SUMMARY
With modern production, Minimum Quantity Lubricant (MQL) technology has emerged as an alternative to conventional liquid cooling. The MQLs is an environmentally friendly lubricant method with low cost while meeting the requirements of machining conditions. In this study, the experimental and analytical results show that the obtained acoustic emission (AE) and vibration signal components can effectively monitor various circumstances in the SCM440 steel turning process with MQL, such as surface quality and chip formation as cutting tool conditions. The AE signals showed a significant response to the tool wear processes. In contrast, the vibration signal showed an excellent ability to reflect the surface roughness during turning with MQL. The chip formation process through the cutting mode parameters (cutting speed, feed and depth of cut) was detected through analysis amplitude of the vibration components Ax, Ay and Az and the AE signal. Finally, Gaussian process regression and adaptive neuro-fuzzy inference systems (GPR-ANFIS) algorithms were combined to predict the surface quality and tool wear parameters of the MQL turning process. Tool condition monitoring devices assist the operator in monitoring tool wear and surface quality limits, stopping the machine in case of imminent tool breakage or lower surface quality. With the unique combination of AE and vibration analysis model and the training and testing samples established by the experimental data, the corresponding average prediction accuracy is 97.57 %. The highest prediction error is not more than 3.8 %, with a confidence percentage of 98 %. The proposed model can be used in industry to predict surface roughness and wear of the tools directly during turning