It is quite important to understand the factors which affect the damaging process of MgO-C bricks during steel making for improving the steel making efficiency. However, the process is too complex to treat with the usual linear analysis. The aim of this study is to analyze the factors with non-linear analyzing technique i.e. the artificial neural network (ANN) analysis. The starting neural network model was constructed using the data set extracted from three previously published literatures. The model was then refined by selecting the factors according to their contribution to the damaging process. The final network model reproduced the experimental results satisfactorily and found new informations from the used data set.

View full abstract