MAXIMUM POWER POINT TRACKING FOR PV SYSTEMS USING ARTIFICIAL NEURAL NETWORKS
Tiago Targino Sepulveda1; Luciana Martinez1; André Pires Nóbrega Tahim1
1 Universidade Federal da Bahia
Baixar PDF doi:10.20906/CPS/CBA2018-0327
Resumo
This paper proposes a method of maximum power point tracking for photovoltaic (PV) panels using neural networks. The PV system consists of a solar panel, a DC-DC converter, a control system and the load. Two neural networks will be trained to integrate into the system. The rst network is responsible for estimating the level of solar irradiance from the electric current, voltage and temperature signals of the solar panel. The second neural network is connected to the rst and uses its output (irradiance) with the temperature to generate a reference voltage, corresponding to the maximum power voltage, to a PI controller. The system responses under variable conditions of irradiance, temperature and load will be analyzed, as well as a performance comparison with the incremental conductance method.
Palavras-chave: Photovoltaic system; Maximum power point tracking; Arti cial neural network