TS fuzzy reconfiguration blocks for fault tolerant control of nonlinear systems
This paper introduces a new Takagi–Sugeno (TS) reconfiguration block structure for control reconfiguration of nonlinear systems with sensor and actuator faults represented by TS models and provides conditions for the existence of such blocks that ensures the stability recovery by fault hiding. The p...
| Autores: | , , |
|---|---|
| Tipo de recurso: | artículo |
| Fecha de publicación: | 2020 |
| 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:2117/340613 |
| Acceso en línea: | https://hdl.handle.net/2117/340613 https://dx.doi.org/10.1016/j.jfranklin.2020.02.002 |
| Access Level: | acceso abierto |
| Palabra clave: | Fault tolerance (Engineering) Nonlinear systems Tolerància als errors (Enginyeria) Sistemes no lineals Àrees temàtiques de la UPC::Informàtica::Automàtica i control |
| Sumario: | This paper introduces a new Takagi–Sugeno (TS) reconfiguration block structure for control reconfiguration of nonlinear systems with sensor and actuator faults represented by TS models and provides conditions for the existence of such blocks that ensures the stability recovery by fault hiding. The proposed reconfiguration block structure consists of a convex sum of linear transformations of the control signals and the sensor measurements hiding the fault occurrence from the nominal controller. Similar static reconfiguration blocks were previously proposed in the literature, however the conditions for their feasibility are more restrictive. This paper presents novel conditions based on the Lyapunov stability theory for obtaining those TS reconfiguration blocks for stability recovery after actuator and sensor faults. These conditions are obtained by assuming that the Lyapunov function that ensures the stability of fault-free closed-loop TS system is a valid Lyapunov candidate for the reconfigured TS system under faulty conditions. In particular, stability recovery conditions for nonlinear systems represented by TS models with nonlinear consequent and networked TS models are provided. Simulation results for a power system and a network of pendulum systems are used to illustrate the effectiveness of the proposed reconfiguration block. |
|---|