Modification of the Adaptive Moving Average Filter for the Signal Parameters Measurement

Moving average filters are among the most popular types of filters used in measuring transducers in the electric power grid. Adaptive modification (with adjustment of the filter length to the interference frequency) allows achieving high measurement accuracy the root mean square value, active power, and reactive power for appropriate measurement methods of these parameters. However, in order to achieve high metrological accuracy, it is necessary to either increase the sampling rate or cascade several identical or different filter stages. The first approach leads to an increase in the computational load on the digital signal processing device used in the implementation. The second one leads to an increase in the filter setting time, which leads to an increase in the dynamic component of the error. In this paper, we consider an alternative approach to the problem, known as the "vernier" method. This method involves approximating the signal for the final sampling step over the entire averaging time. We consider the application of zero-order, first-order, and second-order polynomials, and propose algorithms for their practical implementation. Additionally, we propose variants for constructing simulation models of filters to estimate the error caused by incomplete interference filtering. The magnitude response of the modified moving average filter have been performed, and an assessment of the implementation complexity has been appreciated.