Optimal Structuring Element Selection Guidelines of Morphological Filter for Power Quality Signal Denoising

The structuring element (SE) is a key component of any mathematical morphology (MM) operation and application dependent. However, there are no clear guidelines for the size selection of the SE for denoising application in power quality issues. To solve this problem, voltage sag and swell signals corrupted by white noise and impulse noise are modeled with considerations of the magnitude and position of impulse noise as well as the fault depth. Detailed numerical simulations are carried out to explore the impact of the semicircular SE with a variety of lengths and heights on the performance of morphological opening–closing–closing–opening average (OCCO) filters. Based on simulation results and observations, the impacts of three different factors on the output signal-to-noise ratio (SNR) and the size of the semicircular SE are summarized. Some guidelines for the size selection of the semicircular SE are proposed, which comprehensively considered the denoising effect and processing speed. It will facilitate the rapid implementation of a high-performance morphological filter compared to trial and error.