Dimensional analysis and ANN simulation of chip-tool interface temperature during turning SS304

OBRABOTKAMETALLOV MATERIAL SCIENCE Том 23 № 3 2021 EQUIPMEN . INSTRUM TS Vol. No. 4 2021 Fig. 7 . Neural network: a – Training; b – Validation; c – Test; d – All data set a b c d A comparative evaluation A comparative evaluation of the accuracy of the predicted results of chip-tool interface temperature with the statistical-based model ( SM ), dimensional analysis approach ( DA ), and arti fi cial neural network ( ANN ) is presented in this section. The accuracy of the different models is assessed by obtaining % error between the predicted and experimental values of chip-tool interface temperature at different cutting conditions. Table 6 depicts the results predicted by the developed chip-tool interface temperature models for different tools. Predicted results can be seen in good agreement with the experimental results (Table 2) with an absolute error of less than 5%. However, the results predicted by the ANN model are shown with a better agreement with the experimental results as compared to statistical-based and dimensional analysis models. It has been observed that the chip-tool interface temperature gets more affected with the cutting speed followed by the chip cross-sectional area and the speci fi c cutting pressure. With the increase in the cutting speed, the requirement of the cutting energy increases resulting in high cutting temperature. The thermal conductivity of the cutting tool also has a major in fl uence on the chip-tool interface temperature. Uncoated tool exhibited the lowest cutting temperature. This could be attributed to its higher thermal conductivity and large wear-out area of the tool during machining resulting in rapid dissipation of the interface heat into the tool. A lower cutting temperature observed for single-layer TiAlN coated tool than multi-layer TiN / TiAlN coated tool, could be attributed to its higher thermal conductivity than the equivalent thermal conductivity of TiN / TiAlN coated tool. The lower thermal conductivity of the TiN /Ti A lN coated tool resists heat conduction resulting in more temperature on the rake face. This coated tool also exhibited higher cutting temperature than AlTiCrN and AlTiN coated inserts [21]. However, a higher cutting temperature with a multi-layer tool helps to make the material being machined relatively soft and therefore, can help in improving the machining performance. However, the lower cutting force was observed with the multi-layer TiN / TiAlN