Taylor-Gaussian-Fourier Filters Precise, Fast, and Accurate Frequency Estimation Technique

Accurate frequency estimation is crucial for control and protection applications. However, many existing frequency estimation algorithms face reliability issues when applied to real-time, dynamically varying signals. This paper presents a fast, reliable, and precise method for estimating frequency on the data enriched with DC decaying components, harmonics, noise, and off-nominal frequency. The proposed technique utilizes a Taylor Fourier series-based algorithm combined with a Gaussian filter to effectively remove noise, harmonics, and the DC decaying component from the signals. The results of the proposed method have been compared with the alternative approaches, including Kalman pitch tracking, DFT interpolation, and least squares frequency estimation, demonstrating improved accuracy up to the fifth decimal place. Our findings indicate that this method significantly enhances accuracy compared to the techniques mentioned. Simulations to assess the proposed approach show that it can rapidly and precisely estimate the frequency under various operating conditions. The simulation results reveal that the proposed method's frequency error (FE) consistently stays below 0.0000777 Hz.