Transit Time Estimation by Artificial Neural Networks
The use of interactive activation and competition (IAC) and backpropagation (BP) artificial neural networks (ANNs) for transit time estimation has been investigated in this piece of research. Owing to its competitive nature, the IAC ANN has been found able to correctly estimate the current transit time from short records of signals as well as to quickly follow changes in transit time and to detect when the transit time falls outside a predefined expected range. On the other hand, the interactive nature of the IAC ANN allows it to be robust to significant levels of noise and of the global component. A BP ANN has been appended to the IAC ANN, further allowing for the accurate estimation of decimated transit times.
KeywordsArtificial Neural Network Transit Time Local Component Neutron Density Competitive Nature
Unable to display preview. Download preview PDF.
- M. Antonopoulos, M. Marseguerra, and E. Padovani. On the fast estimation of transit times — application to bwr simulated data. In Proceedings of the SMORN VII (Volume 2), 1995.Google Scholar