Publication

Abstract : Artificial Neural Networks (ANNs) are one of the most powerful computer modeling technique, based on statistical approach, currently used in many fields of engineering for modeling complex relationships which are difficult to describe with physical models. This paper presents development of an ANN model to predict weld bead geometry and dilution for various input process parameters in duplex stainless steel cladding deposited by Flux Cored Arc Welding (FCAW). The chosen input process parameters were welding current, welding speed, contact tip-to-workpiece distance, and welding torch angle. The chosen output parameters were weld bead width, depth of penetration, height of reinforcement, and dilution. The experiments were conducted based on four-factor five level central composite rotatable design with full replications technique. Using the experimental data multi-layer feedforward neural network was developed and it was trained by using back propagation algorithm. This model is very useful for predicting the weld bead geometry and dilution. The performance of the developed neural network model was found to be accurate.