Preliminary communication

Modeling and predicting abrasive wear behaviour of poly oxy methylenes using response surface methodolgy and neural networks

A. Sagbas ; Tarsus Technical Education Faculty, Mersin University, Tarsus, Turkey
F. Kahraman ; Tarsus Technical Education Faculty, Mersin University, Tarsus, Turkey
U. Esme ; Tarsus Technical Education Faculty, Mersin University, Tarsus, Turkey

Abstract

In this study, abrasive wear behaviour of poly oxy methylenes (POM) under various testing conditions was investigated. A central composite design (CCD) was used to describe response and to estimate the parameters in the model. Response surface methodology (RSM) was adopted to obtain an empirical model of wear loss as a function of applied load and sliding distance. Also, a neural network (NN) model was developed for the prediction and testing of the results. Finally, a comparison was made between the results obtained from RSM and NN.

Keywords

abrasive wear; poly oxy methylene; neural network; responce surface methodology

Hrčak ID:

32000

URI

https://hrcak.srce.hr/32000

Publication date:

1.4.2009.

Article data in other languages: croatian

Visits: 3.067 *