Output Behavior Equivalence and Simultaneous Subspace Identification of Systems and Faults

We address the problem of identifying a system subject to additive faults, while simultaneously reconstructing the fault signal via subspace methods. We do not require nominal data for the identification, neither do we impose any assumption on the class of faults, e.g., sensor or actuator faults. We...

Full description

Saved in:
Bibliographic Details
Published inIEEE control systems letters Vol. 9; pp. 1285 - 1290
Main Author de Albuquerquer Gleizer, Gabriel
Format Journal Article
LanguageEnglish
Published IEEE 2025
Subjects
data-driven modeling
Estimation
Fault diagnosis
Instruments
Linear systems
Observability
Probabilistic logic
System identification
system realization
Training
Trajectory
Online AccessGet full text
ISSN2475-1456
2475-1456
DOI10.1109/LCSYS.2025.3581870

Cover

More Information
Summary:We address the problem of identifying a system subject to additive faults, while simultaneously reconstructing the fault signal via subspace methods. We do not require nominal data for the identification, neither do we impose any assumption on the class of faults, e.g., sensor or actuator faults. We show that, under mild assumptions on the fault signal, standard PI-MOESP can recover the system matrices associated to the input-output subsystem. Then we introduce the concept of output behavior equivalence, which characterizes systems with the same output behavior set, and present a method to establish this equivalence from system matrices. Finally, we show how to estimate from data the complete set of fault matrices for which there exist a fault signal with minimal dimension that explains the data.
ISSN:2475-1456
2475-1456
DOI:10.1109/LCSYS.2025.3581870