AN ADAPTIVE MEDIAN POST-FILTER FOR IMPEDANCE ESTIMATION BASED ON DIFFERENTIAL EQUATION ALGORITHM
Rastko Živanović · Adelaide Research & Scholarship (AR&S) (University of Adelaide) · 2005
Abstract – This paper presents an application of the adaptive median post-filter to improve impedance estima-tion based on the differential equation algorithm. The main feature of the post-filter is the automatic selection of the optimal window size that balances between fast track-ing and filtering performance. For sudden impedance change (fault inception), small window size is selected to make fault detection fast. Large window size is selected if the impedance is slow varying or steady-state (pre-fault and fault conditions) to increase filtering efficiency. The post-filter is also very efficient in eliminating impulsive noise created by numerical instability of the differential equation algorithm under certain conditions. The simula-tion results presented in the paper show the adaptive me-dian post-filter enhances filtering performance of the differential equation algorithm while preserving fast tracking of the impedance during fault inception.