force by Using Artificial Neural Networks Based on Experiments of Turning Aluminum," International Journal of Mechanical Engineering and Robotics Research, Vol. 12, No. 6, pp. 410-416, November 2023.

Copyright © 2023 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.