Prediction of Photovoltaic Panels Output Performance Using Artificial Neural Network

Prediction of Photovoltaic Panels Output Performance Using Artificial Neural Network

Abdelouadoud Loukriz, Djamel Saigaa, Abdelhammid Kherbachi, Mustapha Koriker, Ahmed Bendib, Mahmoud Drif
Copyright: © 2022 |Pages: 19
DOI: 10.4018/IJEOE.309417
Article PDF Download
Open access articles are freely available for download

Abstract

To ensure the safe and stable operation of solar photovoltaic system-based power systems, it is essential to predict the PV module output performance under varying operating conditions. In this paper, the interest is to develop an accurate model of a PV module in order to predict its electrical characteristics. For this purpose, an artificial neural network (ANN) based on the backpropagation algorithm is proposed for the performance prediction of a photovoltaic module. In this modeling approach, the temperature and illumination are taken as inputs and the current of the mathematical model as output for the learning of the ANN-PV-Panel. Simulation results showing the performance of the ANN model in obtaining the electrical properties of the chosen PV panel, including I–V curves and P–V curves, in comparison with the mathematical model performance are presented and discussed. The given results show that the error of the maximum power is very small while the current error is about 10-8, which means that the obtained model is able to predict accurately the outputs of the PV panel.
Article Preview
Top

Introduction

Today, energy is at the heart of the economy for all countries as well as the basis for all human activity. Over the years, energy resources have diversified in order to meet ever-increasing needs. The developed countries have moved from wood to coal, to hydrocarbons, to hydroelectricity, and finally to nuclear power. However, the use of fossil fuels is responsible for acid rain and global warming (IEA, 2019). On the other hand, the exploitation of nuclear energy presents risks of serious accidents, in addition to those induced by the management of the resulting waste, which can be radioactively dangerous for several thousand years (Compaan, 2006),(Panchenko, 2021). Numerous countries have committed themselves under the Kyoto protocol to minimize their greenhouse gas emissions by 8% (Luther, 2005). Renewable energy such as wind, solar, hydropower and biomass have an important role to play in achieving this goal. More particularly, the solar energy resource starts to have a significant contribution to global energy production due to the low cost of maintenance and recurring operation of the photovoltaic systems (Luther, 2005) (Antonanzas et al., 2016).

In the photovoltaic field, the manufacturers provide notations for PV modules under various metrological conditions. However, these conditions are not always evident, which rarely occur in outdoor conditions, as they are mostly performed under laboratory conditions using solar simulators (Naeijian et al., 2021). To this end, the accurate and efficient modeling of solar photovoltaic modules represents one of the most important and difficult problems in the field of photovoltaic systems. These problems are mainly caused by the non-linear characteristics of solar cells, and by the lack of availability of all their parameters (Cortés et al., 2020). Several mathematical models have been developed to characterize the PV module under different working conditions; the widely used being the diode-based model. The most well-known models in the literature are single diode and double diode. The model of the single diode is the simplest model that introduces 5 unknown parameters (Chegaar et al., 2001)(Cardenas et al., 2017)(Villalva et al., 2009), while the double diode model takes into account more features than the double diode model with 7 unknown parameters (Chan & Phang, 1987)(Mathew et al., 2018)(Ishaque et al., 2011). However, the main problem that should be dealt with is the estimation of the unknown parameters of the PV panel model.

Complete Article List

Search this Journal:
Reset
Volume 12: 1 Issue (2024): Forthcoming, Available for Pre-Order
Volume 11: 4 Issues (2022): 2 Released, 2 Forthcoming
Volume 10: 4 Issues (2021)
Volume 9: 4 Issues (2020)
Volume 8: 4 Issues (2019)
Volume 7: 4 Issues (2018)
Volume 6: 4 Issues (2017)
Volume 5: 4 Issues (2016)
Volume 4: 4 Issues (2015)
Volume 3: 4 Issues (2014)
Volume 2: 4 Issues (2013)
Volume 1: 4 Issues (2012)
View Complete Journal Contents Listing