Volume 44, No 3, 2022, Pages 374-393

Download full text in PDF

Dry Sliding Friction and Wear Behavior of LM13/Zircon/Carbon (HMMC’s): An Experimental, Statistical and Artificial Neural Network Approach

Authors:

Yellampalli Prakash Ravitej ,
Chikkamaranahalli Boganarasimhaiah Mohan ,
Maravanji Gangadharaiah Ananthaprasad

DOI: 10.24874/ti.1223.11.21.03

Received: 26 November 2021
Revised: 27 December 2021
Accepted: 5 March 2022
Published: 15 September 2022

Abstract:

Dual reinforcement in composite materials leads to the betterment of properties. In the present research, zircon is added from a range of 0 wt. % to 12 wt. % in the steps of 3 wt. % by maintaining constant 3 wt. % graphite powder in addition to the aluminum alloy (LM13) matrix. Reinforcements are preheated at 250°C to remove moisture content and are dispersed in the molten aluminum alloy (LM13) by stir casting process where copper chills are kept at one end of the mold to enable unidirectional solidification. Wear properties are evaluated using pin on disc experiment. Design of the experiment (DOE) is done using Taguchi’s (L25) orthogonal array to assess the effect of (a) applied load (b) sliding speed (c) track radius/sliding distance (d) zircon content on wear loss/volume loss, coefficient of friction (COF)and wear rate. It is observed that wear rate and wear loss are proportional to the applied load, sliding speed and track radius COF is inversely proportional to the same. Wear rate decreases with the addition of zircon up to 9 wt. % and then increases at 12 wt. %. Statistical analysis is done using Minitab software where SN ratio, probability, ANOVA, and a regression model are analyzed. Obtained experimental wear properties are validated using artificial neural networks (ANN) by training the neurons where good agreement is obtained (R2= 0.98). Present research encapsulates the effect of different wear parameters like applied load, sliding speed, and sliding distance on the wear rate of LM13/zircon/C (hybrid metal matrix composites), and experimental wear results are correlated with artificial neural network (ANN) by training the algorithm.

Keywords:

ANOVA, Coefficient of friction, Design of experiments, Stir casting process, Wear properties