Skip to Main content Skip to Navigation
Theses

Design of adaptive observers for LPV systems: Application to fault tolerant control

Abstract : In this thesis, the observer design for linear parameter-varying systems and their applications to fault diagnosis and fault-tolerant control is studied. A linear parameter-varying (LPV) system can approximate the nonlinear dynamic behavior through a set of linear state space models that are interpolated by a mechanism depending on the scheduling variables. The observer used in this research is called generalized dynamic observer (GDO), where the principal idea is to add dynamics structure to increase its degrees of freedom, with the purpose of achieving steady-state accuracy and improve robustness in estimation error against disturbances and parametric uncertainties. Therefore, this structure can be considered as more general than a proportional observer and proportional-integral observer. It addresses the GDO synthesis for LPV systems with measured and unmeasured scheduling variables such as the quasi-LPV system, in which the scheduling variables are functions of endogenous signals such as states, inputs, or outputs instead of exogenous signals. Conditions of existence and stability for each GDO structure are given through the Lyapunov approach using parameter-independent Lyapunov function or parameter-dependent Lyapunov function. The design is obtained in terms of a set of linear matrix inequalities. Engineering applications are used to illustrates the performance and effectiveness of the proposed approaches. It considers a fault-tolerant control (FTC) strategy for polytopic LPV systems to maintain the current system close to the desired performance and preserve stability conditions in the presence of actuator faults. A fault diagnosis unit is built to estimate the states, the actuator faults, and the parameter variation. That information is essential to the FTC law.
Document type :
Theses
Complete list of metadatas

Cited literature [89 references]  Display  Hide  Download

https://hal.univ-lorraine.fr/tel-02960200
Contributor : Thèses Ul <>
Submitted on : Wednesday, October 7, 2020 - 2:39:06 PM
Last modification on : Wednesday, October 14, 2020 - 3:54:48 AM

File

DDOC_T_2020_0069_PEREZ_ESTRADA...
Files produced by the author(s)

Identifiers

  • HAL Id : tel-02960200, version 1

Collections

Citation

Abraham Jashiel Pérez-Estrada. Design of adaptive observers for LPV systems: Application to fault tolerant control. Automatic. Université de Lorraine; Centro Nacional de Investigación y Desarrollo Tecnológico (Cuernavaca, Mor., México), 2020. English. ⟨NNT : 2020LORR0069⟩. ⟨tel-02960200⟩

Share

Metrics

Record views

55

Files downloads

8