Open AccessShort-term forecasting of wind and solar power generation
Author(s) -
В.З. Манусов,
S.K. Khaldarov,
Boris Palagushkin
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2131/5/052050
Subject(s) - renewable energy , probabilistic forecasting , wind power , time horizon , computer science , probabilistic logic , wind speed , term (time) , solar power , environmental science , meteorology , reliability engineering , power (physics) , engineering , electrical engineering , mathematical optimization , artificial intelligence , mathematics , geography , physics , quantum mechanics
It is important to note that the generated power of renewable sources depends on the natural conditions at a particular geographic point, the level of wind flow speeds and solar radiation. The patterns characterizing these parameters depend on the time of year, locality and are purely probabilistic in nature. Taking into account the above-mentioned conditions for the effective implementation of “green” objects in the power supply system, the purpose of this work is to build forecasting models that are more likely to be able to determine what part of the load can be covered by the power supply system based on wind power and solar installations. This purpose was achieved by constructing and training artificial neural networks with data on the speed of wind flow and solar radiation obtained from real renewable energy facilities. The most significant result is the identification of the necessary forecasting horizon, taking into account the preservation of a relatively good quality of metrics, as well as understanding what additional data is required to improve this quality. The significance of the results obtained lies in the fact that they make it possible to determine what reserve capacity is required to be included in the project.