Solar irradiance prediction model based on a statistical approach for microgrid applications

Microgrids are smaller grid networks that supply power locally. They are connected to primary power grids and are able to manage electricity between primary and local electricity generation, maximizing the systems efficiency in order to help successfully supply loads. Although some microgrids use diesel generators, with increasing interests in renewable energy sources, interconnections with wind and solar power have become more common. A drawback from using renewable energy sources is the variability in electricity generation due to their dependency on natural resources. In order to integrate these systems into the network, they require models that aid in predicting their behavior. A prediction model based on the Holt-Winters forecasting method is used in order to predict solar irradiance behavior. In order to verify the prediction model, three different energy management strategies are compared using a typical residential microgrid model.