Quaternions-based detection of power quality events

This paper presents a new method for the detection of voltage disturbances caused by various reasons in households and industries. The aim of the paper is to improve the power quality disturbance methods and implement it in real-time with a very high accuracy and use existing mitigation techniques after classification of the signal. The method involves a novel idea to use the quaternions for the purpose of signal tracking. The envelope formed by Hilbert envelope detection method is used as an input for the quaternions. The quaternions produce a spike whenever there is a deviation from the pure sine wave giving very accurate results in a very short period of time. Combination of SVM and DSmT classifier is used for recognizing and classification of different signals such as sags, swells, transients and other power quality disturbances, hence can be used to monitor the power quality.