An Ensemble Classifier based Power Quality Disturbances Classification
Tiagrajah V. Janahiraman1, Prakash Bala2
1Tiagrajah V. Janahiraman, College of Engineering, Universiti Tenaga Nasional, Jalan IKRAM-UNITEN, Kajang, Selangor, Malaysia.
2Prakash Bala, College of Engineering, Universiti Tenaga Nasional, Jalan IKRAM-UNITEN, Kajang, Selangor, Malaysia.
Manuscript received on November 22, 2019. | Revised Manuscript received on December 15, 2019. | Manuscript published on December 30, 2019. | PP:4161-4167 | Volume-9 Issue-2, December, 2019. | Retrieval Number: B4924129219/2019©BEIESP | DOI: 10.35940/ijeat.B4924.129219
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Evolution of the current modern era demands a huge and good power quality supply day by day. Power utility suppliers and power exchange specialist organizations face a noteworthy test in recognizing the kind of Power Quality Disturbances (PQD). Our research illustrates the technique of PQD classification by utilizing wavelet signal decomposition and Ensemble classification. A normal wave without disturbance and waves with PQD events of single-type and hybrid-type were generated using MATLAB using the mathematical model as per the definition and parameters outlined by IEEE 1159 and IEC61000 customary. Discrete Wavelet Transform (DWT) is pertained to decompose the signal form the generated PQD to get the illustration in time and frequency domain. In this research work, our database consists of 14000 generated signals of a normal wave and the PQDs, which were divided into 80% for the train set and 20% for the test set for each PQDs. An ensemble methodology for multiclass order was chosen as the classifier of the component vector for the PQD. Examinations were conjointly made with elective sorts of classifiers and different kinds of mother wavelet channel capacities to observe and investigate the exhibition qualification. The outcomes demonstrated that the blend of DWT and Ensemble Classifier delivers an optimal solution to recognize the class of PQD with a precision of 100% for each train and test set.
Keywords: Power Quality, Ensemble Method, Discrete Wavelet Transform.