Applying a Moving Average Filter to Reduce the Frequency Measurement Error for the Case Applying by the Zero Crossing Technique

Abstract

The most popular method for the frequency measurement of electrical power grid signals is the zero crossing technique. The main reason for the measurement error of this method is the presence of noise. In the presence of harmonics and flicker, the sensitivity of the method to flicker may increase. One method to reduce the effects of noise and harmonics is the input signal pre-filtration by a low-pass filter. The paper discusses the application of moving average filter (MAF) to solve this problem. A variant of constructing a band-pass filter (BPF) based on a recursive MAF is proposed. Approaches to adaptive adjustment of the order of recursive and non-recursive MAF filter to the current (measured) value of the input signal frequency are considered. It is shown that the synchronization of the moment when the filter order changes and the moment when the output signal crosses zero allows to reduce the duration of the transient process to one sample time. The influence of the number of MAF stages on the frequency response of the resulting BPF and settling time is analyzed. By application of the Simulink software package, a BPF simulation model was performed on the basis of a recursive MAF.