Zonotopic fault detection observer for discrete-time descriptor systems considering - fault sensitivity

This paper addresses a zonotopic fault detection (FD) observer for discrete-time descriptor systems subject to additive actuator faults. The considered descriptor systems are also perturbed by unknown-but-bounded uncertainties including state disturbances and measurement noise. Under a set-based fra...

ver descrição completa

Detalhes bibliográficos
Autores: Wang, Ye|||0000-0003-1395-1676, Puig Cayuela, Vicenç|||0000-0002-6364-6429, Cembrano Gennari, Gabriela|||0000-0003-1436-6022, Zhao, Yuxin
Formato: artículo
Fecha de publicación:2020
País:España
Recursos:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/340512
Acesso em linha:https://hdl.handle.net/2117/340512
https://dx.doi.org/10.1080/00207721.2020.1820098
Access Level:acceso abierto
Palavra-chave:Failure analysis (Engineering)
Fault detection
Zonotopic observer
Robustness
Fault sensitivity
Discrete-time descriptor systems
Linear matrix inequalities
Generalised KYP lemma
Anàlisi de fallades (Enginyeria)
Àrees temàtiques de la UPC::Informàtica::Automàtica i control
Descrição
Resumo:This paper addresses a zonotopic fault detection (FD) observer for discrete-time descriptor systems subject to additive actuator faults. The considered descriptor systems are also perturbed by unknown-but-bounded uncertainties including state disturbances and measurement noise. Under a set-based framework, the effects of uncertainties and faults are separated into two state zonotopes. Based on this decomposition, the FD observer gain is designed to be robust against uncertainties and meanwhile sensitive to faults based on a H- index in a finite-frequency domain. Then, two linear matrix inequality (LMI) conditions are obtained to design an observer that achieves robustness and fault sensitivity at the same time. The generalised KYP lemma is applied to address the fault sensitivity condition. The optimal FD observer gain can be obtained via an offline design procedure. Finally, the proposed fault detection method is applied to a chemical mixing system and the effectiveness is shown.