Wavelet transform and multi-class relevance vector machines based recognition and classification of power quality disturbances

Recognition of power quality events by analyzing the voltage waveform disturbances is a very important task for the power system monitoring. This paper presents an integrated model for recognizing power quality disturbances (PQDs) using a new wavelet multi-class relevance vector machines (WMRVMs). The so-called relevance vector machines (RVMs) is an effective classification tool. The proposed method employs the wavelet transform techniques to extract the most important and significant feature from details and approximation waves. The obtained feature vectors are used for training the RVMs to classify the PQDs. Various transient events are tested, the result shows that the classifier can detect and classify different power quality signal correctly. Finally, sensitivity of the proposed algorithm under noise condition is investigated in this paper.