INVESTIGATION AND MODELING OF CUTTING TOOL TEMPERATURE IN TURNING OF INCONEL 625 STEEL BY USING TAGUCHI METHOD AND LONG SHORT TERM MEMORY NETWORK

In this paper, the results of an experimental investigation of cutting tool temperature in turning operations are presented. The cutting tool temperature is a very relevant in order to avoid excessive tool wear and tool damage. This study explains the impact of cutting speed, feed rate and depth of cut on cutting tool temperature during machining of Inconel 625 steel. Taguchi methodology is used to formulate the experimental layout. A Taguchi L27 design of experiment is applied to analyze the effect of each cutting parameter on the cutting tool temperature. Minitab software is used to process the experimental data and to develop a mathematical regression formula among three main cutting parameters and tool temperature. The obtained data in experimental work are used to perform deep learning using the Long Short-Term Memory (LSTM) network in order to predict the cutting tool temperature. The obtained results indicate that feed rate has the largest effect on cutting tool temperature followed by depth of cut and cutting speed. The predicted cutting tool temperature values obtained from LSTM network are very close to those of experimental, where the average percentage error from test base is 1.93%. According to the results, the LSTM predicts well the expected outcomes, and the standard deviation is in acceptable interval for predicting the cutting tool temperatures.