Fault detection and identification with a new feature selection based on mutual information

Sylvain Verron (1) , Teodor Tiplica (1) , Abdessamad Kobi (1)

1 LASQUO - Laboratoire en Sûreté de Fonctionnement, Qualité & Organisation

This paper presents a fault diagnosis procedure based on discriminant analysis and mutual information. In order to obtain good classification performances, a selection of important features is done with a new developed algorithm based on the mutual information between variables. The application of the new fault diagnosis procedure on a benchmark problem, the Tennessee Eastman Process, shows better results than other well known published methods.

Domaines: Applications [stat.AP]; Intelligence artificielle [cs.AI]

Sylvain Verron, Teodor Tiplica, Abdessamad Kobi. Fault detection and identification with a new feature selection based on mutual information. Journal of Process Control, 2008, 18 (5), pp.479-490. ⟨10.1016/j.jprocont.2007.08.003⟩. ⟨hal-00516996⟩