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

OBRABOTKAMETALLOV Vol. 23 No. 3 2021 MATERIAL SCIENCE EQUIPMENT. INSTRUMENTS 4 The lowest cutting temperature observed for the uncoated carbide tool shows that the maximum heat was penetrated in the cutting tool from the rake surface. However, the higher interface temperature observed with coated tools shows that the coatings contributed to less heat penetration into the base of the tool. On the other hand, amongst the coated tools, the lowest chip-tool interface temperature was observed with the PVD -coated TiAlN carbide tool than the TiN / TiAlN coated tool. This can be also con fi rmed from the higher thermal conductivity value for the uncoated carbide tool followed by the TiAlN coating and TiN / TiAlN coating. Thermal conductivity of uncoated tool, PVD -coated TiAlN and TiN / TiAlN coated tools are 80 W/m-K, 6.7 W/m-K, and 5.1 W/m-K, respectively [12–14]. The thermal conductivity of the uncoated carbide tool is more compared to the TiAlN and TiN / TiAlN coated tools. Despite the fact that the thermal conductivity increases with temperature, at high temperatures the thermal conductivity of the coated tool remains lower than that of the uncoated tool [12]. Therefore, the heat conducted in the tool during machining with the uncoated carbide tool is more compared to that of the TiAlN and TiN / TiAlN coated tools. Hence, the temperature for the uncoated carbide tool is less than that of the coated tools. These results are matching well with those reported by Grzesik [4, 13, 15]. The increase in cutting temperature prominently with the cutting speed could be attributed to an increase in the speci fi c cutting energy. The speci fi c cutting energy can be partitioned into two main components shear energy and frictional energy. Shear energy and frictional energy are directly proportional to the shear velocity and chip velocity respectively [16]. Therefore, an increase in the cutting speed re fl ects directly into the increase in the energy and hence the cutting temperature. In addition, it has been observed that the coating structure greatly affects the cutting temperature. Moreover, it has been noted that uncoated tools wear out quickly compared to coated tools, which increase the chip contact area with the tool, which leads to greater heat conduction to the tool area. While in the case of coated tools, the higher wear resistance of the coatings limits the wear and, consequently, the chip contact area with the tool and allows more heat to be removed with the current chips. The higher thermal conductivity of uncoated tools decreases its hot hardness, which results in earlier failure of the tools [17]. The problems of temperature measurement have led to the research interest in the development of mathematical models for predicting temperature during machining. Suf fi cient studies attempted to predict the cutting temperature using statistical-based models. The mathematical models developed by Boothroyd, Shaw, and Rapier have been also extensively used by researchers to predict the cutting temperature. In this section, simulation using dimensional analysis and arti fi cial neural network to predict a chip-tool interface temperature with uncoated and PVD single-layer TiAlN and multi-layer TiN / TiAlN coated carbide tools are discussed. Dimensional analysis of cutting temperature In dimensional analysis, all independent variables of the problem are written down in the form of its dimensionless combinations. These independent dimensionless variables can be determined based on prior knowledge, reasoning, or experiments. The values of constants are obtained from experimental data [18-19]. In the present work, dimensional analysis is done to develop a mathematical model for obtaining the cutting temperature during the turning of SS304 steel using uncoated and TiAlN coated tools. These relations are developed based on the experimental data. The physical quantities selected for the dimensional analysis are given in Table 4. Physical quantities are expressed in such fundamental units as Mass (M), Length (L), Time (T), and Celsius temperature (D). This is the important step in which the most in fl uencing variables that affect the cutting temperature should be selected. It was assumed that about 80…85 % of the heat is dissipated to- gether with the chips, and, therefore, the thermal conductivity of the tool is not included in present analysis. The variables selected for the analysis are given below in Table 4. The number of fundamental quantities is four and the number of physical quantities selected in the present study is six. According to Buckingham Pi Theorem, the number of dimensionless groups required to correlate all these quantities would be equal to the difference between the number of physical quantities and the fundamental quantities which is two in the present study.