Selection of Optimal Cutting Condition for Wear, Friction and Lubricant using Hybrid Intelligent System.

El-Hamshary, N. A.; El-Midany, Tawfik; Amar, Ibrahim Mohamed Ibrahim Eliwa; Gad El-Mawla, A. S.; B. Abouelatta, O.

doi:10.21608/bfemu.2020.126976

Selection of Optimal Cutting Condition for Wear, Friction and Lubricant using Hybrid Intelligent System.

Document Type : Research Studies

Authors

¹ Production Engineering and Mechanical Design Engineering Department., Faculty of Engineering., El-Mansoura University., Mansoura., Egypt.

² Professor of Production Engineering and Mechanical Design Engineering Department., Faculty of Engineering., El-Mansoura University., Mansoura., Egypt.

³ Professor of Production and Mechanical Design Engineering Department., Faculty of Engineering., El-Mansoura University., Mansoura., Egypt.

10.21608/bfemu.2020.126976

Abstract

Surface technology provides an important and valuable insight into the practical and theoretical applications of a manufactured surface. Cutting conditions play an important role in assessment of functional performance for engineering surfaces. Also, there is an important relation between the surfaces geometry and the functional performance of the surface. The geometry of the surface has a large influences of the surface performance. Topography analysis is primarily concerned with describing a surface in terms of its features. Then the knowledge gained about the geometry of the surface is used to control the surface production process to predict the performance of the component in its functional environment. Roughness parameters which characterize machined surfaces affect on both culling conditions and functional performance. Previous researches introduced a relations between roughness parameters and functional performance of machined surfaces. In present study, three of the surface functional performance that are friction ,wear, and lubrication which known as the tribological properties are concerned. A hybrid Artificial Neural Network (ANN) is used for selection of optimal cutting conditions. Experimental study is made to obtain the required parameters for the constructed ANN. The experimental study is made on 65 specimens free cutting steel 37, 40 mm in diameter. The specimens are divided into three groups, one of them is machined by turning operation. The second group is machined by milling operation. The third group is machined by grinding operation. All specimen are machined at various cutting conditions (feed, speed, depth of cut). All specimens are measured using Mitutoyo Surf Test SJ201 that give the surface profile and some of roughness parameters of the measured surface as a result. A developed Matlab programs Sur/Test SJ201P and SRCP, are used to give a full assessment of surface roughness parameters from the resulted surface profile of the Mitutoyo SurfTest-S/201. An introduced neural network is modeled by computer programs written in Matlab®. Also, OSCC program made by Matlab© is developed for selection of optimal cutting conditions. The maximum difference between measured data and data obtained from introduced hybrid ANN is of ±0.25%.

Keywords

Main Subjects