Indirect prediction of flank wear using ANNs in turning of CK45

This work aims to develop models to investigate the effects of flank wear on cutting forces during the turning of ck45 steel using carbide tools. Therefore, various turning experiments were performed with different cutting conditions. Flank wear and cutting forces were recorded at different stages of each experiment. The data obtained from the experiments showed that the tangential component of the cutting force was not significantly correlated with the tool flank wear. Instead, there was a good correlation between the axial and radial components of the cutting forces against flank wear. Since the cutting forces depend on both the cutting conditions and the tool flank wear, different cutting forces and cutting condition ratios were used to find the cutting force models that are more sensitive to tool wear. These ratios were used to develop artificial neural network models. The statistical results showed that the tool wear obtained from the artificial neural network models was very close to the results obtained from the experiments. In addition, the accuracy of the models including the axial component of cutting force was higher than in other models.