Detection and classification of power disturbances using mathematical morphology with trapezoid structuring elements and signal envelopes

Abstract

This paper presents a novel method of multiresolution morphological gradient (MMG), which uses trapezoid structuring elements (SE), for the identification of power disturbances. MMG is a nonlinear morphological transform which presents the gradient information in the output. Another simple morphological operator, closing operator which can extract envelopes of signals, is also applied in combination with MMG for the classification of different types of power disturbances. A variety of power disturbances have been simulated to evaluate the reliability of this hybrid approach and noisy environment has been taken into consideration during simulations. Applied to power systems, MMG extracts the features of disturbances and detects their location and duration. Moreover, with the characteristics extracted by MMG, signal envelopes obtained by closing operator can be another criterion for the classification of power disturbances.