Optimization of Weld-Bead Parameters of Plasma Arc Welding Using GA and IWO

Abstract

Plasma arc welding (PAW) of Inconel 617 plates is an important and critical process for many engineering applications such as combustion cans, high-temperature nuclear reactors, and transition liners in aircraft due to its high depth-to-width ratio. Therefore, finding the combination of optimal input process parameters of the said welding process is an essential task to be carried out before employing it in various applications. In the present study, bead-on-plate (BoP) trails of PAW are performed on Inconel 617 plates after conducting the experiments designed based on the central composite design of experiments (CCD). During experimentation, welding speed, welding current, and gas flow rate are considered as input process parameters, and bead width and bead height of BoP trails are treated as responses of the PAW process. The nonlinear regression equations developed for both the bead width and bead height are optimized with the help of two population-based optimization algorithms, namely genetic algorithm (GA) and invasive weed optimization (IWO) algorithms.