Application of a Combined Scheme for Estimating Power System Frequency from Distorted Signals

Abstract In present-day power systems, fast and accurate frequency estimation has become vital. Any mistake in accurate estimation of frequency could lead to system operation problems or system instability. Fast and accurate frequency estimation has become even more important due to the increasing level of harmonics and distortions and the need to be able to control the system more precisely. The level of system harmonics is increased in modern power systems. In addition, the presence of outside inference and noise coming from the environment and instruments are unavoidable. Accuracy of frequency estimation algorithms would be improved considerably if the effect of harmonics could be reduced and the signal noise diminished. In this article, a new combined scheme for estimating the frequency of a power system is presented and analyzed. The proposed method is based on a combination of the least square and zero crossing methods. An appropriate wavelet signal de-noising scheme is applied as well. Some post-processing units are also considered to modify the calculated results. The performance of the proposed scheme is evaluated using different data. Various tests using simulated samples and real data recorded from a power system are performed, and capabilities of the proposed algorithm are investigated. The obtained results confirm the capabilities of the proposed method.