Reduction of harmonic monitors and estimation of voltage harmonics in distribution networks using wavelet analysis and NARX

Abstract

The use of power quality monitors is necessitated for optimizing power systems to improve efficiency and decrease power system-related failures. These monitors are expensive and their installation at all buses is not economic. With regard to optimum allocation approaches, power quality monitors should be installed for the assessment of the power quality of sensitive loads. Monitoring of sensitive loads is eliminated in this paper, and an estimator is used for virtually monitoring voltage harmonics. This paper presents an intelligent system based on the nonlinear autoregressive model with exogenous input (NARX) and wavelet analysis. The wavelet transform is utilized for extracting the feature of waveforms, and NARX is employed to detect the voltage harmonic distortion waveforms (VHDWs) at non-monitored sensitive loads in distribution networks using the information gained through the nearest monitoring data. The proposed estimator can be considered as a virtual monitor, eliminating the monitor of sensitive load such that the optimum number of power quality monitors and monitoring cost of the distribution network would be reduced. The estimator is examined on the IEEE test system and the performance of the intelligent system has been verified in many different ways.