Harmonic distortion prediction model of a grid - Connected photovoltaic using grey wolf optimizer – Least square support vector machine

Abstract

This paper depicts a new technique for prediction of the total harmonic distortion (THD) in Grid-Connected Photovoltaic System. Global environmental awareness, increasing demand for energy and down price tendency has led to new opportunities for utilization of renewable energy resources such as photovoltaic (PV) system. The integration of PV system to the grid must comply with the relevant standards given by the utility company. However, the output of PV somehow causes a harmonic distortion as the installation of inverter. The output of PV mainly depends on solar irradiation. Therefore, solar irradiation is selected as one of the input to the prediction model. The hybridize method of heuristic-algorithm namely Grey Wolf Optimizer-Least Square Support Vector (GWO-LSSVM) is introduced in order to improve the prediction accuracy. GWO is inspired by the leadership hierarchy and hunting mechanism of grey wolf in nature. The top hierarchy of grey wolf that considered as the fittest solution is alpha, followed by beta, delta and omega. The optimization process implementing three main steps such as hunting, searching for prey, encircling prey and attacking prey. GWO is utilized to optimize the parameters in LS-SVM model. The results showed that GWO-LSSVM predict more accurate than PSO-LSSVM and LSVM. © 2019 IEEE.