Fault Detection and Estimation System Integrated To Fault Tolerant

DAVID A. ZUMOFFEN AND MARTA S. BASUALDO

Río Gallegos, Argentina.

Congreso; XII Reunión de Trabajo en Procesamiento de la Información y Control (RPIC),; 2007

Universidad Nacional de la Patagonia Austral

In this paper an expert system for fault detection, isolation and estimation (FDIE) approach is presented. It connects different research areas like statistical monitoring, knowledge-based systems and historical data-based systems. Adaptive Principal Components Analysis (APCA), Fuzzy-Logic (FL) and Artificial Neural Network (ANN) are integrated here in order to detect, classify and estimate the faulty process elements. The information given by the FDIE system is very useful for supporting the control structure reconfiguration so it turns as as a fault tolerant control (FTC) by applying a proper integration methodology . In this paper typical faults such as bias at sensors and extra variable dead time at actuators are considered. Several simulation results are presented obtained by an implementation on a benchmark case study of a Waste Water Treatment Plant (WWTP)