Wind speed estimation of Bingöl province using deep learning method

The global increase in energy demand has brought about the efforts of countries to turn to renewable energy sources and use these resources in the most effective way. Therefore, the use of technological opportunities that exist in every aspect of life and make life easier has an important role in this search. Studies have demonstrated the technology's ability to effectively use renewable energy sources and even its potential to predict how much energy can be generated before using them. It is very difficult to predict the wind speed and therefore the power to be obtained from the wind. It is unstable in nature and can vary depending on many parameters. Many factors such as time, season, temperature, humidity, and weather conditions affect wind speed. Therefore, being able to accurately predict the energy that can be produced from wind is an important issue for energy producers. For production planning, it is necessary to make the best estimate with the lowest margin of error. Today, the best method to estimate wind power and wind speed under these conditions is predictions made with deep learning methods. Future wind speed predictions can be made with deep learning architectures trained with the wind speed data of the region from previous periods. In this study, the wind speed data of Bingöl province within a certain date range was processed with the Convolutional Neural Network (CNN) model and a short-term wind speed forecast was made. In the thesis study, short-term wind speed prediction was made by processing the wind speed data measured between 01.01.2020 and 01.02.2021 of Bingöl province, obtained from Bingöl Meteorology Directorate, with deep learning methods. The study provides information about the wind potential of the region and the wind power that can be produced. As a result of the study, the CNN model, which is a deep learning method, was trained with past wind speed data of the region and the predictions it produced about the future wind speed of the region were observed. The study provides information about the wind potential of the region to be used in many areas by energy investors and researchers.