skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


Title: See the Light: Modeling Solar Performance using Multispectral Satellite Data
Developing accurate solar performance models, which infer solar output from widely available external data sources, is increasingly important as the grid's solar capacity rises. These models are important for a wide range of solar analytics, including solar forecasting, resource estimation, and fault detection. The most significant error in existing models is inaccurate estimates of clouds' effect on solar output, as cloud formations and their impact on solar radiation are highly complex. In 2018 and 2019, respectively, the National Oceanic and Atmospheric Administration (NOAA) in the U.S. began releasing multispectral data comprising 16 different light wavelengths (or channels) from the GOES-16 and GOES-17 satellites every 5 minutes. Enough channel data is now available to learn solar performance models using machine learning (ML). In this paper, we show how to develop both local and global solar performance models using ML on multispectral data, and compare their accuracy to existing physical models based on ground-level weather readings and on NOAA's estimates of downward shortwave radiation (DSR), which also derive from multispectral data but using a physical model. We show that ML-based solar performance models based on multispectral data are much more accurate than weather- or DSR-based models, improving the average MAPE across 29 solar sites by over 50% for local models and 25% for global models.  more » « less
Award ID(s):
1645952
PAR ID:
10298239
Author(s) / Creator(s):
;
Date Published:
Journal Name:
Proceedings of the 7th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation (BuildSys)
Page Range / eLocation ID:
1 to 10
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; (, 2020 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm))
    null (Ed.)
    Developing accurate solar performance models, which infer solar power output in real time based on the current environmental conditions, are an important prerequisite for many advanced energy analytics. Recent work has developed sophisticated data-driven techniques that generate customized models for complex rooftop solar sites by combining well-known physical models with both system and public weather station data. However, inferring solar generation from public weather station data has two drawbacks: not all solar sites are near a public weather station, and public weather data generally quantifies cloud cover-the most significant weather metric that affects solar-using highly coarse and imprecise measurements.In this paper, we develop and evaluate solar performance models that use satellite-based estimates of downward shortwave (solar) radiation (DSR) at the Earth's surface, which NOAA began publicly releasing after the launch of the GOES-R geostationary satellites in 2017. Unlike public weather data, DSR estimates are available for every 0.5km 2 area. As we show, the accuracy of solar performance modeling using satellite data and public weather station data depends on the cloud conditions, with DSR-based modeling being more accurate under clear skies and station-based modeling being more accurate under overcast skies. Surprisingly, our results show that, overall, pure satellite-based modeling yields similar accuracy as pure station-based modeling, although the relationship is a function of conditions and the local climate. We also show that a hybrid approach that combines the best of both approaches can also modestly improve accuracy. 
    more » « less
  2. ; ; ; (, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing)
    This study investigates high-frequency mapping of downward shortwave radiation (DSR) at the Earth’s surface using the advanced baseline imager (ABI) instrument mounted on Geo- stationary Operational Environmental Satellite—R Series (GOES- R). The existing GOES-R DSR product (DSRABI) offers hourly temporal resolution and spatial resolution of 0.25°. To enhance these resolutions, we explore machine learning (ML) for DSR estimation at the native temporal resolution of GOES-R Level-2 cloud and moisture imagery product (5 min) and its native spatial resolution of 2 km at nadir. We compared four common ML regres- sion models through the leave-one-out cross-validation algorithm for robust model assessment against ground measurements from AmeriFlux and SURFRAD networks. Results show that gradient boosting regression (GBR) achieves the best performance (R2 = 0.916, RMSE = 88.05 W·m−2) with more efficient computation compared to long short-term memory, which exhibited similar performance. DSR estimates from the GBR model through the ABI live imaging of vegetated ecosystems workflow (DSRALIVE) outperform DSRABI across various temporal resolutions and sky conditions. DSRALIVE agreement with ground measurements at SURFRAD networks exhibits high accuracy at high temporal res- olutions (5-min intervals) with R2 exceeding 0.85 and RMSE = 122 W·m−2 . We conclude that GBR offers a promising approach for high-frequency DSR mapping from GOES-R, enabling improved applications for near-real-time monitoring of terrestrial carbon and water fluxes. 
    more » « less
  3. ; ; ; ; ; ; ; (, Advances in Neural Information Processing Systems)
    Solar power is a critical source of renewable energy, offering significant potential to lower greenhouse gas emissions and mitigate climate change. However, the cloud induced-variability of solar radiation reaching the earth’s surface presents a challenge for integrating solar power into the grid (e.g., storage and backup management). The new generation of geostationary satellites such as GOES-16 has become an important data source for large-scale and high temporal frequency solar radiation forecasting. However, no machine-learning-ready dataset has integrated geostationary satellite data with fine-grained solar radiation information to support forecasting model development and benchmarking with consistent metrics. We present SolarCube, a new ML-ready benchmark dataset for solar radiation forecasting. SolarCube covers 19 study areas distributed over multiple continents: North America, South America, Asia, and Oceania. The dataset supports short (i.e., 30 minutes to 6 hours) and long-term (i.e., day-ahead or longer) solar radiation forecasting at both point-level (i.e., specific locations of monitoring stations) and area-level, by processing and integrating data from multiple sources, including geostationary satellite images, physics-derived solar radiation, and ground station observations from different monitoring networks over the globe. We also evaluated a set of forecasting models for point- and image-based time-series data to develop performance benchmarks under different testing scenarios. The dataset is available at https://doi.org/10.5281/zenodo.11498739. A Python library is available to conveniently generate different variations of the dataset based on user needs, along with baseline models at https://github.com/Ruohan-Li/SolarCube. 
    more » « less
  4. ; ; (, Proceedings of the Thirteenth ACM International Conference on Future Energy Systems (e-Energy))
    Solar energy is now the cheapest form of electricity in history. Unfortunately, significantly increasing the electric grid's fraction of solar energy remains challenging due to its variability, which makes balancing electricity's supply and demand more difficult. While thermal generators' ramp rate---the maximum rate at which they can change their energy generation---is finite, solar energy's ramp rate is essentially infinite. Thus, accurate near-term solar forecasting, or nowcasting, is important to provide advance warnings to adjust thermal generator output in response to variations in solar generation to ensure a balanced supply and demand. To address the problem, this paper develops a general model for solar nowcasting from abundant and readily available multispectral satellite data using self-supervised learning. Specifically, we develop deep auto-regressive models using convolutional neural networks (CNN) and long short-term memory networks (LSTM) that are globally trained across multiple locations to predict raw future observations of the spatio-temporal spectral data collected by the recently launched GOES-R series of satellites. Our model estimates a location's near-term future solar irradiance based on satellite observations, which we feed to a regression model trained on smaller site-specific solar data to provide near-term solar photovoltaic (PV) forecasts that account for site-specific characteristics. We evaluate our approach for different coverage areas and forecast horizons across 25 solar sites and show that it yields errors close to that of a model using ground-truth observations. 
    more » « less
  5. ; ; ; (, Frontiers in Astronomy and Space Sciences)
    Supervised Machine Learning (ML) models for solar flare prediction rely on accurate labels for a given input data set, commonly obtained from the GOES/XRS X-ray flare catalog. With increasing interest in utilizing ultraviolet (UV) and extreme ultraviolet (EUV) image data as input to these models, we seek to understand if flaring activity can be defined and quantified using EUV data alone. This would allow us to move away from the GOES single pixel measurement definition of flares and use the same data we use for flare prediction for label creation. In this work, we present a Solar Dynamics Observatory (SDO) Atmospheric Imaging Assembly (AIA)-based flare catalog covering flare of GOES X-ray magnitudes C, M and X from 2010 to 2017. We use active region (AR) cutouts of full disk AIA images to match the corresponding SDO/Helioseismic and Magnetic Imager (HMI) SHARPS (Space weather HMI Active Region Patches) that have been extensively used in ML flare prediction studies, thus allowing for labeling of AR number as well as flare magnitude and timing. Flare start, peak, and end times are defined using a peak-finding algorithm on AIA time series data obtained by summing the intensity across the AIA cutouts. An extremely randomized trees (ERT) regression model is used to map SDO/AIA flare magnitudes to GOES X-ray magnitude, achieving a low-variance regression. We find an accurate overlap on 85% of M/X flares between our resulting AIA catalog and the GOES flare catalog. However, we also discover a number of large flares unrecorded or mislabeled in the GOES catalog. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email