Time-Domain Identification of the Power Quality Disturbances Based on the Event-Driven Processing

The Power quality (PQ) disturbances originates problems in smart grids and industries. The identification and prevention of such disturbances is mandatory. This paper suggests an original approach, based on event-driven processing, for time-domain PQ signals features extraction and recognition. The incoming PQ signal is digitized with an event-driven A/D converter (EDADC). A novel selection mechanism is employed to efficiently segment the EDADC pertinent output. Later on, features of these segments are explored by performing only the time-domain analysis. The identification is performed with a specifically developed voting based classifier. Results demonstrate a first order of magnitude reduction in collected samples as compared to the traditional counterparts. It aptitudes a significant processing and power consumption effectiveness of the designed solution compared to the counterparts. The proposed system attains an average recognition accuracy of 98.06%, for the case of a three class PQ disturbances. It demonstrates the benefit of embedding the proposed solution for the development of effective automatic PQ disturbances recognizers.