Title :
Detection of the time of failure of a hybrid system by particle filtering in a log-likelihood ratio approach
Author :
Cadini, Francesco ; Peloni, Giovanni ; Zio, Enrico
Author_Institution :
Dipt. di Energia, Politec. di Milano, Milan, Italy
Abstract :
A particle filtering method is implemented within a log-likelihood ratio approach here specifically tailored to handle hybrid dynamic systems, i.e, systems characterized by transitions among discrete modes of operation and failure, each one giving rise to a specific continuous dynamics of evolution. An application to a case study of literature proves the capabilities of the newly devised algorithm.
Keywords :
particle filtering (numerical methods); continuous dynamics; hybrid dynamic systems; log-likelihood ratio approach; particle filtering; Actuators; Fault detection; Fault diagnosis; Filling; Filtering; Maintenance; Modems; Power generation economics; Probability distribution; Random variables;
Conference_Titel :
Prognostics and Health Management Conference, 2010. PHM '10.
Conference_Location :
Macao
Print_ISBN :
978-1-4244-4756-5
Electronic_ISBN :
978-1-4244-4758-9
DOI :
10.1109/PHM.2010.5413504
