The objective of the study is the development of the system for effective prediction of residual stresses using the back propagation algorithm from the neural network.
The achieve of this goal, the series experiment were carried out and measured the ...
The objective of the study is the development of the system for effective prediction of residual stresses using the back propagation algorithm from the neural network.
The achieve of this goal, the series experiment were carried out and measured the residual stresses using sectional method. Using the experimental results, the optional control algorithms using a neural network should be developed in order to reduce than the effect of the external distribution during GMA welding processes. Also, comparison with the measured and the calculated results from the FEM(finite element method) and verification of the developed system was carried out. This system can not only help to understand the interaction between the process parameters and residual stress, but also, improve the quantity control for welded structures.
Then the results obtained from this study are as follows. Through comparison between the measured and calculated results, the neural network based on back propagation algorithm is the best techniques to predict the process parameter. A new techniques which predict the process parameter such as welding voltage, arc current, welding speed using the training the raw dates, will be proposed.