Robust Pelican Optimization Approach for Single-Diode Photovoltaic Module Parameter Estimation

Abstract

Single-diode photovoltaic (PV) models require the accurate estimation of the parameters to predict the performance of the photovoltaic with reliability and maximum power point tracking(MPPT). The high level of nonlinearity of PV characteristics and unpredictability of operating conditions pose challenges to the traditional optimization methods. This paper recommends the Pelican Optimization Algorithm (POA), a nature-based metaheuristic based on the pelican hunting behavior, to approximate the five most important single-diode PV model parameters: photocurrent, diode saturation current, series resistance, shunt resistance and ideality factor. POA is formulated to establish an efficient balance in the world between global exploration and local exploitation resulting in a speedy and stable convergence to quality solutions. The proposed method is checked based on measured and simulated I-V and P-V curves, and there is an agreement indicating the correctness of the identified parameters. The root mean square error (RMSE) values of both current and power are very low which implies significant amount of numerical accuracy in the estimates of the parameters. Additional experiments during different levels of irradiance and temperature also provide consistently small errors, which proves how well POA is resistant to changes in the environment. Altogether, the findings show that POA is an effective and reliable PV parameter identification, outperforms the traditional optimization approaches, and has a good potential to be used in real-time PV models, control, and MPPT.