2664-Article Text-5499-3-10-20190212.pdf (1.14 MB)
Enhanced recurrent neural network for short-term wind farm power output prediction
journal contribution
posted on 2023-06-09, 17:09 authored by Ethelbert Chinedu Eze, Chris ChatwinChris ChatwinScientists, investors and policy makers have become aware of the importance of providing near accurate spatial estimates of renewable energies. This is why current studies show improvements in methodologies to provide more precise energy predictions. Wind energy is tied to weather patterns, which are irregular, especially in climates with erratic weather patterns. This can lead to errors in the predicted potentials. Therefore, recurrent neural networks (RNN) are exploited for enhanced wind-farm power output prediction. A model involving a combination of RNN regularization methods using dropout and long short-term memory (LSTM) is presented. In this model, the regularization scheme modifies and adapts to the stochastic nature of wind and is optimised for the wind farm power output (WFPO) prediction. This algorithm implements a dropout method to suit non-deterministic wind speed by applying LSTM to prevent RNN from overfitting. A demonstration for accuracy using the proposed method is performed on a 14-turbines wind farm. The model out performs the ARIMA model with up to 80% accuracy.
History
Publication status
- Published
File Version
- Published version
Journal
IJRDO - Journal of Applied ScienceISSN
2455-6653Publisher
IJRDOPublisher URL
Issue
2Volume
5Page range
28-35Department affiliated with
- Engineering and Design Publications
Research groups affiliated with
- Industrial Informatics and Signal Processing Research Group Publications
Full text available
- Yes
Peer reviewed?
- Yes