ABSTRACT
Electricity usage is a substantial source of carbon emissions worldwide. There has been significant interest in reducing the carbon impact of energy usage through supply-side shifts to cleaner generation sources and through demand-side optimizations to reduce carbon usage. An essential building block for these optimizations is future knowledge of the carbon intensity of the supplied electricity. In this paper, we present a Day-Ahead Carbon Forecasting system (DACF) that predicts the carbon intensity from scope 2 emissions in the power grids using machine learning. DACF first computes production forecasts for all the electricity-generating sources and then combines them with the carbon-emission rate of each source to generate a carbon intensity forecast. DACF provides a general approach that works well across a range of geographically distributed regions. DACF has a mean MAPE of 6.4% across the regions. It also achieves an average decrease of 6.4% and a maximum decrease of 8.6% in MAPE compared to the state-of-the-art. We make DACF publicly available so that it is easily accessible to researchers.
- Martín Abadi, Ashish Agarwal, Paul Barham, Eugene Brevdo, Zhifeng Chen, Craig Citro, Greg S. Corrado, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Ian Goodfellow, Andrew Harp, Geoffrey Irving, Michael Isard, Yangqing Jia, Rafal Jozefowicz, Lukasz Kaiser, Manjunath Kudlur, Josh Levenberg, Dan Mané, Rajat Monga, Sherry Moore, Derek Murray, Chris Olah, Mike Schuster, Jonathon Shlens, Benoit Steiner, Ilya Sutskever, Kunal Talwar, Paul Tucker, Vincent Vanhoucke, Vijay Vasudevan, Fernanda Viégas, Oriol Vinyals, Pete Warden, Martin Wattenberg, Martin Wicke, Yuan Yu, and Xiaoqiang Zheng. 2015. TensorFlow: Large-Scale Machine Learning on Heterogeneous Systems. http://tensorflow.org/ Software available from tensorflow.org.Google Scholar
- US Energy Information Administration. 2018. Real-time Operating Grid. Retrieved February 6, 2022 from https://www.eia.gov/electricity/gridmonitor/dashboard/electric_overview/US48/US48Google Scholar
- Neeraj Dhanraj Bokde, Bo Tranberg, and Gorm Bruun Andresen. 2021. Short-term CO2 emissions forecasting based on decomposition approaches and its impact on electricity market scheduling. Applied Energy 281 (2021), 116061.Google ScholarCross Ref
- A Bruce, Lyndon Ruff, James Kelloway, Fraser MacMillan, and Alex Rogers. 2021. Carbon intensity forecast methodology. National Grid ESO: Warwick, UK. Available online: https://github.com/carbon-intensity/methodology/raw/master/Carbon%20Intensity%20Forecast 20 (2021).Google Scholar
- California ISO. 2005--2022. Open Access Same-time Information System (OASIS). Retrieved February 8, 2022 from http://oasis.caiso.com/mrioasis/logon.doGoogle Scholar
- François Chollet et al. 2015. Keras. https://keras.io.Google Scholar
- Climate Change. 2014. Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. [Core Writing Team, R.K. Pachauri and L.A. Meyer (eds.)]. IPCC, Geneva, Switzerland, 151 pp. Retrieved February 8, 2022 from https://archive.ipcc.ch/pdf/assessment-report/ar5/wg3/ipcc_wg3_ar5_annex-iii.pdf#page=7Google Scholar
- Department of Business, Energy and Industrial Startegy. 2021. 2021 Government Greenhouse Gas Conversion Factors for Company Reporting. Retrieved April 26, 2022 from https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1049346/2021-ghg-conversion-factors-methodology.pdfGoogle Scholar
- ElectricityMap. 2020. Zone bounding boxes. Retrieved February 8, 2022 from https://github.com/electricityMap/electricitymap-contrib/blob/master/config/zones.jsonGoogle Scholar
- ElectricityMap. 2022. ElectricityMap. Retrieved February 8, 2022 from https://electricitymap.org/Google Scholar
- European association for the cooperation of transmission system operators. 2008. ENTSOE transparency platform. Retrieved February 8, 2022 from https://transparency.entsoe.eu/Google Scholar
- Julian Huber, Kai Lohmann, Marc Schmidt, and Christof Weinhardt. 2021. Carbon efficient smart charging using forecasts of marginal emission factors. Journal of Cleaner Production 284 (2021), 124766.Google ScholarCross Ref
- International Energy Agency. 2019. Global Energy & CO2 Status Report 2019: Emissions. Retrieved February 8, 2022 from https://www.epa.gov/ghgemissions/global-greenhouse-gas-emissions-dataGoogle Scholar
- Kenneth Leerbeck, Peder Bacher, Rune Grønborg Junker, Goran Goranović, Olivier Corradi, Razgar Ebrahimy, Anna Tveit, and Henrik Madsen. 2020. Short-term forecasting of CO2 emission intensity in power grids by machine learning. Applied Energy 277 (2020), 115527.Google ScholarCross Ref
- Gordon Lowry. 2018. Day-ahead forecasting of grid carbon intensity in support of heating, ventilation and air-conditioning plant demand response decision-making to reduce carbon emissions. Building Services Engineering Research and Technology 39, 6 (2018), 749--760.Google ScholarCross Ref
- Luke George. 2021. The Correct Way to Average the Globe. Retrieved February 8, 2022 from https://towardsdatascience.com/the-correct-way-to-average-the-globe-92ceecd172b7Google Scholar
- National Centers for Environmental Prediction, National Weather Service, NOAA, U.S. Department of Commerce. 2015. NCEP GFS 0.25 Degree Global Forecast Grids Historical Archive. Retrieved February 8, 2022 from Google ScholarCross Ref
- United Satets Environmental Protection Agency. 2021. Greenhouse Gas Protocol: Scope 2 Guidance. Retrieved April 26, 2022 from https://ghgprotocol.org/sites/default/files/standards/Scope%202%20Guidance_Final_Sept26.pdfGoogle Scholar
- US Environmental Protection Agency. 2021. Global Greenhouse Gas Emissions Data. Retrieved February 8, 2022 from https://www.epa.gov/ghgemissions/global-greenhouse-gas-emissions-dataGoogle Scholar
- US Environmental Protection Agency. 2021. Sources of Greenhouse Gas Emissions. Retrieved February 8, 2022 from https://www.epa.gov/ghgemissions/sources-greenhouse-gas-emissions#:~:text=Larger%20image%20to%20save%20or%20print%20The%20Electricity%20sector%20involves,2O)%20are%20also%20emitted.Google Scholar
- Watttime. 2022. Watttime. Retrieved February 8, 2022 from https://www.watttime.org/Google Scholar
Index Terms
- DACF: day-ahead carbon intensity forecasting of power grids using machine learning
Recommendations
CarbonCast: multi-day forecasting of grid carbon intensity
BuildSys '22: Proceedings of the 9th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and TransportationThe ever-increasing demand for energy is resulting in considerable carbon emissions from the electricity grid. In recent years, there has been growing attention on demand-side optimizations to reduce carbon emissions from electricity usage. A vital ...
Multi-Day Forecasting of Electric Grid Carbon Intensity Using Machine Learning
The ever-increasing demand for energy is resulting in considerable carbon emissions from the electricity grid. In recent years, there has been growing attention on demand-side optimizations to reduce carbon emissions from electricity usage. A vital ...
A Cloud-Based Black-Box Solar Predictor for Smart Homes
Special Issue on Smart Homes, Buildings and InfrastructuresThe popularity of rooftop solar for homes is rapidly growing. However, accurately forecasting solar generation is critical to fully exploiting the benefits of locally generated solar energy. In this article, we present two machine-learning techniques to ...
Comments