Spatio-temporal wind power prediction using recurrent neural networks
Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt
While wind is an abundant source of energy, integrating wind power into existing electricity grids is a major challenge due to its inherent variability. The ability to accurately predict future generation output would greatly mitigate this problem and is thus extremely valuable. Numerical Weather Prediction (NWP) techniques have been the basis of many wind prediction approaches, but the use of machine learning techniques is steadily gaining ground. Deep Learning (DL) is a sub-class of machine learning which has been particularly successful and is now the state of the art for a variety of classification and regression problems, notably image processing and natural language processing. In this paper, we demonstrate the use of Recurrent Neural Networks, a type of DL architecture, to extract patterns from the spatio-temporal information collected from neighboring turbines. These are used to generate short term wind energy forecasts which are then benchmarked against various prediction algorithms. The results show significant improvements over forecasts produced using state of the art algorithms.
Originalsprog | Engelsk |
---|---|
Titel | Neural Information Processing - 24th International Conference, ICONIP 2017, Proceedings |
Redaktører | Dongbin Zhao, Yuanqing Li, El-Sayed M. El-Alfy, Derong Liu, Shengli Xie |
Antal sider | 8 |
Forlag | Springer Verlag, |
Publikationsdato | 1 jan. 2017 |
Sider | 556-563 |
ISBN (Trykt) | 9783319701387 |
DOI | |
Status | Udgivet - 1 jan. 2017 |
Eksternt udgivet | Ja |
Begivenhed | 24th International Conference on Neural Information Processing, ICONIP 2017 - Guangzhou, Kina Varighed: 14 nov. 2017 → 18 nov. 2017 |
Konference
Konference | 24th International Conference on Neural Information Processing, ICONIP 2017 |
---|---|
Land | Kina |
By | Guangzhou |
Periode | 14/11/2017 → 18/11/2017 |
Navn | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
---|---|
Vol/bind | 10638 LNCS |
ISSN | 0302-9743 |
ID: 223196128