New Algorithms for the Estimation of Distorted Time-varying Power System Signal Parameters in Noise

Abstract

This article proposes an improved two-step augmented Gauss-Newton type algorithm for the estimation of distorted power system signals. The proposed algorithm is capable of accurately estimating the amplitude, phase, and frequency of the harmonics and interharmonics present in the power system signal in the presence of high additive white Gaussian noise. The estimation of all the parameters is divided into two steps. In the first step, the frequency of the signal is computed using an improved recursive Newton-type algorithm; in the second step, the amplitude and phase of the power system voltage or current are computed using a recursive Gauss-Newton-type algorithm. Further, to reduce the computational complexity of the standard Gauss-Newton algorithm, certain approximations are carried out, making the algorithm suitable for on-line applications. Also, a variable forgetting factor is used to improve the speed of convergence and accuracy of the algorithm. To demonstrate the robustness and accuracy in the estimation of the distorted fundamental and harmonic phasors using the proposed algorithm, the results obtained from computer simulations are presented.