Enhancing Wind Turbine Power Forecast via Convolutional Neural Network
The rapid development in wind power comes with new technical challenges. Reliable and accurate wind power forecast is of considerable significance to the electricity system’s daily dispatching and production. Traditional forecast methods usually utilize wind speed and turbine parameters as the model...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-01-01
|
Series: | Electronics |
Subjects: | |
Online Access: | https://www.mdpi.com/2079-9292/10/3/261 |
id |
doaj-d90c87f6d47c431990cf228f494089e4 |
---|---|
record_format |
Article |
spelling |
doaj-d90c87f6d47c431990cf228f494089e42021-01-23T00:03:58ZengMDPI AGElectronics2079-92922021-01-011026126110.3390/electronics10030261Enhancing Wind Turbine Power Forecast via Convolutional Neural NetworkTianyang Liu0Zunkai Huang1Li Tian2Yongxin Zhu3Hui Wang4Songlin Feng5Shanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, ChinaShanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, ChinaShanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, ChinaShanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, ChinaShanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, ChinaShanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, ChinaThe rapid development in wind power comes with new technical challenges. Reliable and accurate wind power forecast is of considerable significance to the electricity system’s daily dispatching and production. Traditional forecast methods usually utilize wind speed and turbine parameters as the model inputs. However, they are not sufficient to account for complex weather variability and the various wind turbine features in the real world. Inspired by the excellent performance of convolutional neural networks (CNN) in computer vision, we propose a novel approach to predicting short-term wind power by converting time series into images and exploit a CNN to analyze them. In our approach, we first propose two transformation methods to map wind speed and precipitation data time series into image matrices. After integrating multi-dimensional information and extracting features, we design a novel CNN framework to forecast 24-h wind turbine power. Our method is implemented on the Keras deep learning platform and tested on 10 sets of 3-year wind turbine data from Hangzhou, China. The superior performance of the proposed method is demonstrated through comparisons using state-of-the-art techniques in wind turbine power forecasting.https://www.mdpi.com/2079-9292/10/3/261convolutional neural networkGWF matricesripple encodingwind power generation forecast |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Tianyang Liu Zunkai Huang Li Tian Yongxin Zhu Hui Wang Songlin Feng |
spellingShingle |
Tianyang Liu Zunkai Huang Li Tian Yongxin Zhu Hui Wang Songlin Feng Enhancing Wind Turbine Power Forecast via Convolutional Neural Network Electronics convolutional neural network GWF matrices ripple encoding wind power generation forecast |
author_facet |
Tianyang Liu Zunkai Huang Li Tian Yongxin Zhu Hui Wang Songlin Feng |
author_sort |
Tianyang Liu |
title |
Enhancing Wind Turbine Power Forecast via Convolutional Neural Network |
title_short |
Enhancing Wind Turbine Power Forecast via Convolutional Neural Network |
title_full |
Enhancing Wind Turbine Power Forecast via Convolutional Neural Network |
title_fullStr |
Enhancing Wind Turbine Power Forecast via Convolutional Neural Network |
title_full_unstemmed |
Enhancing Wind Turbine Power Forecast via Convolutional Neural Network |
title_sort |
enhancing wind turbine power forecast via convolutional neural network |
publisher |
MDPI AG |
series |
Electronics |
issn |
2079-9292 |
publishDate |
2021-01-01 |
description |
The rapid development in wind power comes with new technical challenges. Reliable and accurate wind power forecast is of considerable significance to the electricity system’s daily dispatching and production. Traditional forecast methods usually utilize wind speed and turbine parameters as the model inputs. However, they are not sufficient to account for complex weather variability and the various wind turbine features in the real world. Inspired by the excellent performance of convolutional neural networks (CNN) in computer vision, we propose a novel approach to predicting short-term wind power by converting time series into images and exploit a CNN to analyze them. In our approach, we first propose two transformation methods to map wind speed and precipitation data time series into image matrices. After integrating multi-dimensional information and extracting features, we design a novel CNN framework to forecast 24-h wind turbine power. Our method is implemented on the Keras deep learning platform and tested on 10 sets of 3-year wind turbine data from Hangzhou, China. The superior performance of the proposed method is demonstrated through comparisons using state-of-the-art techniques in wind turbine power forecasting. |
topic |
convolutional neural network GWF matrices ripple encoding wind power generation forecast |
url |
https://www.mdpi.com/2079-9292/10/3/261 |
work_keys_str_mv |
AT tianyangliu enhancingwindturbinepowerforecastviaconvolutionalneuralnetwork AT zunkaihuang enhancingwindturbinepowerforecastviaconvolutionalneuralnetwork AT litian enhancingwindturbinepowerforecastviaconvolutionalneuralnetwork AT yongxinzhu enhancingwindturbinepowerforecastviaconvolutionalneuralnetwork AT huiwang enhancingwindturbinepowerforecastviaconvolutionalneuralnetwork AT songlinfeng enhancingwindturbinepowerforecastviaconvolutionalneuralnetwork |
_version_ |
1724327387718483968 |