Passivation blocks for fault tolerant control of nonlinear systems

This paper describes a novel passivity/dissipativity based approach for fault tolerant control (FTC) of nonlinear systems by means of fault hiding, i.e., by inserting reconfiguration blocks (RBs) between the plant and controller to mitigate the fault effects. The proposed approach is used to design...

Descripción completa

Detalles Bibliográficos
Autores: Bessa, Iury, Puig Cayuela, Vicenç|||0000-0002-6364-6429, Martínez Palhares, Reinaldo
Tipo de recurso: artículo
Fecha de publicación:2021
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/362696
Acceso en línea:https://hdl.handle.net/2117/362696
https://dx.doi.org/10.1016/j.automatica.2020.109450
Access Level:acceso abierto
Palabra clave:Fault-tolerant computing
Fault tolerance (Engineering)
Automatic control
Fault-tolerant systems
Fault hiding
Nonlinear control systems
Dissipativity theory
Passivity indices
Tolerància als errors (Informàtica)
Tolerància als errors (Enginyeria)
Control automàtic
Àrees temàtiques de la UPC::Informàtica::Automàtica i control
Descripción
Sumario:This paper describes a novel passivity/dissipativity based approach for fault tolerant control (FTC) of nonlinear systems by means of fault hiding, i.e., by inserting reconfiguration blocks (RBs) between the plant and controller to mitigate the fault effects. The proposed approach is used to design a new kind of RB, called passivation block (PB), which is generically employed for sensor and actuator faults and achieves simultaneously series, feedback and feedfoward passivation of the controller during a fault occurrence. Based on the dissipativity theory, new conditions are obtained to design a dynamic PB (DPB) which requires minimum information on the system model. In particular, the proposed DPB can be systematically obtained by combining the LMI-based conditions based on the knowledge about the passivity indices. Numerical simulations are carried out and indicate that the PBs are able to stabilize an example of a faulty nonlinear system.