Maximum Power Point Estimation for Photovoltaic Systems Using Neural Networks

Solar panels are the power sources in photovoltaic applications which provide electrical power. Solar panel characteristics depend on environmental conditions (solar radiation level, temperature and etc.). In this paper, estimation of maximum power point of silicon solar panels is presented. We applied two different neural networks (back propagation and RBF) for the purpose of estimation in different environmental conditions. These neural networks estimate Maximum power point of solar panels accurately. We used Matlab environment for the purpose of simulation, training and evaluation of these neural networks. It is shown that the responses of RBF neural network are faster and more accurate than back propagation.