DocumentCode :
3778854
Title :
MLP-neural network based detection and classification of Power Quality Disturbances
Author :
Swapnil B. Mohod;Vilas N. Ghate
Author_Institution :
Electrical Engineering Department, S.G.B. University, Amravati, India
fYear :
2015
Firstpage :
124
Lastpage :
129
Abstract :
This paper deal with the innovative method for the detection and classification of power quality disturbance events with good classification accuracy. Proposed method is reliable to distinguish normally occurring six types of disturbances. Extreme Proliferation of automated power electronics equipments are hampered the quality of power supply. Hence diagnosis of these disturbances within a stipulated time is an urgent need. Deterioration of power quality often termed as Power Quality (PQ) Disturbance. This paper presents a Multilayer Perceptron Neural Network based classifier for effective classification of power quality disturbances. For feature extraction and dimensionality reduction Wavelet Transform technique and Sensitivity Analysis are used respectively. Optimised classifier classifies the six fundamental PQ disturbances with classification accuracy of 99.81%.
Keywords :
"Training","Testing","Artificial neural networks","Power quality","Feature extraction","Wavelet transforms"
Publisher :
ieee
Conference_Titel :
Energy Systems and Applications, 2015 International Conference on
Type :
conf
DOI :
10.1109/ICESA.2015.7503325
Filename :
7503325
Link To Document :
بازگشت