The paper presents the preliminary results of an on-going research activity concerning the application of ANN (Artificial Neural Network) for on-line identification of potential harmful states in power systems (contingency screening). The necessity of online Static Security Assessment with a good degree of confidence and the evaluation of indices for successive indications to power system operators of possible remedial actions is deeply felt in the new context of liberalized electrical energy markets. The paper addresses primarily the use of ANN for contingency screening. It describes and reports the activity for choosing the most suitable ANN structure and training the ANN for a realistic power system. The chosen ANNs are then trained and used to assess the security state of a larger power system representing an equivalent model of the Italian HV grid.
ANN application for on-line power system security assessment
GRILLO, SAMUELE;
2006-01-01
Abstract
The paper presents the preliminary results of an on-going research activity concerning the application of ANN (Artificial Neural Network) for on-line identification of potential harmful states in power systems (contingency screening). The necessity of online Static Security Assessment with a good degree of confidence and the evaluation of indices for successive indications to power system operators of possible remedial actions is deeply felt in the new context of liberalized electrical energy markets. The paper addresses primarily the use of ANN for contingency screening. It describes and reports the activity for choosing the most suitable ANN structure and training the ANN for a realistic power system. The chosen ANNs are then trained and used to assess the security state of a larger power system representing an equivalent model of the Italian HV grid.File | Dimensione | Formato | |
---|---|---|---|
PMAPS 06.pdf
Accesso riservato
:
Post-Print (DRAFT o Author’s Accepted Manuscript-AAM)
Dimensione
279.31 kB
Formato
Adobe PDF
|
279.31 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.