Title: Modelling frictional noise using artificial neural network and regression: the case of steel on steel reciprocating sliding

Authors: Mir Mohsin John; M. Hanief

Addresses: Tribology Laboratory, Department of Mechanical Engineering, National Institute of Technology, Srinagar, Kashmir, India ' Tribology Laboratory, Department of Mechanical Engineering, National Institute of Technology, Srinagar, Kashmir, India

Abstract: The present work investigates the influence of the operating parameters on the frictional noise during reciprocating sliding of steel on steel using a ball-on-disc configuration. All the experiments were carried out in an anechoic chamber to mitigate the effect of surrounding noise to arrive at reliable frictional noise measurements during sliding experiments. The experimental outcomes reveal that the frictional noise increases with increase in surface roughness, frequency and the applied load. The regression and artificial neural network models were developed using the data generated by experimentation. The artificial neural network model predicts the frictional noise more accurately and closely to the experimental results as compared to the regression models. It has been found that the MSE, MAPE and R² for the ANN model are 0.001100, 1.35 and 0.98385 respectively, while the corresponding values for the 2nd order regression model are 0.0661, 4.07 and 0.8995, respectively.

Keywords: frictional noise; surface roughness; modelling; regression; artificial neural network; ANN; contour interpolation.

DOI: 10.1504/IJMATEI.2023.130131

International Journal of Materials Engineering Innovation, 2023 Vol.14 No.2, pp.149 - 162

Received: 12 Feb 2022
Accepted: 26 Mar 2022
Published online: 05 Apr 2023 *

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