Sparse Bayesian Harmonic State Estimation

Abstract

This paper presents a novel iterative method for harmonic state estimation based on the sparse Bayesian learning framework. The proposed method can locate harmonic sources and estimate the distribution of harmonic voltages using fewer harmonic meters than buses, despite the strong correlation between the columns of the system matrix. Extensive simulations are performed on a benchmark transmission system to corroborate the efficacy of this method when measurements are noise free. Our results show that the proposed state estimator achieves an identification error of less than $1.6\times 10^{-6}$ and can locate harmonic sources with an average success rate of 97.92%, outperforming state-of-the-art harmonic state estimators.