An effective way to mitigate climate change is to electrify most of our energy demand and supply the necessary electricity from renewable wind and solar power plants. Spatio-temporal predictions of electric load become increasingly important for planning this transition, while deep learning prediction models provide increasingly accurate predictions for it. The data that are used for training deep learning models, however, are usually collected at random using a passive learning approach. This naturally results in a large demand for data and associated costs for sensors such as smart meters, posing a large barrier for electric utilities when decarbonizing their grids. Here we investigate whether electric utilities can use active learning to collect a more informative subset of data by leveraging additional computation for better distributing smart meters. We predict ground-truth electric load profiles for single buildings using only remotely sensed data from aerial imagery of these buildings and meteorological conditions in the area of these buildings at different times. We find that active learning can enable 26–81% more accurate predictions using 29–46% less data at the price of 4–11 times more computation compared with passive learning.
This is a preview of subscription content, access via your institution
Subscribe to Nature+
Get immediate online access to Nature and 55 other Nature journal
Subscribe to Journal
Get full journal access for 1 year
only $8.25 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Get time limited or full article access on ReadCube.
All prices are NET prices.
All of our data and results can be accessed on the Harvard Dataverse under a CC0 1.0 license (https://doi.org/10.7910/DVN/3VYYET). For privacy maintenance, we only provide load profiles that are sampled from the original data and histograms of pixel values of building images, which can be used to reproduce all the elements of our original experiments.
All results, figures and tables can be reproduced using step-by-step instructions in Jupyter notebook sessions that we provide in a public Github repository (https://github.com/ArsamAryandoust/DataSelectionMaps). We further maintain a Python package (https://pypi.org/project/altility) and a Docker container (https://hub.docker.com/r/aryandoustarsam/altility) implementation of our algorithm. All code is available under an MIT license.
Patt, A. Transforming Energy – Solving Climate Change with Technology Policy (Cambridge Univ. Press, 2015).
IPCC Special Report on Global Warming of 1.5 °C (eds Masson-Delmotte, V. et al.) (WMO, 2018).
IPCC Climate Change 2021: The Physical Science Basis (eds Masson-Delmotte, V. et al.) (Cambridge Univ. Press, 2021).
Hahn, H., Meyer-Nieberg, S. & Pickl, S. Electric load forecasting methods: tools for decision making. Eur. J. Oper. Res. https://doi.org/10.1016/j.ejor.2009.01.062 (2009).
Soliman, A.-h. S. & Al-Kandari, A. M. Electric Load Forecasting (Butterworth-Heinemann, 2010).
Alfares, H. K. & Nazeeruddin, M. Electric load forecasting: literature survey and classification of methods. Int. J. Syst. Sci. 33, 23–24 (2002).
Kofi Nti, I., Teimeh, M., Nyarko-Boateng, O. & Adekoya, A. F. Electricity load forecasting: a systematic review. J. Electr. Syst. Inform. Technol. 7, 13 (2020).
Shi, J., Liu, Y. & Yu, N. Spatio-temporal modeling of electric loads. IEEE https://doi.org/10.1109/NAPS.2017.8107311 (2017).
Tascikaraoglu, A. Evaluation of spatio-temporal forecasting methods in various smart city applications. Renew. Sustain. Energy Rev. 82, 424–435 (2018).
Severiano, C. A., Cândido de Lima eSilva, P., Cohen, M. W. & Gadelha Guimarãesae, F. Evolving fuzzy time series for spatio-temporal forecasting in renewable energy systems. Renew. Energy 171, 764–783 (2021).
Willis, H. L. Spatial Electric Load Forecasting (Marcel Dekker, 2002).
Rolf, E. et al. A generalizable and accessible approach to machine learning with global satellite imagery. Nat. Commun. 12, 4392 (2021).
Burke, M., Driscoll, A.Lobell, D. B. & Ermon, S. Using satellite imagery to understand and promote sustainable development. Science https://doi.org/10.1126/science.abe8628 (2021).
Melo, J. D. & Carreno, E. M. Data Issues in Spatial Electric Load Forecasting (IEEE, 2014).
Milam, M. & Venayagamoorthy, G. K. Smart meter deployment: US initiatives. IEEE https://doi.org/10.1109/ISGT.2014.6816507 (2014).
Sovacool, B. K., Hook, A., Sareen, S. & Geels, F. W. Global sustainability, innovation and governance dynamics of national smart electricity meter transitions. Glob. Environ. Change https://doi.org/10.1016/j.gloenvcha.2021.102272 (2021).
Kezunovic, M., Xie, L. & Grijalva, S. The role of big data in improving power system operation and protection. IEEE https://doi.org/10.1109/IREP.2013.6629368 (2013).
Yu, N. et al. The role of big data in improving power system operation and protection. IEEE https://doi.org/10.1109/ISGT.2015.7131868 (2015).
Stein, A. L. Artificial intelligence and climate change. Yale J. Regul. 37, 890–934 (2020).
Rolnick, D. et al. Tackling climate change with machine learning. ACM Comput. Surv. https://doi.org/10.1145/3485128 (2022).
Kuo, P., Liang, D., Gao, L. & Lou, J. Probabilistic electricity price forecasting with variational heteroscedastic Gaussian process and active learning. Energy Convers. Manage. 89, 298–308 (2015).
Wang, Z., Zhao, B., Guo, H., Tang, L. & Peng, Y. Deep ensemble learning model for short-term load forecasting within active learning framework. Energies https://doi.org/10.3390/en12203809 (2019).
Zhang, L. & Wen, J. Active learning strategy for high fidelity short-term data-driven building energy forecasting. Energy Build. https://doi.org/10.1016/j.enbuild.2021.111026 (2021).
Kuster, C., Rezgui, Y. & Mourshed, M. Electrical load forecasting models: a critical systematic review. Sustain. Cities Soc. 35, 257–270 (2017).
Panamtash, H., Mahdavi, S., Dimitrovski, A. & Zhou, Q. Comparison of Probabilistic Forecasts for Predictive Voltage Control (North American Power Symposium, 2021).
Chen, T. et al. Big self-supervised models are strong semi-supervised learners. In Proc. of the 34th Conference on Advances in Neural Information Processing Systems 33 (eds Larochelle, H. et al.) 22243–22255 (NeurIPS, 2020).
Yue, X. et al. Prototypical cross-domain self-supervised learning for few-shot unsupervised domain adaptation. In Proc. of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 13834–13844 (CVF, 2021).
Yang, W. et al. Class distribution alignment for adversarial domain adaptation. Preprint at https://arxiv.org/abs/2004.09403 (2020).
Saito, K., Saenko, K. & Liu, M.-Y. COCO-FUNIT: few-shot unsupervised image translation with a content conditioned style encoder. In Proc. of 16th European Conference on Computer Vision (eds Vedaldi, A. et al.) 382–398 (Springer, 2020).
Jain, A., Zamir, A. R., Savarese, S. & Saxena, A. Structural-RNN: deep learning on spatio-temporal graphs. In 2016 IEEE Conference on Computer Vision and Pattern Recognition (IEEE, 2016).
Mikolov, T., Chen, K., Corrado, G. & Dean, J. Efficient estimation of word representations in vector space. Preprint at https://arxiv.org/abs/1301.3781 (2013).
Frome, A. et al. DeViSE: a deep visual-semantic embedding model. In Proc. of the 26th Advances in Neural Information Processing Systems 26 (eds Burges, C.J. et al.) 2121–2129 (NIPS, 2013).
Pennington, J., Socher, R. & Manning, C. D. GloVe: Global vectors for word representation. In Proc. of the 2014 Conference on Empirical Methods in Natural Language Processing (eds Moschitti, A. et al.) 1532–1543 (ACL, 2014).
Perozzi, B., Al-Rfou, R. & Skiena, S. DeepWalk: online learning of social represenations. Preprint at https://arxiv.org/abs/1403.6652 (2014).
Devlin, J., Chang, M.-W., Lee, K. & Toutanova, K. BERT: pre-training of deep bidirectional transformers for language understanding. Preprint at https://arxiv.org/abs/1810.04805 (2019).
GeoVITE – user-friendly geodata service. Swiss Federal Office of Topography https://geovite.ethz.ch/ (2020).
Pfenninger, S. & Staffel, I. Long-term patterns of European PV output using 30 years of validated hourly reanalysis and satellite data. Energy 114, 1251–1265 (2016).
Staffel, I. & Pfenninger, S. Using bias-corected reanalysis to simulate current and future wind power output. Energy 114, 1224–1239 (2016).
Ash, J. T., Zhang, C., Krishnamurthy, A., Langford, J. & Agarwal, A. Deep batch active learning by diverse, uncertain gradient lower bounds. In Proc. of International Conference on Learning Representations https://openreview.net/forum?id=ryghZJBKPS (2020).
We acknowledge funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement no. 837089 (A.A., A.P. and S.P.), for the SENTINEL project.
The authors declare no competing interests.
Peer review information
Nature Machine Intelligence thanks Valentin Robu and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Aryandoust, A., Patt, A. & Pfenninger, S. Enhanced spatio-temporal electric load forecasts using less data with active deep learning. Nat Mach Intell 4, 977–991 (2022). https://doi.org/10.1038/s42256-022-00552-x