- Karadeniz Fen Bilimleri Dergisi
- Vol: 7 Issue: 1
- K-Nearest Neighbor Classification of Harmonics Using Akaike Information Criterion
K-Nearest Neighbor Classification of Harmonics Using Akaike Information Criterion
Authors : Özgür TOMAK, Onur Özdal MENGİ
Pages : 1-8
View : 13 | Download : 3
Publication Date : 2017-06-15
Article Type : Research
Abstract :If power quality can be maintained, high performance is possible for electrical devices. Harmonics is one of such problems that can cause performance to drop. Therefore, harmonics must be detected and prevented. This study aimed to be part of detection system designed to maintain power quality. Akaike Information Criterion is used to calculate features for power quality analysis. And for classification k-Nearest Neighbor classification is used. %94.8 accuracy is obtained from training set and %91.7 accuracy is obtained from test set. MATLAB is the program which is used for classification and feature calculation.Keywords : Akaike Information, Harmonic Detection, k-Nearest Neighbor Classification, Power Quality