Precedent-free Fault Isolation in a Diesel Engine Egr Valve System

Content Provider	Semantic Scholar
Author	Cholette, Michael E. Djurdjanovic, Dragan
Copyright Year	2009
Abstract	In this paper, a framework for isolating unprecedented faul ts for an EGR valve system is presented. Using normal behavior data generated by a high fidelity engine simulation, the r ecently introduced Growing Structure Multiple Model System (GSMMS) is used to construct models of normal behavior for an EGR valve system and its various subsystems. Using the GSMMS models as a foundation, anomalous behavior of the entire system is then detected as statistically significant de partures of the most recent modeling residuals from the modeling resi duals during normal behavior. By reconnecting anomaly detect ors to the constituent subsystems, the anomaly can be isolated w i hout the need for prior training using faulty data. Furthermo e, faults that were previously encountered (and modeled) are r ecognized using the same approach as the anomaly detectors.
File Format	PDF HTM / HTML
Alternate Webpage(s)	https://faculty.engr.utexas.edu/sites/default/files/limes/files/rc6.pdf
Language	English
Access Restriction	Open
Subject Keyword	Anomaly detection Detectors Diesel Ebstein Anomaly Simulation
Content Type	Text
Resource Type	Article