skip to main content
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Wind Speed Prediction with Spatio–Temporal Correlation: A Deep Learning Approach

Abstract

Wind speed prediction with spatio–temporal correlation is among the most challenging tasks in wind speed prediction. In this paper, the problem of predicting wind speed for multiple sites simultaneously is investigated by using spatio–temporal correlation. This paper proposes a model for wind speed prediction with spatio–temporal correlation, i.e., the predictive deep convolutional neural network (PDCNN). The model is a unified framework, integrating convolutional neural networks (CNNs) and a multi-layer perceptron (MLP). Firstly, the spatial features are extracted by CNNs located at the bottom of the model. Then, the temporal dependencies among these extracted spatial features are captured by the MLP. In this way, the spatial and temporal correlations are captured by PDCNN intrinsically. Finally, PDCNN generates the predicted wind speed by using the learnt spatio–temporal correlations. In addition, three error indices are defined to evaluate the prediction accuracy of the model on the wind turbine array. Experiment results on real-world data show that PDCNN can capture the spatio–temporal correlation effectively, and it outperforms the conventional machine learning models, including multi-layer perceptron, support vector regressor, decision tree, etc.

Authors:
ORCiD logo [1];  [1];  [2];  [1];  [2]
  1. Huazhong Univ. of Science and Technology, Wuhan (China)
  2. Univ. of Tennessee, Knoxville, TN (United States)
Publication Date:
Research Org.:
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1474549
Grant/Contract Number:  
AC05-00OR22725
Resource Type:
Accepted Manuscript
Journal Name:
Energies (Basel)
Additional Journal Information:
Journal Name: Energies (Basel); Journal Volume: 11; Journal Issue: 4; Journal ID: ISSN 1996-1073
Publisher:
MDPI AG
Country of Publication:
United States
Language:
English
Subject:
17 WIND ENERGY

Citation Formats

Zhu, Qiaomu, Chen, Jinfu, Zhu, Lin, Duan, Xianzhong, and Liu, Yilu. Wind Speed Prediction with Spatio–Temporal Correlation: A Deep Learning Approach. United States: N. p., 2018. Web. doi:10.3390/en11040705.
Zhu, Qiaomu, Chen, Jinfu, Zhu, Lin, Duan, Xianzhong, & Liu, Yilu. Wind Speed Prediction with Spatio–Temporal Correlation: A Deep Learning Approach. United States. doi:10.3390/en11040705.
Zhu, Qiaomu, Chen, Jinfu, Zhu, Lin, Duan, Xianzhong, and Liu, Yilu. Wed . "Wind Speed Prediction with Spatio–Temporal Correlation: A Deep Learning Approach". United States. doi:10.3390/en11040705. https://www.osti.gov/servlets/purl/1474549.
@article{osti_1474549,
title = {Wind Speed Prediction with Spatio–Temporal Correlation: A Deep Learning Approach},
author = {Zhu, Qiaomu and Chen, Jinfu and Zhu, Lin and Duan, Xianzhong and Liu, Yilu},
abstractNote = {Wind speed prediction with spatio–temporal correlation is among the most challenging tasks in wind speed prediction. In this paper, the problem of predicting wind speed for multiple sites simultaneously is investigated by using spatio–temporal correlation. This paper proposes a model for wind speed prediction with spatio–temporal correlation, i.e., the predictive deep convolutional neural network (PDCNN). The model is a unified framework, integrating convolutional neural networks (CNNs) and a multi-layer perceptron (MLP). Firstly, the spatial features are extracted by CNNs located at the bottom of the model. Then, the temporal dependencies among these extracted spatial features are captured by the MLP. In this way, the spatial and temporal correlations are captured by PDCNN intrinsically. Finally, PDCNN generates the predicted wind speed by using the learnt spatio–temporal correlations. In addition, three error indices are defined to evaluate the prediction accuracy of the model on the wind turbine array. Experiment results on real-world data show that PDCNN can capture the spatio–temporal correlation effectively, and it outperforms the conventional machine learning models, including multi-layer perceptron, support vector regressor, decision tree, etc.},
doi = {10.3390/en11040705},
journal = {Energies (Basel)},
number = 4,
volume = 11,
place = {United States},
year = {2018},
month = {3}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Citation Metrics:
Cited by: 2 works
Citation information provided by
Web of Science

Save / Share:

Works referenced in this record:

Short-term prediction of wind power with a clustering approach
journal, October 2010


Method selection in different regions for short-term wind speed prediction in Japan
conference, July 2015

  • Tanaka, Ikki; Ohmori, Hiromitsu
  • 2015 54th Annual Conference of the Society of Instrument and Control Engineers of Japan (SICE)
  • DOI: 10.1109/SICE.2015.7285424

Deep neural networks for ultra-short-term wind forecasting
conference, March 2015

  • Dalto, Mladen; Matusko, Jadranko; Vasak, Mario
  • 2015 IEEE International Conference on Industrial Technology (ICIT)
  • DOI: 10.1109/ICIT.2015.7125335

An Efficient Approach to Power System Uncertainty Analysis With High-Dimensional Dependencies
journal, May 2018


Predictive Deep Boltzmann Machine for Multiperiod Wind Speed Forecasting
journal, October 2015

  • Zhang, Chun-Yang; Chen, C. L. Philip; Gan, Min
  • IEEE Transactions on Sustainable Energy, Vol. 6, Issue 4
  • DOI: 10.1109/TSTE.2015.2434387

Deep Learning for Consumer Devices and Services: Pushing the limits for machine learning, artificial intelligence, and computer vision.
journal, April 2017

  • Lemley, Joe; Bazrafkan, Shabab; Corcoran, Peter
  • IEEE Consumer Electronics Magazine, Vol. 6, Issue 2
  • DOI: 10.1109/MCE.2016.2640698

Deep belief network based deterministic and probabilistic wind speed forecasting approach
journal, November 2016


Statistical modelling of directional wind speeds using mixtures of von Mises distributions: Case study
journal, May 2008


An End-to-End Trainable Neural Network for Image-Based Sequence Recognition and Its Application to Scene Text Recognition
journal, November 2017

  • Shi, Baoguang; Bai, Xiang; Yao, Cong
  • IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 39, Issue 11
  • DOI: 10.1109/TPAMI.2016.2646371

A Comparative Study of Empirical Mode Decomposition-Based Short-Term Wind Speed Forecasting Methods
journal, January 2015

  • Ren, Ye; Suganthan, P. N.; Srikanth, Narasimalu
  • IEEE Transactions on Sustainable Energy, Vol. 6, Issue 1
  • DOI: 10.1109/TSTE.2014.2365580

A Fuzzy Model for Wind Speed Prediction and Power Generation in Wind Parks Using Spatial Correlation
journal, June 2004

  • Damousis, I. G.; Alexiadis, M. C.; Theocharis, J. B.
  • IEEE Transactions on Energy Conversion, Vol. 19, Issue 2
  • DOI: 10.1109/TEC.2003.821865

Spatial prediction of wind farm outputs using the Augmented Kriging-based Model
conference, July 2012

  • Jin Hur, ; Baldick, R.
  • 2012 IEEE Power & Energy Society General Meeting. New Energy Horizons - Opportunities and Challenges, 2012 IEEE Power and Energy Society General Meeting
  • DOI: 10.1109/PESGM.2012.6345117

Schedulability Analysis of Conditional Parallel Task Graphs in Multicore Systems
journal, January 2016

  • Melani, Alessandra; Bertogna, Marko; Bonifaci, Vincenzo
  • IEEE Transactions on Computers
  • DOI: 10.1109/TC.2016.2584064

A review of wind power and wind speed forecasting methods with different time horizons
conference, September 2010

  • Soman, Saurabh S.; Zareipour, Hamidreza; Malik, Om
  • 2010 North American Power Symposium (NAPS 2010), North American Power Symposium 2010
  • DOI: 10.1109/NAPS.2010.5619586

The influence of wind direction on short-term wind power prediction: A case study in north China
conference, May 2012

  • Yuyu Ding,
  • 2012 IEEE Innovative Smart Grid Technologies - Asia (ISGT Asia), IEEE PES Innovative Smart Grid Technologies
  • DOI: 10.1109/ISGT-Asia.2012.6303282

Wind speed spatial estimation for energy planning in Sicily: Introduction and statistical analysis
journal, June 2008


The Wind Integration National Dataset (WIND) Toolkit
journal, August 2015


Learning Deep Architectures for AI
journal, January 2009

  • Bengio, Y.
  • Foundations and Trends® in Machine Learning, Vol. 2, Issue 1
  • DOI: 10.1561/2200000006

Analytical Assessment for Transient Stability Under Stochastic Continuous Disturbances
journal, March 2018


Deep Convolutional Neural Networks and Data Augmentation for Environmental Sound Classification
journal, March 2017


Correlation of Wind Speed Between Neighboring Measuring Stations
journal, June 2004

  • Bechrakis, D. A.; Sparis, P. D.
  • IEEE Transactions on Energy Conversion, Vol. 19, Issue 2
  • DOI: 10.1109/TEC.2004.827040

A method for coherency identification based on singular value decomposition
conference, July 2016


Transfer learning for short-term wind speed prediction with deep neural networks
journal, January 2016


Input variable selection: mutual information and linear mixing measures
journal, January 2006

  • Trappenberg, T.; Ouyang, J.; Back, A.
  • IEEE Transactions on Knowledge and Data Engineering, Vol. 18, Issue 1
  • DOI: 10.1109/TKDE.2006.11

Rough Deep Neural Architecture for Short-Term Wind Speed Forecasting
journal, December 2017

  • Khodayar, Mahdi; Kaynak, Okyay; Khodayar, Mohammad E.
  • IEEE Transactions on Industrial Informatics, Vol. 13, Issue 6
  • DOI: 10.1109/TII.2017.2730846

Two Machine Learning Approaches for Short-Term Wind Speed Time-Series Prediction
journal, August 2016

  • Ak, Ronay; Fink, Olga; Zio, Enrico
  • IEEE Transactions on Neural Networks and Learning Systems, Vol. 27, Issue 8
  • DOI: 10.1109/TNNLS.2015.2418739

Multi-Layer Interaction Graph for Analysis and Mitigation of Cascading Outages
journal, June 2017

  • Ju, Wenyun; Sun, Kai; Qi, Junjian
  • IEEE Journal on Emerging and Selected Topics in Circuits and Systems, Vol. 7, Issue 2
  • DOI: 10.1109/JETCAS.2017.2703948

Locally recurrent neural networks optimal filtering algorithms: application to wind speed prediction using spatial correlation
conference, January 2005

  • Barbounis, T. G.; Theocharis, J. B.
  • Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005.
  • DOI: 10.1109/IJCNN.2005.1556353

Optimizing the efficiency of deep learning through accelerator virtualization
journal, July 2017

  • Gschwind, M.; Kaldewey, T.; Tam, D. K.
  • IBM Journal of Research and Development, Vol. 61, Issue 4/5
  • DOI: 10.1147/JRD.2017.2716598

A Review of Wind Power Forecasting Models
journal, January 2011


Wind Power Forecasts Using Gaussian Processes and Numerical Weather Prediction
journal, March 2014


Short-term prediction of the aggregated power output of wind farms—a statistical analysis of the reduction of the prediction error by spatial smoothing effects
journal, March 2002

  • Focken, Ulrich; Lange, Matthias; Mönnich, Kai
  • Journal of Wind Engineering and Industrial Aerodynamics, Vol. 90, Issue 3
  • DOI: 10.1016/S0167-6105(01)00222-7

Impact of Wind Forecast Error Statistics Upon Unit Commitment
journal, October 2012


Wind power prediction using deep neural network based meta regression and transfer learning
journal, September 2017


Injurious or Noninjurious Defect Identification From MFL Images in Pipeline Inspection Using Convolutional Neural Network
journal, July 2017

  • Feng, Jian; Li, Fangming; Lu, Senxiang
  • IEEE Transactions on Instrumentation and Measurement, Vol. 66, Issue 7
  • DOI: 10.1109/TIM.2017.2673024

Pattern-Based Wind Speed Prediction Based on Generalized Principal Component Analysis
journal, July 2014


    Works referencing / citing this record:

    A Review of the Use of Artificial Neural Network Models for Energy and Reliability Prediction. A Study of the Solar PV, Hydraulic and Wind Energy Sources
    journal, May 2019

    • Ferrero Bermejo, Jesús; Gómez Fernández, Juan F.; Olivencia Polo, Fernando
    • Applied Sciences, Vol. 9, Issue 9
    • DOI: 10.3390/app9091844