In this paper, a fault diagnosis method is developed for systems described by multimodels. The main contribution consists in the design of a new Fault Detection and Isolation scheme (FDI) through an adaptive filter for such systems. Based on the assumption that dynamic behavior of the process is described by a multi-model approach around different operating points, a set of residual is established in order to generate weighting functions robust to faults. These robust weighting functions are directly linked with the adaptive filter effectiveness which provides multiple fault magnitude estimations for the whole operating range of the system. Stability conditions of the adaptive filter are studied and its performances are tested using an hydraulic system.

No abstract

HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt età la diffusion de documents scientifiques de niveau recherche, publiés ou non, emanant desétablissements d'enseignement et de recherche français ouétrangers, des laboratoires publics ou privés.

In this paper, the main goal is to design an approach that performs fault detection, isolation and estimation for a large class of nonlinear systems. Fault diagnosis is established by regarding system as a convex combination of linear time invariant (LTI) stochastic models and not as a single global model. The nonlinear representation is based on a bank of decoupled Kalman filters. This paper consists in generating a robust model selection of the "best" representative linear model. Under fault isolation conditions, the main contribution is to design an adaptive filter which makes possible multiple faults detection which appear simultaneously or in a sequential way, isolation and estimation over the whole operating range of nonlinear system. The stability conditions of the adaptive filter are developed. These conditions result in convex linear matrix inequalities (LMIs) that can be solved efficiently with optimization techniques. Performances of the method are tested on an academic example. Copyright@2003 EJC.

No abstract

There are different schemes based on observers to detect and isolate faults in dynamic processes. In the case of fault diagnosis in instruments (FDI) there are different diagnosis schemes based on the number of observers: the Simplified Observer Scheme (SOS) only requires one observer, uses all the inputs and only one output, detecting faults in one detector; the Dedicated Observer Scheme (DOS), which again uses all the inputs and just one output, but this time there is a bank of observers capable of locating multiple faults in sensors, and the Generalized Observer Scheme (GOS) which involves a reduced bank of observers, where each observer uses all the inputs and m-1 outputs, and allows the localization of unique faults. This work proposes a new scheme named Simplified Interval Observer SIOS-FDI, which does not requires the measurement of any input and just with just one output allows the detection of unique faults in sensors and because it does not require any input, it simplifies in an important way the diagnosis of faults in processes in which it is difficult to measure all the inputs, as in the case of biologic reactors.

No abstract

International audienceThe aim of this study is to provide a full comprehensive study of the singular linear parameter-varying (LPV) systems and observer synthesis in a real process. The states and unknown inputs estimation problem for a binary distillation column is solved by using an LPV proportional-integral observer (PI-Observer). A singular LPV model of the process is developed. This model facilitates the design of the PI-Observer for singular LPV systems. The observer proposed is tested using experimental data of a distillation column separating an ethanol??water mixture. The results indicate that the observer is able to accurately reconstruct the product composition dynamics and the unknown inputs of the process