Comparison of fault detection and isolation methods: A review

Abstract

Fault Detection and Isolation (FDI) is important in many industries to provide safe operation of a process. To determine the kind, size, location and time of fault, many Fault detection and Identification (FDI) Techniques are proposed. The Characteristic of FDI techniques include robustness, fast detection and isolation of faults. In this paper a comparison of fault diagnosis system based on Artificial Neural Network (ANN), Observer, Fuzzy, Kalman filter is presented. To achieve fault detection and isolation, a set of residuals need to be determined. Residual indicates the state of the system and provide information about the source of possible faults. A comparison of residual generation methods such as observer based residual generation, parity relation, Kalman filter and structural analysis is also presented in this paper.