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 a 50× 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.

NREL prints on paper that contains recycled content.

No abstract

NREL prints on paper that contains recycled content.

This report is one in a series of Electrification Futures Study (EFS) publications. The EFS is a multi-year research project to explore widespread electrification in the future energy system of the United States. This report documents a new model, the demand-side grid (dsgrid) model, which was developed for the EFS and in recognition of a general need for a more detailed understanding of electricity load. dsgrid utilizes a suite of bottom-up engineering models across all major economic sectors-transportation, residential and commercial buildings, and industry-to develop hourly electricity consumption profiles for every county in the contiguous United States (CONUS). The consumption profiles are available by subsector and end use as well as in aggregate. This report documents a bottom-up modeling assessment of historical ( 2012) consumption and explains the key inputs, methodology, assumptions, and limitations of dsgrid.The EFS is specifically designed to examine electric technology cost advancement and adoption for end uses across all major economic sectors as well as electricity consumption growth and load profiles, future power system infrastructure development and operations, and the economic and environmental implications of electrification. Because of the expansive scope and the multiyear duration of the study, research findings and supporting data will be published as a series of reports, with each report released on its own timeframe. Future research to be presented in future planned EFS publications will rely on dsgrid to analyze the hourly electricity consumption under scenarios with various levels of electrification. In addition to providing electricity consumption data for the planned EFS analysis, dsgrid can be used for other analysis outside the EFS research umbrella.More information and the supporting data associated with this report, links to other reports in the EFS study, and information about the broader study are available at www.nrel.gov/efs.