A Novel Digital Signal Processing Algorithm for On-line Assessment of Power System Frequency

An innovative, computationally efficient digital signal processing algorithm has been proposed to evaluate the fundamental frequency of a non-sinusoidal power system signal at every sample instant. This approach adopts backward difference approximation as second derivative FIR filter to estimate the instantaneous frequency. Besides this, an efficient square root algorithm, based on iterative Newton-Raphson inverse (NRI) method is also presented to get high accuracy in fixed point processor. The method is illustrated and evaluated in real time by means of a Texas Instruments (TI) TMS320VC5416 digital signal processor (DSP). The experimental results confirm the validity and accurate performance of the proposed approach even under slow magnitude and frequency variations.