An ANN-Based Method to Predict Surface Roughness in Turning Operations

ARAPOĞLU R. A., SOFUOĞLU M. A., ORAK S.

ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING, vol.42, no.5, pp.1929-1940, 2017 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 42 Issue: 5
  • Publication Date: 2017
  • Doi Number: 10.1007/s13369-016-2385-y
  • Journal Name: ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.1929-1940
  • Keywords: ANN, Prediction, Surface roughness, Hypothesis test, METAL-MATRIX COMPOSITES, ARTIFICIAL NEURAL-NETWORK, CUTTING PARAMETERS, STAINLESS-STEEL, OPTIMIZATION, SELECTION, DESIGN, FORCE, MODEL
  • Eskisehir Osmangazi University Affiliated: Yes

Abstract

In recent years, there has been a growing interest for the prediction of machining characteristics (such as surface roughness and tool wear) during machining. Several machining parameters such as cutting speed and cutting depth are known to affect the surface characteristics. Various methods are used to investigate the relative contribution of these parameters on the surface characteristics. Therefore, selecting a set of parameters according to the relative contributions is important in the prediction of the surface characteristics effectively. In this paper, a new alternative parameter selection method based on artificial neural networks is suggested. Within this scope, forward and stepwise selection methods are proposed. A statistical hypothesis test is used as an elimination criterion. The suggested methods are used to predict the surface roughness in turning operations effectively. Successful results were obtained in the prediction of surface roughness by using these methods.