Model based Fault-Diagnosis for Linear Systems - 5AMAC514

Goal(s)

Fault detection and isolation (FDI)is a subfield of control engineering which involves monitoring a system, identifying when a fault has occurred, and pinpointing the type of fault and its location. Model-based techniques of fault detection and isolation use a model to investigate/analyze the occurrence of faults. The system model may be mathematical or knowledge-based. We focus our attention on mathematical models.

Content(s)

There ar only tree main approaches. The observer-based approach, the parity-space approach, and parameter identification-based methods. In order to optimize FDI indications, the following two step are developped :

The first step is to design a filter based on a model of the plant to generate a vector known as the residual. The residual should ideally be zero (or zero mean) under no-fault conditions.

The second step is to make decisions on whether a fault has occurred. This step is usually done using statistical tools to test if the residual has significantly deviated from zero.

State space representation
Observer design
Identification H2
Algebre of matrice : Rank, ker, eigenvalue ...