Optimal Harmonic Estimation Using A Particle Swarm Optimizer

This paper presents a new algorithm for harmonic estimation. It utilizes the particle swarm optimizer with passive congregation (PSOPC) to estimate the phases of the harmonics, alongside a least-square (LS) method that is used to estimate the amplitudes. The PSOPC and LS method are executed alternately to minimize the error between the original signal and the signal reconstructed from the estimated parameters during the estimation process. Simulation results are presented to demonstrate that the estimation accuracy is greatly improved in comparison with that of the conventional discrete Fourier transform and genetic algorithms. The proposed algorithm is also used to estimate interhar- monics and the harmonics with frequency deviation. The results show that this new method, working in a corporative manner between PSOPC and LS, is capable of estimating power system integral harmonics and interharmonics, even in the case of the deviation of fundamental frequency.