Vol. 2, Issue 6 (2017)
Discretization and artificial neural network approach in resistance spot welding of aluminium alloy AA6063 T6 sheets used in automotive applications
Author(s): BDY Sunil, Dr. B Balu Naik, Dr. K Sammaiah, Dr. KGK Murti, N Ananth
Abstract: The present research work investigates the approach for predicting the mechanical properties – tension shear; nugget diameter; direct tension; torsion shear for a resistance spot weld of aluminum 6063 T6 sheets of 0.5 mm. In this study, the discrete analysis viz., finite element method; finite difference method was used to predict the parameters for experimentation and heat affected zone. Various levels of the three important spot welding parameters - weld force, weld current and weld time were selected to be used for analysis and experimentation. In order to validate the numerical results a series of experiments were carried out and the mechanical properties were measured. The results obtained with the experimentation were used to build up a back propagation, artificial neural network model for the mechanical properties prediction. The results revealed that the combination of these two developed models – discretization and artificial neural network can accurately and rapidly predict the mechanical properties for a resistance spot weld.