Subspace-based Fault Detection and Isolation Methods - Application to Vibration Monitoring Michèle Basseville, Maher Abdelghani, Albert Benveniste
نویسندگان
چکیده
We address the problem of detecting and isolating faults modeled as changes in the eigenstructure of a linear dynamical system. The purpose of the paper is to describe and analyze new fault detection and isolation algorithms, based on recent stochastic subspace-based identi-cation methods and the statistical local approach to the design of detection algorithms. The application to vibration monitoring of mechanical structures and rotating machines is discussed. A conceptual comparison is made with another detection algorithm based on the instrumental variables identiication method, and previously proposed by two of the authors. D etection et diagnostic de pannes a l'aide de m ethodes par sous-espaces-Application a la surveillance des vibrations. R esum e : On consid ere le probl eme de la d etection et du diagnostic de pannes mod elis ees par des changements de la structure propre d'un syst eme dynamique lin eaire. Dans cet article sont d ecrits et analys es de nouveaux algorithmes de d etection et diagnostic de telles pannes, bas es sur les m ethodes r ecentes d'identiication par sous-espaces et sur l'approche statistique locale. L'application a la surveillance vibratoire des structures m ecaniques et machines tournantes est discut ee. Ces m ethodes sont compar ees conceptuellement a un autre algorithme bas e sur la m ethode d'identiication par variables instrumentales, et propos e pr ec edemment par deux des auteurs. Ce travail est eeectu e dans le cadre du projet Eureka no 1562 Sinopsys (Model based Structural monitoring using in-operation system identiication), qui est coordonn e par la soci et e Lms, Leuven, Belgique. Mots cl es : D etection et diagnostic de pannes, m ethodes stochastiques d'identiication par sous-espaces, approche statistique locale, surveillance des vibrations.
منابع مشابه
Subspace-based fault detection algorithms for vibration monitoring
We address the problem of detecting faults modeled as changes in the eigenstructure of a linear dynamical system. This problem is of primary interest for structural vibration monitoring. The purpose of the paper is to describe and analyze new fault detection algorithms, based on recent stochastic subspace-based identiication methods and the statistical local approach to the design of detection ...
متن کاملFault Detection and Isolation in Nonlinear Dynamic Systems: A Combined Input-Output and Local Approach
The problem of fault detection and isolation (FDI) is addressed for nonlinear dynamic systems modeled by polynomial diierential-algebraic equations. Ritt's algorithm is used to obtain an input-output representation of the monitored system by eliminating unknown variables. A local approach to change detection is then applied to the input-output representation for FDI residual generation and eval...
متن کاملPro O F Co Py 022104 Jds
We address the problem of structural model identification during normal operating conditions and thus with uncontrolled, unmeasured, and nonstationary excitation. We advocate the use of output-only and covariance-driven subspace-based stochastic identification methods. We explain how to handle nonsimultaneously measured data from multiple sensor setups, and how robustness with respect to nonsta...
متن کاملThe local approach to change detection, diagnosis, and model validation: application to vibration mechanics∗
In this paper we survey our results on model validation, change detection, and diagnosis, based on the local approach. We see this work as a counterpart of the central contribution by Lennart Ljung to a systematic approach to system identification. We reuse concepts brought by Lennart such as: model set, true system model, identification method, and we enrich this set of concepts by considering...
متن کاملOn-board Component Fault Detection and Isolation Using the Statistical Local Approach
We describe both the key principles and real application examples of a uniied theory which allows us to perform the on-board incipient fault detection and isolation tasks involved in monitoring for condition-based maintenance. We stress that, when designing detection algorithms , the main conceptual task is to select a convenient estimating function. ml, ls, iv and subspace identiication method...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1997