Adversarial super-resolution of climatological wind and solar data
Abstract
Accurate and high-resolution data reflecting different climate scenarios are vital for policy makers when deciding on the development of future energy resources, electrical infrastructure, transportation networks, agriculture, and many other societally important systems. However, state-of-the-art long-term global climate simulations are unable to resolve the spatiotemporal characteristics necessary for resource assessment or operational planning. We introduce an adversarial deep learning approach to super resolve wind velocity and solar irradiance outputs from global climate models to scales sufficient for renewable energy resource assessment. Using adversarial training to improve the physical and perceptual performance of our networks, we demonstrate up to a50×resolution enhancement of wind and solar data. In validation studies, the inferred fields are robust to input noise, possess the correct small-scale properties of atmospheric turbulent flow and solar irradiance, and retain consistency at large scales with coarse data. An additional advantage of our fully convolutional architecture is that it allows for training on small domains and evaluation on arbitrarily-sized inputs, including global scale. We conclude with a super-resolution study of renewable energy resources based on climate scenario data from the Intergovernmental Panel on Climate Change’s Fifth Assessment Report.
- Authors:
-
- National Renewable Energy Lab. (NREL), Golden, CO (United States)
- Publication Date:
- Research Org.:
- National Renewable Energy Laboratory (NREL), Golden, CO (United States)
- Sponsoring Org.:
- USDOE Laboratory Directed Research and Development (LDRD) Program
- OSTI Identifier:
- 1660090
- Report Number(s):
- NREL/JA-2C00-75625
Journal ID: ISSN 0027-8424; MainId:6859;UUID:c2361be1-9c1b-ea11-9c2a-ac162d87dfe5;MainAdminID:14074
- Grant/Contract Number:
- AC36-08GO28308
- Resource Type:
- Accepted Manuscript
- Journal Name:
- Proceedings of the National Academy of Sciences of the United States of America
- Additional Journal Information:
- Journal Volume: 117; Journal Issue: 29; Journal ID: ISSN 0027-8424
- Publisher:
- National Academy of Sciences
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 29 ENERGY PLANNING, POLICY, AND ECONOMY; 17 WIND ENERGY; 14 SOLAR ENERGY; adversarial training; climate downscaling; deep learning
Citation Formats
Stengel, Karen, Glaws, Andrew, Hettinger, Dylan, and King, Ryan N. Adversarial super-resolution of climatological wind and solar data. United States: N. p., 2020.
Web. doi:10.1073/pnas.1918964117.
Stengel, Karen, Glaws, Andrew, Hettinger, Dylan, & King, Ryan N. Adversarial super-resolution of climatological wind and solar data. United States. https://doi.org/10.1073/pnas.1918964117
Stengel, Karen, Glaws, Andrew, Hettinger, Dylan, and King, Ryan N. Mon .
"Adversarial super-resolution of climatological wind and solar data". United States. https://doi.org/10.1073/pnas.1918964117. https://www.osti.gov/servlets/purl/1660090.
@article{osti_1660090,
title = {Adversarial super-resolution of climatological wind and solar data},
author = {Stengel, Karen and Glaws, Andrew and Hettinger, Dylan and King, Ryan N.},
abstractNote = {Accurate and high-resolution data reflecting different climate scenarios are vital for policy makers when deciding on the development of future energy resources, electrical infrastructure, transportation networks, agriculture, and many other societally important systems. However, state-of-the-art long-term global climate simulations are unable to resolve the spatiotemporal characteristics necessary for resource assessment or operational planning. We introduce an adversarial deep learning approach to super resolve wind velocity and solar irradiance outputs from global climate models to scales sufficient for renewable energy resource assessment. Using adversarial training to improve the physical and perceptual performance of our networks, we demonstrate up to a50×resolution enhancement of wind and solar data. In validation studies, the inferred fields are robust to input noise, possess the correct small-scale properties of atmospheric turbulent flow and solar irradiance, and retain consistency at large scales with coarse data. An additional advantage of our fully convolutional architecture is that it allows for training on small domains and evaluation on arbitrarily-sized inputs, including global scale. We conclude with a super-resolution study of renewable energy resources based on climate scenario data from the Intergovernmental Panel on Climate Change’s Fifth Assessment Report.},
doi = {10.1073/pnas.1918964117},
journal = {Proceedings of the National Academy of Sciences of the United States of America},
number = 29,
volume = 117,
place = {United States},
year = {Mon Jul 06 00:00:00 EDT 2020},
month = {Mon Jul 06 00:00:00 EDT 2020}
}
Web of Science
Works referenced in this record:
Image upsampling via imposed edge statistics
journal, July 2007
- Fattal, Raanan
- ACM Transactions on Graphics, Vol. 26, Issue 3
The impact of climate change on the electricity market: A review
journal, July 2010
- Mideksa, Torben K.; Kallbekken, Steffen
- Energy Policy, Vol. 38, Issue 7
The Perception-Distortion Tradeoff
conference, June 2018
- Blau, Yochai; Michaeli, Tomer
- 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Using Machine Learning to Parameterize Moist Convection: Potential for Modeling of Climate, Climate Change, and Extreme Events
journal, October 2018
- O'Gorman, Paul A.; Dwyer, John G.
- Journal of Advances in Modeling Earth Systems, Vol. 10, Issue 10
Image Superresolution Using Support Vector Regression
journal, June 2007
- Ni, Karl S.; Nguyen, Truong Q.
- IEEE Transactions on Image Processing, Vol. 16, Issue 6
Downscaling and Assimilation of Surface Soil Moisture Using Ground Truth Measurements
journal, May 2008
- Kaheil, Yasir H.; Gill, M. Kashif; McKee, Mac
- IEEE Transactions on Geoscience and Remote Sensing, Vol. 46, Issue 5
Effects on power system operations of potential changes in wind and solar generation potential under climate change
journal, March 2019
- Craig, Michael T.; Losada Carreño, Ignacio; Rossol, Michael
- Environmental Research Letters, Vol. 14, Issue 3
A Neural Network Approach for a Fast and Accurate Computation of a Longwave Radiative Budget
journal, November 1998
- Chevallier, F.; Chéruy, F.; Scott, N. A.
- Journal of Applied Meteorology, Vol. 37, Issue 11
An Overview of CMIP5 and the Experiment Design
journal, April 2012
- Taylor, Karl E.; Stouffer, Ronald J.; Meehl, Gerald A.
- Bulletin of the American Meteorological Society, Vol. 93, Issue 4
The Wind Integration National Dataset (WIND) Toolkit
journal, August 2015
- Draxl, Caroline; Clifton, Andrew; Hodge, Bri-Mathias
- Applied Energy, Vol. 151, p. 355-366
The National Solar Radiation Data Base (NSRDB)
journal, June 2018
- Sengupta, Manajit; Xie, Yu; Lopez, Anthony
- Renewable and Sustainable Energy Reviews, Vol. 89
ArtGAN: Artwork synthesis with conditional categorical GANs
conference, September 2017
- Tan, Wei Ren; Chan, Chee Seng; Aguirre, Hernan E.
- 2017 IEEE International Conference on Image Processing (ICIP)
Southward shift of the global wind energy resource under high carbon dioxide emissions
journal, December 2017
- Karnauskas, Kristopher B.; Lundquist, Julie K.; Zhang, Lei
- Nature Geoscience, Vol. 11, Issue 1
Deeply-Recursive Convolutional Network for Image Super-Resolution
conference, June 2016
- Kim, Jiwon; Lee, Jung Kwon; Lee, Kyoung Mu
- 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Accurate Image Super-Resolution Using Very Deep Convolutional Networks
conference, June 2016
- Kim, Jiwon; Lee, Jung Kwon; Lee, Kyoung Mu
- 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Landsat Super-Resolution Enhancement Using Convolution Neural Networks and Sentinel-2 for Training
journal, March 2018
- Pouliot, Darren; Latifovic, Rasim; Pasher, Jon
- Remote Sensing, Vol. 10, Issue 3
Deep learning to represent subgrid processes in climate models
journal, September 2018
- Rasp, Stephan; Pritchard, Michael S.; Gentine, Pierre
- Proceedings of the National Academy of Sciences, Vol. 115, Issue 39
Do climate models project changes in solar resources?
journal, May 2016
- Huber, Isabelle; Bugliaro, Luca; Ponater, Michael
- Solar Energy, Vol. 129
Learning a Deep Convolutional Network for Image Super-Resolution
book, January 2014
- Dong, Chao; Loy, Chen Change; He, Kaiming
- Computer Vision – ECCV 2014
Image up-sampling using total-variation regularization with a new observation model
journal, October 2005
- Aly, H. A.; Dubois, E.
- IEEE Transactions on Image Processing, Vol. 14, Issue 10
Dynamical downscaling: Assessment of model system dependent retained and added variability for two different regional climate models
journal, January 2008
- Rockel, Burkhardt; Castro, Christopher L.; Pielke, Roger A.
- Journal of Geophysical Research, Vol. 113, Issue D21
Image Super-Resolution Using Deep Convolutional Networks
journal, February 2016
- Dong, Chao; Loy, Chen Change; He, Kaiming
- IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 38, Issue 2
Single image super-resolution using Gaussian process regression
conference, June 2011
- He, He; Siu, Wan-Chi
- 2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), CVPR 2011
tempoGAN: a temporally coherent, volumetric GAN for super-resolution fluid flow
journal, August 2018
- Xie, You; Franz, Erik; Chu, Mengyu
- ACM Transactions on Graphics, Vol. 37, Issue 4
Principles of geostatistics
journal, December 1963
- Matheron, Georges
- Economic Geology, Vol. 58, Issue 8
Deep learning and process understanding for data-driven Earth system science
journal, February 2019
- Reichstein, Markus; Camps-Valls, Gustau; Stevens, Bjorn
- Nature, Vol. 566, Issue 7743
Super-Resolution Simulation for Real-Time Prediction of Urban Micrometeorology
journal, January 2019
- Onishi, Ryo; Sugiyama, Daisuke; Matsuda, Keigo
- SOLA, Vol. 15, Issue 0
Perceptual Losses for Real-Time Style Transfer and Super-Resolution
book, September 2016
- Johnson, Justin; Alahi, Alexandre; Fei-Fei, Li
- Computer Vision – ECCV 2016
Development of neural network convection parameterizations for numerical climate and weather prediction models using cloud resolving model simulations
conference, July 2010
- Krasnopolsky, Vladimir M.; Fox-Rabinovitz, Michael S.; Belochitski, Alexei A.
- The 2010 International Joint Conference on Neural Networks (IJCNN)
Overview and Meteorological Validation of the Wind Integration National Dataset toolkit
report, April 2015
- Draxl, C.; Hodge, B. M.; Clifton, A.
TAG: A time-dependent, autoregressive, Gaussian model for generating synthetic hourly radiation
journal, September 1992
- Aguiar, R.; Collares-Pereira, M.
- Solar Energy, Vol. 49, Issue 3
Prognostic Validation of a Neural Network Unified Physics Parameterization
journal, June 2018
- Brenowitz, N. D.; Bretherton, C. S.
- Geophysical Research Letters, Vol. 45, Issue 12
DeepDownscale: A Deep Learning Strategy for High-Resolution Weather Forecast
conference, October 2018
- Rocha Rodrigues, Eduardo; Oliveira, Igor; Cunha, Renato
- 2018 IEEE 14th International Conference on e-Science (e-Science)
The Phenomenology of Small-Scale Turbulence
journal, January 1997
- Sreenivasan, K. R.; Antonia, R. A.
- Annual Review of Fluid Mechanics, Vol. 29, Issue 1
DeepSD: Generating High Resolution Climate Change Projections through Single Image Super-Resolution
conference, January 2017
- Vandal, Thomas; Kodra, Evan; Ganguly, Sangram
- Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '17
Deep Residual Learning for Image Recognition
conference, June 2016
- He, Kaiming; Zhang, Xiangyu; Ren, Shaoqing
- 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Perceptual Adversarial Networks for Image-to-Image Transformation
journal, August 2018
- Wang, Chaoyue; Xu, Chang; Wang, Chaohui
- IEEE Transactions on Image Processing, Vol. 27, Issue 8
A review of the potential impacts of climate change on bulk power system planning and operations in the United States
journal, December 2018
- Craig, Michael T.; Cohen, Stuart; Macknick, Jordan
- Renewable and Sustainable Energy Reviews, Vol. 98
Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network
conference, July 2017
- Ledig, Christian; Theis, Lucas; Huszar, Ferenc
- 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Super-resolution reconstruction of turbulent flows with machine learning
journal, May 2019
- Fukami, Kai; Fukagata, Koji; Taira, Kunihiko
- Journal of Fluid Mechanics, Vol. 870
Tree approximation of the long wave radiation parameterization in the NCAR CAM global climate model
journal, September 2011
- Belochitski, Alexei; Binev, Peter; DeVore, Ronald
- Journal of Computational and Applied Mathematics, Vol. 236, Issue 4
Deep learning for ocean remote sensing: an application of convolutional neural networks for super-resolution on satellite-derived SST data
conference, December 2016
- Ducournau, Aurelien; Fablet, Ronan
- 2016 9th IAPR Workshop on Pattern Recogniton in Remote Sensing (PRRS)
From GCMs to river flow: a review of downscaling methods and hydrologic modelling approaches
journal, June 1999
- Xu, Chong-yu
- Progress in Physical Geography: Earth and Environment, Vol. 23, Issue 2
Validation of Power Output for the WIND Toolkit
report, September 2014
- King, J.; Clifton, A.; Hodge, B.
Works referencing / citing this record:
Will Artificial Intelligence supersede Earth System and Climate Models?
text, January 2021
- Irrgang, Christopher; Boers, Niklas; Sonnewald, Maike
- arXiv
CIRA Guide to Custom Loss Functions for Neural Networks in Environmental Sciences -- Version 1
preprint, January 2021
- Ebert-Uphoff, Imme; Lagerquist, Ryan; Hilburn, Kyle
- arXiv