System identification and validation of aggregate models of renewable resources

Κρυσταλλίδης, Απόστολος; Krystallidis, Apostolos

dc.contributor.author	Κρυσταλλίδης, Απόστολος	el
dc.contributor.author	Krystallidis, Apostolos	en
dc.date.accessioned	2016-05-19T07:01:25Z
dc.date.available	2016-05-19T07:01:25Z
dc.date.issued	2016-05-19
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/42526
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.10078
dc.rights	Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα	*
dc.rights	Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα	*
dc.rights.uri	http://creativecommons.org/licenses/by-nc-nd/3.0/gr/	*
dc.subject	System Identification	en
dc.subject	Aggregate renewable resources	en
dc.subject	ARX	en
dc.subject	ARMAX	en
dc.subject	ERA	en
dc.subject	Power systems	en
dc.subject	SVD	en
dc.subject	Identification methods	en
dc.subject	Αναγνώριση συστημάτων	el
dc.subject	Μέθοδοι αυτοπαλινδρόμησης	el
dc.subject	Συγκεντρωτικά μοντέλα ανανεώσιμων πηγών ενέργειας	el
dc.subject	Μέθοδοι αναγνώρισης συστημάτων	el
dc.title	System identification and validation of aggregate models of renewable resources	en
heal.type	bachelorThesis
heal.classification	Control theory	en
heal.classification	Electric power systems--Periodicals	en
heal.classification	System identification	en
heal.classification	Renewable resources	en
heal.classificationURI	http://id.loc.gov/authorities/subjects/sh85031658
heal.classificationURI	http://id.loc.gov/authorities/subjects/sh2008119240
heal.classificationURI	http://id.loc.gov/authorities/subjects/sh85131740
heal.classificationURI	http://skos.um.es/unescothes/C03349
heal.language	en
heal.access	free
heal.recordProvider	ntua	el
heal.publicationDate	2015-10-19
heal.abstract	The process of arriving to a dynamical model (difference or differential equations) from observed data is called System Identification and is of great importance in the area of control theory. The complexity of modern power systems, which is augmented by the increasing penetration of renewable resources along with the necessity to secure power system's reliability, has been a strong motivation for extensive research in control and system identification area in particular. The objective of the thesis is two-fold: 1) provide an insight to different identification methods by presenting the mandatory related theoretical background and 2) apply them to synthetic time domain data derived from simulations using PSAT software in order to identify dynamic systems. The first study is based on a 24-state space representation of Kundur power system and is an introductory and illustrative example for the better understanding of the identification techniques. ARX, ARMAX and ERA methods are applied, identified models are extracted while for validation purposes different input signals are introduced and finally a comparison between the eigenvalues of the identified and the original model is conducted. The second test case corresponds to a wind farm model in a 14-bus system. After selecting the proper input-output signals for the identification, the three aforementioned methods are applied and the identified models are constructed. Validation with completely different wind speed time series present accurate estimates and Hankel Singular Value Decomposition (a mode truncating technique) is applied for the reduction of the model order and consequently, for the additional simplification of the system. As a final step for the assessment of the results, measurement noise is introduced to the output signals, the methods are applied and residual tests are conducted for every identification procedure.	en
heal.advisorName	Χατζηαργυρίου, Νικόλαος	el
heal.advisorName	Vanfretti, Luigi	en
heal.committeeMemberName	Χατζηαργυρίου, Νικόλαος	el
heal.committeeMemberName	Παπαθανασίου, Σταύρος	el
heal.committeeMemberName	Γεωργιλάκης, Παύλος	el
heal.academicPublisher	Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Ηλεκτρολόγων Μηχανικών και Μηχανικών Υπολογιστών. Τομέας Ηλεκτρικής Ισχύος	el
heal.academicPublisherID	ntua
heal.numberOfPages	99 σ.	el
heal.fullTextAvailability	true