Investigation of reciprocating wear characteristics of carbon steel at varying load, relative humidity, and operating temperature using ANN-NSGA methodology

Abstract

Steel's friction and wear characteristics depend not only on the load applied and the nature of the contact surface but also on the environmental circumstances directly and indirectly. This research deals with an investigation of tribological behavior and the wear mechanism of low carbon steel and the effect of load (10 N–100 N), varying temperature (room temperature—200 °C), and relative humidity ranging between 30% and 70% under high-frequency reciprocating condition. Variation of wear rate and coefficient of friction with test parameters was evaluated by generating 3D surface and contour plots. Wear surface morphologies were also characterized. The synergistic effects of the parameters led to a very complex trend in the variation of wear rate and coefficient of friction. Due to the complexity of the problem, an artificial neural network was coupled with the genetic algorithm for the modeling and optimization of tribological characteristics. Simultaneous minimization of wear rate and coefficient of friction was obtained at 18.7349 N load, 199.99 °C temperature, and 49.8332% relative humidity. The worn surface comprised a low shear oxide film and pressed debris patches. The artificial neural network model was also obtained with a significantly high accuracy having an overall R-value close to 1 for both wear rate and coefficient of friction.