Testing and evaluation on friction surfacing of Alloy 36 over spheroidal graphite iron and its effect on coating thickness and width using artificial neural networks and response surface methodology

Abstract

AbstractFriction surfacing finds its application in areas which require joining of dissimilar metals with superior wear property. The current work investigates the solid‐state coating of Alloy 36 over ductile cast iron (SG Iron) by friction surfacing process. The effect of process factors on the actual geometry of the coating was investigated through testing and a desirability‐based technique. Spindle speed, axial force, and table traverse speed were the three most important variables for bonding fidelity. Analysis of variance was used to examine how the input parameters affected the output parameters. Experimental findings are validated using artificial neural networks and response surface methodology which are then used to predict how the system will respond to various operating scenarios. The research demonstrates that speed, rather than force or traverse action, has a more substantial effect on both materials. The response surface methodology statistician approach was utilized to identify the most important factors influencing the process parameters. As a result, the load is determined as the most crucial component that impacts bonding to the substrate. According to the data, the output parameters have the most impact on lowering the bonding geometry.