1- School of Electrical and Computer Engineering, Shiraz University, Shiraz, Iran
2- School of Electrical and Computer Engineering, Shiraz University, Shiraz, Iran, m.mohammadi@shirazu.ac.ir.
Abstract: (5762 Views)
Since it is essential to deliver smoothed sinusoidal voltage to the customers, diagnosing power quality (PQ) events has played important role in power delivery and conversion. This diagnostic scheme should be accurate to classify PQ events from other events in power system. Also it should be fast enough to rapidly mitigate PQ events. In this paper, an algorithm based on Core Vector Machine (CVM) has been introduced to classify power quality events. Feature selection method has also been applied to increase the accuracy of the classifier’s algorithm. Some features have been selected among several others extracted by wavelet transform. In addition, eight different classes are simulated due to the corresponding equations used in previous studies. Evaluating the performance of the algorithm, different indices have been used to assess the operation of the classifier’s algorithm. Simulations results show the robust capability of the proposed algorithm to classify the PQ events
Received: 2015/10/20 | Accepted: 2015/12/26 | Published: 2016/03/5