Statistical root cause analysis of novel faults based on digraph models

► Neither fault mechanism nor relevant historical data are available for novel faults. ► We combine multivariate statistical process monitoring (MSPM) and causal reasoning. ► Normal historical data are incorporated into causal reasoning through MSPM. ► Signed digraph enables automatic variable selec...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Chemical engineering research & design 2013-01, Vol.91 (1), p.87-99
Hauptverfasser:	Wan, Yiming, Yang, Fan, Lv, Ning, Xu, Haipeng, Ye, Hao, Li, Weichang, Xu, Peng, Song, Liming, Usadi, Adam K.
Format:	Artikel
Sprache:	eng
Schlagworte:	Causal reasoning Digraph model Fault diagnosis Faults Graph theory Mathematical models Monitoring Multivariate statistical process monitoring Novel fault Qualitative reasoning Root cause analysis Roots
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!