Tribological behavior prediction of composite MoS2-TiO2 coating using Taguchi coupled artificial neural network approach

Solid lubricants have been popularly used in tribological applications by virtue of their excellent lubricating properties in extreme environmental conditions. The tribological performance of such lubricated surfaces can be significantly enhanced by optimization of the coating material properties. In the present study, the composite MoS2-TiO2 coating is fabricated and studied the different parameters affecting the tribological behavior of the developed coating using design of experiment Techniques such as Taguchi and artificial neural network (ANN) approach. The optimization study was performed by considering the various input parameters such as addition of wt.% of TiO2 into the MoS2 base matrix, contact pressure, sliding speed, and pin surface temperature which affects the coating's material tribological properties. The Taguchi approach reveals that pin surface temperature has the highest influence, followed by contact pressure, and wt.% addition of TiO2. The responses predicted using both Taguchi and ANN were compared and it has been observed that ANN helps to predict the responses more accurately compared to the Taguchi approach.

Automated summary

Solid lubricants are commonly used in tribological applications due to their excellent lubricating properties in extreme environmental conditions. The tribological performance of lubricated surfaces can be significantly improved by optimizing the coating material properties. In this study, a composite MoS2-TiO2 coating was fabricated and the effects of different parameters on the tribological behavior of the coating were investigated using design of experiment techniques such as Taguchi and artificial neural network (ANN) approach. The study showed that ANN provides more accurate predictions of the responses compared to the Taguchi approach.