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 Ta b l e 3 Statistical-based models to predict cutting temperature Type of tool Statistical-based models R-squared Eq. no. Uncoated carbide T =169.0517V 0.3814 f 0.01144 0.98 (1) PVD -coated TiAlN carbide tool T =265.5113V 0.3074 f 0.0874 0.93 (2) PVD -coated TiN / TiAlN carbide tool T =299.4988V 0.2889 f 0.0897 0.99 (3) Fig. 4. Cutting temperature for ( a ) Uncoated, ( b ) TiAlN coated, ( c ) TiN / TiAlN coated tool a b c The surface plots are drawn for a better understanding of the effect of process parameters on the cutting temperature. The values of the coef fi cients involved in the statistical-based models for different tools were calculated by regression method using the Data- fi t software . The R -squared values for all the developed statistical-based models above 0.9 (Table 3) shows that developed models could be reliably used to predict the chip-tool interface temperature during the turning of SS304 for the given combination of tool and workpiece pair. In order to have a clear idea of the effect of the input parameters on the cutting temperature, three- dimensional (3-D) surface plots are constructed by changing the process parameters for uncoated and PVD -coated tools. The plots are constructed using developed empirical equations. Fig. 4 depicts the 3-D surface plots of cutting temperature during turning of SS304 for different tools plotted using Eqs. (1) to (3). It can be seen from Fig. 4, a–c that the cutting temperature increases with increasing cutting speed and feed. An increase in the cutting temperature can be seen as more prominent with the cutting speed followed by the feed. However, this effect can be seen as more signi fi cant with the uncoated carbide tool followed by TiAlN coated carbide tool. This can be also con fi rmed from the higher positive exponent value for the cutting speed followed by a feed from Eqs. (1) to (3).