HYBRID METHODOLOGIES FOR THE ANALYSIS OF FAULTS IN COMPLEX DYNAMIC SYSTEMS

DURATION: From January 1, 1999 to December 31, 2002

CHIEF RESEARCHER: Ángela Nebot Castells

KEYWORDS:

Complex dynamic systems; analysis and prediction of time series; fuzzy inductive reasoning; heterogeneous neural networks; delta neural networks; failure monitoring; Water distribution network.

SUMMARY

Based on the group's experience in Fuzzy Inductive Reasoning (FIR), Reconstruction Analysis (RA), Heterogeneous Neural Networks (RNH) and Delta Neural Networks (RND), the project aims to conduct a comparative study of such modeling and forecasting methodologies, detect the strengths of each and integrate them into a hybrid methodology for failure analysis and its possible prediction. The results will be experienced on "benchmarks", laboratory models and finally they will be applied to the model of a water distribution network.

PARTICIPANTS:

Researchers from the following institutions:

• Departamento de Lenguajes y Sistemas Informáticos (UPC)
• Departamento de Ingeniería de Sistemas, Automática e Informática Industrial (UPC)
• Departamento de Estadística e Investigación Operativa (UPC)
• Instituto de Robótica e Informática Industrial (UPC)
• Department of Information Technology (National Research Council of Canada)
• Department of Electrical and Computer Engineering (University of Arizona)