Fault detection and isolation in switched linear systems and applications

Fault detection and isolation;reduced-order observers;switched linear systems;photovoltaic systems;differential power processing converters

Zholbaryssov, Madi ; Domínguez-García ; Alejandro D.

关键词: Fault detection and isolation;    reduced-order observers;    switched linear systems;    photovoltaic systems;    differential power processing converters;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/49551/Madi_Zholbaryssov.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

This thesis proposes a fault detection and isolation (FDI) method for switched linear systems. Themethod builds on a class of FDI filters for linear time-invariant (LTI) systems based on reduced-orderobservers. If certain conditions are satisfied, then it is possible to apply these filters toswitched linear systems. When system parameters change slowly, a system is considered to be ina faulty mode. In the absence of faults, the system dynamics are described by a switched linearstate space model. In a faulty mode, the state space model is modified by adding disturbanceterms associated with parameter changes and component degradations. An FDI filter consists of abank of reduced-order observers with residual generators which have certain geometric propertiesthat allow detecting and isolating faults. In order to distinguish different faults, an FDI filter usesdifferent residual generators and observers. We demonstrate practical feasibility of our approachby applying it to a photovoltaic (PV) system with differential power processing (DPP) converters.Our simulation results confirmed the fact that an FDI filter can detect and pinpoint multiple faultswhich can simultaneously affect a system. Finally, we experimentally demonstrate the feasibilityof our approach.