Detection and location of voltage sags based on mathematical morphology and grille fractal

A detection method based on mathematical morphology filtering and grille fractal is proposed to detect and locate the power quality disturbance signals which are represented by voltage sag. An improved morphological filter with good performance in suppressing noise is designed in which the least mean square (LMS) arithmetic is adopted to calculate the weight coefficients of open-closing and close-opening filter adaptively. Then, according to the theory of grille fractal, voltage sag signals could be located exactly and the phase variation could be detected by analyzing the change of the grille. The simulation and actual application results show that the proposed method is effective and valid. Moreover, the grill number can be applied as an auxiliary measure for the detection of different disturbances of voltage sags and compensation strategy to improve the power quality.