The use of neural networks in predicting turning forces

Abstract The purpose of this research is to develop a predictive turning-force model based on neural networks. In the first stage of the research, a cutting-force model based on orthogonal machining theory is studied. Turning forces can be estimated from this model using complex computational procedures when a knowledge of the flow stress and thermal properties of the work material and the cutting conditions is given. In the second stage of the research, a feed-forward neural network is trained by the cutting-force model. After the training process is finished, the neural network becomes a knowledge-based turning-force system. Good correlation between the neural prediction and experimental verification of the turning forces is shown.