Neural Networks Design for Control in CO2 Welding using Backpropagation and Levenberg-Marquardt Algorithm

Abstract

CO2 welding is a joining process which is used to produce high quality joints and has a capability to be utilized in automation systems to enhance productivity. Despite its wide spread use in the various manufacturing industries, the full automation of the CO2 welding has not yet been achieved partly because mathematical models for the process parameters for a given welding task are not fully understood and quantified. In this paper a neural network model is developed to predict the weld bead width as a function of key process parameters in CO2 welding. The neural network model is developed using two different training algorithms, namely, the error backpropagation algorithm and the Levenberg-Marquardt approximation algorithm. The accuracy of the neural network models developed in this study has to be tested by comparing the simulated data obtained from the neural network model with that obtained from the actual CO2 welding experiments. The result will show that the Levenberg-Marquardt approximation algorithm is the preferred method, as this algorithm reduces the root of the mean sum of square (RMS) error to a significantly small value.