Abstract
This chapter reviews two fault detection schemes based on recursive subspace identification. Both schemes have the ability to detect a fault of a system under surveillance in real time, without being disturbed by changes in the dynamics of the input source. One scheme is based on a hypotheses testing on the distance between the two estimates of a parameter of a system, one of which is made by a subspace identification algorithm having the ability to fast track the variation of the parameter and the other is made by a recursive subspace identification algorithm without forgetting. This scheme is computationally more expensive than the other one because it is required that two parameter estimation algorithms are activated in parallel. Nevertheless, it has great potential as a fault isolation scheme. The other scheme is based on a hypotheses testing on the so-called residual, which is a by-product of the recursive subspace identification algorithm. This scheme is computationally cheaper than the former one. Moreover, it can deal with systems represented by the innovation form. The innovation form is a wide class of models, which includes, as a part, the output-error model. However, a reasonable signal-to-noise-ratio is desirable for this scheme.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Systems and Human Science - For Safety, Security and Dependability
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
