Machine Learning Based Cutting Force Prediction in Thin-Wall Turning Operation

Many complex components may contain thin-walled sections which are mainly produced by machining operation. These thin-walled sections are likely to deform during machining due to cutting forces resulting into form error in the final component. During machining, cutting force is the major controlling parameter determining the part quality, productivity and energy consumption. The current study mainly aims to estimate longitudinal, tangential and radial force components in thin-wall turning through machine learning algorithms. The required data for the development of the predictive models are obtained through turning investigations conducted for various cutting conditions. Results acquired by the predictive models are compared with the experimental ones over a wide range of cutting conditions. The present work will be helpful to predict the form error of thin-walled turned components. The expense of conducting multiple machining operation will be reduced with the help of such force model by predicting various force components, which will help in lowering the energy consumption and decreasing carbon foot prints.