Paper title: AN OFFLINE TRAINED ARTIFICIAL NEURAL NETWORK TO PREDICT A PHOTOVOLTAIC PANEL MAXIMUM POWER POINT
Author(s): HOCINE ABDELHAK AZZEDDINE, MUSTAPHA TIOURSI, DJAMEL-EDDINE CHAOUCH, BRAHIM KHIARI,
Abstract: In this work, we develop a radial basis artificial neural network to predict the voltage and the current at maximum power point
of a photovoltaic panel under different cell temperature and solar irradiance conditions. For training the proposed artificial
neural network, we generate a group of maximum power points defined by their corresponding current and voltage values using
the photovoltaic panel single diode model. To ensure the validity of the artificial neural network, we compare the obtained
results to those obtained by using the photovoltaic panel single diode model for cell temperature and solar irradiance conditions
other than those used for the training phase. Results show that the developed artificial neural network can predict accurately
and quickly the current and the voltage of the photovoltaic panel at the maximum power point for any cell temperature and
solar irradiance conditions.
Keywords: Artificial neural network (ANN), Photovoltaic (PV), Maximum power point tracking (MPPT) Year: 2016 | Tome: 61 | Issue: 3 | Pp.: 255-257
Full text : PDF (192 KB) |