A New Back-Propagation Algorithm for Diagnosing Defective Casting

Abstract

Abstract-The diagnosis of defective castings has always been a centre of attention in the manufacturing industry. An intelligent diagnosis system should be able to diagnose effectively the causal representation and also justify its diagnosis. Recently, the artificial-neural networks (ANN), or simply neural-networks (NN), technique has gained more popularity in learning cause and effect analysis in casting processes However, the algorithm comes with problems like slow convergence and convergence to local minima and unable to model exponential increase/decrease in belief values in cause and effect relationships. This paper proposed a new algorithm for improving the conventional back-propagation algorithm by introducing gain parameters during the learning process. The abilities of the proposed method to capture the exponential change in the belief variation of the cause when the belief in the effect is at its minimum is compared with the outputs from both the conventional back propagation on a real casting data set.