An ultra-short-term wind power prediction model independent of meteorological data is proposed to solve the problem that there are some wild farms do not have meteorological sensors. The model consists of a two-stream network module and an attention module, one stream is used to extract spatial and time series features, the other stream is used to extract temporal features; the features of the two streams are fused with the attention mechanism to get final feature. Finally, the dimension of the final feature is reduced by the fully connection layers to get ultra-short-time wind power prediction. The validity and practicability of the prediction model are proved by an example analysis of a wind farm in east China for one year, the accuracy of prediction is high, which also provides a strong support for making power generation plan and power dispatching in actual scenarios.
The accurate prediction of wind power is related to the accuracy of power generation plan and overall scheduling, and wind power is a clean energy with randomness, intermittence and volatility. Therefore, we propose a novel short-term wind power prediction method based on hybrid deep neural network. This method does not require meteorological data and only uses a two-stream framework. The spatial features were extracted by using one-dimensional convolutional network, the long short-term memory network (LSTM) was used to extract temporal features, and then fuses the two parts of features for prediction. Through the analysis of three years' data of a wind farm in east China, we prove the validity and practicability of the prediction model, which provides strong support for the reliability analysis of power prediction.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.