This review-paper focuses on the development the intelligent technology for modelling (Multi-Model Approach (MMA)) and control (Artificial Neural Networks Controller) of grid connected photovoltaic energy conversion system. This approach (MMA) is based on a black box modeling. A database consists of input variables (sunshine, temperature and voltage at the terminals of photovoltaic generator (PVG) and output (PVG current) is obtained by characterization of a photovoltaic module Sharp installed type at the "Polytechnic Higher School" (PHS) in Dakar in March 2012. Indeed 70% of this database is used to train the multi-model and 30% of the database is reserved for validation of the multi-model. The proposed model has a correlation of 89% and a Nash criterion (NS) average of 75.65%. Learning is performed with oil operating area. Each area of operation is made by a local affine model structure and function of validity sigmoid. These results show the good performance of the proposed model. Control design of a single phase grid-connected photovoltaic (PV) system including the PV array and the electronic power conditioning (PCS) system, based on Artificial Neural Networks Controller (ANNC) is presented. The developed controller is compared with a Proportional Integral (PI) controller through computer simulation. The obtained results show that the NNC have faster response and lower THD without overshoots.
Key words: Black box modelling, photovoltaic generator, inverter, maximum power point tracking (MPPT), neural network controller.
Copyright © 2018 Author(s) retain the copyright of this article.
This article is published under the terms of the Creative Commons Attribution License 4.0