Combination of morphology edge detection and skeletonization in detecting time location of power disturbances

This paper proposes a new approach for detecting the time location of power disturbances by combining skeletonization and morphology edge detection. Signals with disturbances were filtered using morphology edge detection to find the time location of the disturbances; however the results were not very accurate. Adding skeletonization to the system after applying the morphology edge detection improved the accuracy in detecting the time location of the disturbances. A Matlab simulation has been undertaken and the results show that the proposed method has the capability to detect power quality issues more accurate than the morphology edge detection method for both noise-free signals and signals that contain noise. A reliability analysis has shown that the proposed method produces accurate results when detecting the changing of a block signal.