Recursive subspace identification for fault detection

Future electrical grids will require new ways to identify faults, as inverters are not capable of supplying large fault currents to support existing fault detection methods, and because distributed resources may feed faults from the edge of the grid. This master’s thesis proposes using online, subsp...

Descripción completa

Detalles Bibliográficos
Autor: El Yaagoubi El Hassouni, Soufiane
Tipo de recurso: tesis de maestría
Fecha de publicación:2024
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:2117/428836
Acceso en línea:https://hdl.handle.net/2117/428836
Access Level:acceso abierto
Palabra clave:MIMO systems
Electric power systems
System analysis
Sistemes MIMO
Sistemes de distribució d'energia elèctrica
Anàlisi de sistemes
Àrees temàtiques de la UPC::Enginyeria elèctrica::Distribució d’energia elèctrica::Xarxes elèctriques
Descripción
Sumario:Future electrical grids will require new ways to identify faults, as inverters are not capable of supplying large fault currents to support existing fault detection methods, and because distributed resources may feed faults from the edge of the grid. This master’s thesis proposes using online, subspace/time-domain system identification for online power system fault detection. This innovative approach can detect high impedance faults, thus improving the reliability and safety of modern electrical grids. State-Space and ARX model methods are compared, demonstrating that ARX model methods are better suited for the task. Furthermore, a recursive ARX method is proposed which increases the detection speed. A discussion of the theoretical foundations, as well as practical implementations, is presented with simulation results validating the effectiveness and robustness of the proposed methods.