1998 | OriginalPaper | Buchkapitel
Using Artificial Neural Networks to Model Non-Linearity in a Complex System
verfasst von : P. Weller, A. Thompson, R. Summers
Erschienen in: Artificial Neural Nets and Genetic Algorithms
Verlag: Springer Vienna
Enthalten in: Professional Book Archive
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
This paper describes an investigation into using artificial neural networks (ANNs) to model the non-linearities in a complex system, a nuclear reactor. A simple one compartment finite difference model of the plant is developed and an exact ANN equivalent formed directly without training. Conventional training using standard transfer functions available in ANN packages is compared with this. A novel method is used to produce the ANN training and test sets. A twenty five compartment model is built from directly formed ANNs and the results compared to a simulator model.