Application of computational methods for harmonic state estimation of power system networks

Abstract

In this study, a novel technique is used to estimate the power system harmonic state, as one of the biggest risks in a power system network. Nonlinear loads are widely used, which inject harmonics into a system. Such injected harmonics make networks unstable and increase power loss. The main objective of this work is to develop a new harmonic state estimator system to increase power system accuracy, stability and the wall operation state. Three computational methods are used in this study, that is, the i) proposed particle swarm optimisation-recursive least squares (PSO-RLS) algorithm, which is developed, presented and compared with the ii) discrete fourier transform (DFT) and iii) PSO algorithms. The three algorithms are tested on an IEEE 14-bus system, and simulation results show that the new PSO-RLS algorithm is more accurate than the other two algorithms (i.e. DFT and PSO algorithms), with a lower error percentage. The proposed algorithm is tested to prove its validity and effectiveness in power system networks. The capability of the PSO-RLS algorithm is apparent in the error percentage compared with that of the other two computational methods, which can be used to provide an excellent prediction rate for measurement errors in system buses.