Solar radiation forecasting using artificial neural network for local power reserve

RÉSUMÉ / ABSTRACT :

Renewable energy sources have a variable nature and are depending on weather conditions. To satisfy the instantaneous balancing between electrical production and consumption, the operation of electrical networks requires a power reserve to compensate unforeseen imbalances. This one must be minimized in order to reduce the system cost while ensuring a satisfying security level. A back propagation Artificial Neural Network (ANN) is proposed to predict the Global Solar Radiations (GSR). Predictions have been analyzed according to weather classification with some error indexes, which are also used to evaluate performances. These forecasting results can be used for power reserve  quantification by analysis of forecasting uncertainty errors of both generation and load.