Very short-term wind power density forecasting through artificial neural networks for microgrid control

The aim of this study was to develop an artificial intelligence-based tool that is able to predict wind power density. Wind power density is volatile in nature, and this creates certain challenges, such as grid controlling problems or obstacles to guaranteeing power generation capacity. In order to ensure the proper control of the traditional network, energy generation and demand must be balanced, yet the variability of wind power density poses difficulties for fulfilling this requirement. This study addresses the complex control in systems based on wind energies by proposing a tool that is able to predict future wind power density in the near future, specifically, the next 10 min, allowing microgrid's control to be optimized. The tool is validated by examining the root mean square error value of the prediction. The deviation between the actual and forecasted wind power density was less than 6% for 81% of the examined days in the validation step, from January 2017 to August 2017. In addition, the obtained average deviation for the same period was 3.75%. After analysing the results, it was determined that the forecaster is accurate enough to be installed in systems that have wind turbines in order to improve their control strategy. (c) 2019 Elsevier Ltd. All rights reserved.