Neural network based maximum power point tracking of photovoltaic arrays

Photovoltaic energy has become the most prospective source of renewable energy at this moment of energy crisis in the world. To utilize this energy efficiently a photovoltaic array should be used at its maximum power point (MPP) all the time. Therefore, introducing an effective technique of tracking MPP is of immense importance. There are various techniques to track MPP reported in several literatures. Among them artificial neural network based MPP tracking is of less concern. In this paper a novel technique of tracking MPP using backpropagation neural network is proposed. The proposed model is designed and simulated in Matlab/Simulink. The design considers the change in solar irradiance and ambient temperature successfully tracks the maximum power point more accurately and quicker than conventional perturb and observe based controller and even better than fuzzy logic based controller in these situations. The overall test results validate the efficiency of the model as well as the backpropagation neural network method which can be used in related research works.