Title: Neural network modelling of Abbott-Firestone roughness parameters in honing processes

Authors: Maurici Sivatte-Adroer; Irene Buj-Corral; Xavier Llanas-Parra

Addresses: Department of Mechanical Engineering, School of Engineering of Vilanova i la Geltrú (EPSEVG), Universitat Politècnica de Catalunya, Av. Víctor Balaguer, 1. 08800 Vilanova i la Geltrú, Barcelona, Spain ' Department of Mechanical Engineering, School of Engineering of Barcelona (ETSEIB), Universitat Politècnica de Catalunya, Av. Diagonal 647. 08028, Barcelona, Spain ' Automatic Control Department, School of Engineering of Vilanova i la Geltrú (EPSEVG), Universitat Politècnica de Catalunya, Av. Víctor Balaguer, 1. 08800 Vilanova i la Geltrú, Barcelona, Spain

Abstract: In present study, three roughness parameters defined in the Abbott-Firestone or bearing area curve, Rk, Rpk and Rvk, were modelled for rough honing processes by means of artificial neural networks (ANN). Input variables were grain size and density of abrasive, pressure of abrasive stones on the workpiece's surface, tangential or rotation speed of the workpiece and linear speed of the honing head. Two strategies were considered, either use of one network for modelling the three parameters at the same time or use of three networks, one for each parameter. Overall best neural network consists of three networks, one for each roughness parameter, with one hidden layer having 25, nine and five neurons for Rk, Rpk and Rvk respectively. However, use of one network for the three roughness parameters would allow addressing an indirect model. In this case, best solution corresponds to two hidden layers having 26 and 11 neurons.

Keywords: surface roughness; honing; artificial neural networks; ANN; backpropagation algorithm; Abbott-Firestone roughness parameters; back propagation algorithm; grain size; density of abrasive; linear speed; tangential speed; pressure.

DOI: 10.1504/IJSURFSE.2017.088973

International Journal of Surface Science and Engineering, 2017 Vol.11 No.6, pp.512 - 530

Received: 16 Dec 2016
Accepted: 21 May 2017

Published online: 29 Dec 2017 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article