Determination of tool tip steady-state temperature in dry turning process based on artificial neural network. (July 2022)

• Record Type: Journal Article
• Authors: Zhou, Guo; Xu, Chao; Wang, Xiaohao; Feng, Pingfa; Zhang, Min
• Is Part Of: Journal of manufacturing processes. Volume 79(2022)
• Page Start: 600
• Page End: 613
• Publication Date: 2022-07
• Subjects: Cutting temperature -- Measurement -- Turning -- Neural network -- Finite element model; Production management -- Data processing -- Periodicals; Manufacturing processes -- Periodicals; Procestechnologie; Productietechniek; Production -- Gestion -- Informatique -- Périodiques; Fabrication -- Périodiques; Manufacturing processes; Production management -- Data processing; Periodicals; 670.5
• DOI: 10.1016/j.jmapro.2022.05.021
• Languages: English
• ISSNs: 1526-6125

Abstract

Abstract: Effective and accurate measurement of cutting temperature of the tool tip is of great significance in the machining fields. However, the current measuring methods, including the embedded thermocouple and thermography, have some inevitable shortages. For example, the hole for displacing the embedded thermocouple damages the cutting edge strength, and the chip blocks the view contact between the tool tip and the thermal imager. This paper proposes two feed-forward multi-layered perceptron artificial neural network (ANN) models with Levenberg-Marquardt backpropagation training algorithms to determine the tool steady-state tip temperature for the dry turning process. The first model (called TC_ANN) calculates the steady-state tip temperature based on the temperatures of four specific points on the rake face. The second model (called TP_ANN) predicts the steady-state tip temperature via the cutting parameters. First, a fully thermo-mechanical finite element simulation model of the dry turning process is established. Next, the simulation model is repeated with different combinations of cutting parameters for sufficient times. Then, collection of the steady-state temperatures of the tool tip and these four specific points, and corresponding cutting parameters for each simulation model is conducted as the dataset to train these two ANN models. Based on the evaluation results with experimental data, the simulation model is validated to be reliable, and the mean relative errors of the tip temperature calculation for the actual turning process via TC_ANN and TP_ANN are 3.25% and 4.03%, respectively. Finally, the influences of cutting parameters on the tool tip temperature are investigated based on TP_ANN. It is found that the cutting speed and depth of cut have the most and least significant effects on the tip temperature, respectively, and the feed rate has almost the same influence as the depth of cut.