Dynamic process modeling and fault prediction using artificial neural networks

This case study presents two practical applications where artificial neural networks (ANNs) have been used to solve difficult process engineering problems. Firstly, ANNs are shown to provide a more accurate process model of a vitrification process than was possible using linear techniques. In the second application ANNs are applied in a novel way in which the residuals of the models are monitored in order to detect the imminent failure of a vessel used in the vitrification process.