Full Length Research Paper

Derivation of statistical models to predict roughness parameters during machining process of PEEK composites using PCD and K10 tools

Francisco Mata¹*, Issam Hanafi², Abdellatif  Khamlichi², Abdallah Jabbouri³

and Mohammed Bezzazi³

¹Polytechnic School of Almadén, University of Castilla-La Mancha, Plaza Manuel Meca,1, 13412 Almadén, Spain.

²EMS2M, Faculty of Sciences at Tetouan, BP. 2121, M'hannech, Tetouan, Morocco.

³EMMS, Faculty of Sciences and technology at Tangier, BP 416, Tangier, Morocco

*Corresponding author. E-mail: francisco.mcabrera@uclm.es.

Accepted 08 June, 2009.

   Abstract

In many scientific fields, non-linear regression based models are of great utility to perform curve adjustment of experimental data. This concept is used in the present study in order to construct adequate adjusted models enabling to make predictions for the different roughness parameters characterizing machining of PEEK composites when using PCD and K10 tools. The adjusted data were obtained by using design of experimental methods and only the main factors affecting roughness during machining of PEEK composites were retained. Since, analysis of variance performed on experimental results has revealed that feed is the main cutting factor that influences surface roughness, nonlinear regression is conducted only in terms of this parameter.

Key words: Non-linear regression, machining process, PEEK composites, PCD tool, K10 tool, roughness parameter.