Optimal input design for autoregressive model discrimination with constrained output variance

Uosaki, Katsuji; Tanaka, I.; Sugiyama, Hiroshi

doi:10.1109/tac.1984.1103525

Cited by 25 publications

(11 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Even though the design of auxiliary signals has been an issue in the system identification literature, the point of view adopted here is closer to that found in failure detection problems, in particular [12,5,6]; see also [11] . However these works, and most problem formulations for auxiliary signal design in the identification literature, use statistical methods and consider model uncertainties as stochastic processes.…”

Section: Introductionmentioning

confidence: 88%

Auxiliary signal design for robust multimodel identification

Nikoukhah

Campbell

Horton

et al. 2002

IEEE Trans. Automat. Contr.

View full text Add to dashboard Cite

Under the assumption that one of two given linear models is the real underlying model of the system, a proper auxiliary signal is defined as an input signal that allows the selection of the correct model. Under the assumption that model noises have bounded energy, the separability index is defined as the inverse of the energy of the proper auxiliary signal of least energy. A method for the computation of this index and the construction of the auxiliary signal is presented. Finally, an efficient procedure for on-line identification called the hyperplane test is developed.

show abstract

Section: Introductionmentioning

confidence: 88%

Auxiliary signal design for robust multimodel identification

Nikoukhah

Campbell

Horton

et al. 2002

IEEE Trans. Automat. Contr.

View full text Add to dashboard Cite

show abstract

“…Some interesting theories in the design and implementation of detection signals are given in [79][80][81][82][83][84].…”

Section: Signal-based Fdimentioning

confidence: 99%

System health monitoring and prognostics — a review of current paradigms and practices

Kothamasu

Huang

VerDuin

2006

Int J Adv Manuf Technol

275

121

View full text Add to dashboard Cite

System health monitoring is a set of activities performed on a system to maintain it in operable condition. Monitoring may be limited to the observation of current system states, with maintenance and repair actions prompted by these observations. Alternatively, monitoring of current system states is being augmented with prediction of future operating states and predictive diagnosis of future failure states.Such predictive diagnosis or prognosis is motivated by the need for manufacturers and other operators of complex systems to optimize equipment performance and reduce costs and unscheduled downtime. Prognosis is a difficult task requiting precise, adaptive and intuitive models to predict future machine health states. Numerous modeling techniques have been proposed in the literature and implemented in practice. This paper reviews the philosophies and techniques that focus on improving reliability and reducing unscheduled downtime by monitoring and predicting machine health.Keywords Condition-based maintenance 9 Health monitoring -Model-based diagnosis -OSA-CBM 9 Preventive maintenance 9 Reliability-centered maintenance

show abstract

“…used for fault detection and isolation (see e.g. (Basseville, 1997(Basseville, , 1998Huang and Tamayo, 2000)), model discrimination (Uosaki et al, 1984) or model validation (Ljung, 1999). In order to decide between the two (contradictory) hypotheses, a test (or decision rule) is designed based on a so-called test statistic.…”

Section: Introductionmentioning

confidence: 99%