Detection and Analysis of Power Quality Variations for initiating control actions in DGs

Identification and classification of Power Quality (PQ) events have been a subject of recent interest for researchers from the point of view of initiating suitable control actions in Distributed Generation Systems (DGs) for achieving improved performance. This requires carrying out a detailed analysis of the various characteristics of real time electrical signals through a set of signal processing techniques. In this context, two major parameters are considered to arrive at suitable control actions needed for monitoring and control of DGs. These parameters include PQ distortions due to environmental factors, such as change in wind speed and solar irradiations. Signal features are extracted using S-Transform while signal-classification is done by Least Square Support Vector Machine (LS-SVM) technique. A 17-bus test system is modeled using the open source software, Open Distribution System Simulator (OpenDSS). LS-SVM and S-Transform are implemented using MatLab. Smart converter control is realized with inputs received from signal classifier, so as to initiate proper grid-support functions.