NEURAL NETWORKS FOR FAULT-DIAGNOSIS OF A NUCLEAR-FUEL PROCESSING PLANT AT DIFFERENT OPERATING POINTS

Industrial processes often produce at various operating points; howeve r, demonstrated applications of neural networks for fault diagnosis us ually consider only a single (primary) operating point. Developing a s tandard neural network scheme for fault diagnosis at all operating poi nts may be impractical due to the unavailability of suitable training data for less frequently used (secondary) operating points. This paper investigates the application of a single neural-network for the diagn osis of non-catastrophic faults in an industrial nuclear processing pl ant operating at different points. Data-conditioning methods are inves tigated to facilitate fault classification, and to reduce the complexi ty of the neural networks. Results illustrate the performance of train ed neural networks for classifying process faults using simulated and real industrial data. (C) 1998 Elsevier Science Ltd. All rights reserv ed.