Title :
Diagnosis of batch processes based on parameter estimation of discrete event models
Author :
Schullerus, Gernot ; Krebs, Volker
Author_Institution :
Inst. fur Regelungs- und Steuerungssyst., Univ. Karlsruhe, Karlsruhe, Germany
Abstract :
This paper presents a model based diagnosis approach for batch processes that can be adequately modeled by timed event graphs. The evolution of the event times for this particular class of discrete event systems can be described by an algebraic model using the max-plus algebra. We present and discuss a diagnosis method based on parameter estimation of such algebraic discrete event process models and demonstrate the application by diagnosing valve faults in a batch process.
Keywords :
algebra; batch processing (industrial); fault diagnosis; graph theory; parameter estimation; valves; algebraic discrete event process models; algebraic model; batch processes; discrete event models; max-plus algebra; model based diagnosis approach; parameter estimation; timed event graphs; valve fault diagnosis; Algebra; Batch production systems; Inductors; Mathematical model; Parameter estimation; Solvents; Valves; Applications; Identification Methods; Modelling; Petri Nets; Qualitative Methods of Fault Diagnosis;
Conference_Titel :
Control Conference (ECC), 2001 European
Conference_Location :
Porto
Print_ISBN :
978-3-9524173-6-2
