UE Supervision and diagnosis

Degrees incorporating this pedagocial element :

Description

Safety, supervision and diagnosis of industrial plants 

The objective of this class is to introduce the concept of fault detection and fault diagnosis for complex systems and to present different classes of methods which have proven their performances in practical applications. 

Lesson 

Topic 

Introduction to supervision 

 

Tasks of supervision, terminology.

Model-based fault detection 

 

Parity equations, observers, on line estimation of model parameters.

Signal-based fault detection 

 

Features extraction using time, frequency and time-frequency transformation, pattern comparison. Temporal change detection.

Data-driven fault detection methods 

 

Fault diagnosis with pattern recognition, fault diagnosis with principal component analysis.

 

SUPERVISION LABS 

 

Lab 1 

Fault detection in a two-tanks system using a bank of observers 

Lab 2 

On-line detection of deep sleep using EEG spectral power 

Lab 3 

Diagnosis of a mineral treatment unit using pattern recognition 

Lab 4 

Sensor fault detection in an air quality monitoring network using principal component analysis 

Bibliography

  • S. Gentil (Ed.), "Supervision des procédés complexes", HERMES Systèmes automatisés, 2007.
  • Isermann, "Fault diagnosis systems", Springer, 2006.
  • Blanke, Kinnaert, Lunze, Staroswiecki, "Diagnosis and fault tolerant control", Springer, 2003.