Large Scale, Mid Term Wind Farms Power Generation Prediction

Renewable energy sources, such as wind turbines, have become much more prevalent in recent years, and thus a popular form of generation. This is part due to the ‘Fit for 55’ EU initiative, part, rising fossil fuel prices, well perceived requirement nations power independence, influence renewable sources we can see marked increase large farms particular. However, by their very nature are highly inconsistent regarding generation weather-dependent, presenting several challenges transmission system operators. One options overcome these issues being able forecast generated wide-ranging period—ranging from 15 min up 36 h, with an adequate resolution. Such would better help manage grid allow greater utilization green produced. In this document, present process development system, along comparison various steps process, including data preparation, feature importance analysis, impact on horizon. Lastly, also compare multiple machine learning models quality execution time. Additionally, propose ensemble that concatenates predictions over The conducted experiments been evaluated seven located Central Europe. Out conducted, most efficient solution lowest error rate required computational resources has obtained random forest regression, two independent models; one short-term horizon, other, mid- long-term which was combined into forecasting system.