Process optimization for maximizing bushing length in thermal drilling using integrated ANN-SA approach

Abstract

Thermal drilling is a new hole producing method in sheet material applications that uses the frictional heat generated at the intersection of workpiece and drilling tool with an intention of softening and subsequently, penetrating the hole. According to L27-Taguchi orthogonal array approach, the thermal drilling experiments were executed on galvanized steel which is engaged in boat yards construction and automobile body manufacture applications. All the way through experimentation of thermal drilling with the noteworthy process conditions like rotational speed (S), workpiece thickness (Wt) as well as tool angle (α) were varied in different levels. Using the experimental outcomes, an appropriate artificial neural network technique (ANN) is established to develop bushing length model. Bushing length (BL) is considered as the most important output parameter in thermal drilling, since it is directly linked by tapping operation. Thus, the effects of thermal drilling input parameters on bushing length is studied since it is involved in fastening the galvanized steel by means of fasteners as required in numerous engineering problems. Furthermore, process optimization is established by means of simulated annealing (SA) algorithm approach under constraint boundaries with a view to maximization of bushing length. Besides, compare the optimal value of bushing length as predicted by SA algorithm with experimental bushing length value and this validates the development of thermal drilling on galvanized steel. An outstanding conformity has been detected between the predicted optimum and experimental value of bushing length.