Mostrar el registro sencillo del ítem

dc.contributor.authorRodríguez, Alejandro
dc.contributor.authorAguado-Sánchez, José Antonio 
dc.contributor.authorLópez, José Jesús
dc.contributor.authorMartín, Francisco I.
dc.contributor.authorMuñoz, Francisco
dc.contributor.authorRuiz, Jose Ernesto
dc.contributor.editorEberhard, Andreas
dc.date.accessioned2011-07-04T08:06:45Z
dc.date.available2011-07-04T08:06:45Z
dc.date.issued2011
dc.identifier.isbn978-953-307-180-0
dc.identifier.urihttp://hdl.handle.net/10630/4670
dc.description.abstractThe usual operations on the distribution network such as switching loads and circuits, the proliferation of power electronic equipment and non-linear loads and the distributed generation with renewable energy are several of the most common causes that are leading to an increasing polluted power system in terms of voltage signal distortion. One way of improving the power quality (PQ) parameters consists of analyzing these disturbances efficiently and understanding them deeply and PQ monitoring is one major task in order to achieve it. PQ monitoring is not an easy task usually involving sophisticated hardware instrumentation and software packages. Many recent approaches in PQ monitoring try to achieve it through the automated classification of different disturbances. The different approaches in this field lead their efforts in two directions, the main parts that form an automated classification. The first make focus to obtain a suitable pattern that allow distinguish clearly each disturbance, by the use of time-frequency transforms to get feature extraction, as Wavelet transform (WT) and S-transform (ST). The second is oriented to use a classifier able to assign each disturbance correctly in its class, so the most of the artificial intelligent techniques have been combined with WT or ST, as Artificial Neural Networks (ANN), Decision Tree, Fuzzy Logic, Hidden Markov Model, Support Vector Machines, etc. In this work an automated classification system based on time-frequency transform as a feature extraction tool in combination with Artificial Neural Network (ANN) as algorithm classifier is presented. As feature extraction tool have been used both, WT and ST, in order to make a comparison between them. ANN is discussed as an example of a robust classification algorithm, so it is chosen for obtaining experimental results. In addition, two variants, backpropagation (BP) and probabilistic (PNN) have been used for more completeness of the results. An automated classification system based on Wavelet transform as a feature extraction tool in combination with Artificial Neural Network as algorithm classifier is presented.The most usual PQ disturbances have been generated according to mathematical models to obtain experimental results. Noise is added to the signals from 40dB to 20dB. At last, signals generated by power network simulation using PSCAD/EMTDC environment have been used to check the reliability of the resulting systems based on different time-frequency transforms.es_ES
dc.language.isoenges_ES
dc.publisherInTeches_ES
dc.rightsinfo:eu-repo/semantics/openAccess
dc.subjectEnergía eléctrica - Calidades_ES
dc.subject.otherPower qualityen
dc.subject.otherClassificationen
dc.subject.otherTime-frequency transformsen
dc.titleTime-Frequency Transforms for Classification of Power Quality Disturbancesen
dc.title.alternativeTransformadas Tiempo-Frecuencia para la Clasificación de Perturbaciones Eléctricases_ES
dc.typeinfo:eu-repo/semantics/bookPartes_ES


Ficheros en el ítem

Este ítem aparece en la(s) siguiente(s) colección(ones)

Mostrar el registro sencillo del ítem