Breakthrough machine learning approach quickly produces higher-resolution climate data

July 07, 2020

Researchers at the U.S. Department of Energy's (DOE's) National Renewable Energy Laboratory (NREL) have developed a novel machine learning approach to quickly enhance the resolution of wind velocity data by 50 times and solar irradiance data by 25 times--an enhancement that has never been achieved before with climate data.

The researchers took an alternative approach by using adversarial training, in which the model produces physically realistic details by observing entire fields at a time, providing high-resolution climate data at a much faster rate. This approach will enable scientists to complete renewable energy studies in future climate scenarios faster and with more accuracy.

"To be able to enhance the spatial and temporal resolution of climate forecasts hugely impacts not only energy planning, but agriculture, transportation, and so much more," said Ryan King, a senior computational scientist at NREL who specializes in physics-informed deep learning.

King and NREL colleagues Karen Stengel, Andrew Glaws, and Dylan Hettinger authored a new article detailing their approach, titled "Adversarial super-resolution of climatological wind and solar data," which appears in the journal Proceedings of the National Academy of Sciences of the United States of America.

Accurate, high-resolution climate forecasts are important for predicting variations in wind, clouds, rain, and sea currents that fuel renewable energies. Short-term forecasts drive operational decision-making; medium-term weather forecasts guide scheduling and resource allocations; and long-term climate forecasts inform infrastructure planning and policymaking.

However, it is very difficult to preserve temporal and spatial quality in climate forecasts, according to King. The lack of high-resolution data for different scenarios has been a major challenge in energy resilience planning. Various machine learning techniques have emerged to enhance the coarse data through super resolution--the classic imaging process of sharpening a fuzzy image by adding pixels. But until now, no one had used adversarial training to super-resolve climate data.

"Adversarial training is the key to this breakthrough," said Glaws, an NREL postdoc who specializes in machine learning.

Adversarial training is a way of improving the performance of neural networks by having them compete with one another to generate new, more realistic data. The NREL researchers trained two types of neural networks in the model--one to recognize physical characteristics of high-resolution solar irradiance and wind velocity data and another to insert those characteristics into the coarse data. Over time, the networks produce more realistic data and improve at distinguishing between real and fake inputs. The NREL researchers were able to add 2,500 pixels for every original pixel.

"By using adversarial training--as opposed to the traditional numerical approach to climate forecasts, which can involve solving many physics equations--it saves computing time, data storage costs, and makes high-resolution climate data more accessible," said Stengel, an NREL graduate intern who specializes in machine learning.

This approach can be applied to a wide range of climate scenarios from regional to global scales, changing the paradigm for climate model forecasting.

NREL is the U.S. Department of Energy's primary national laboratory for renewable energy and energy efficiency research and development. NREL is operated for the Energy Department by the Alliance for Sustainable Energy, LLC.
-end-


DOE/National Renewable Energy Laboratory

Related Renewable Energy Articles from Brightsurf:

Creating higher energy density lithium-ion batteries for renewable energy applications
Lithium-ion batteries that function as high-performance power sources for renewable applications, such as electric vehicles and consumer electronics, require electrodes that deliver high energy density without compromising cell lifetimes.

Renewable energy targets can undermine sustainable intentions
Renewable energy targets (RETs) may be too blunt a tool for ensuring a sustainable future, according to University of Queensland-led research.

Intelligent software for district renewable energy management
CSEM has developed Maestro, an intelligent software application that can manage and schedule the production and use of renewable energies for an entire neighborhood.

Renewable energy transition makes dollars and sense
New UNSW research has disproved the claim that the transition to renewable electricity systems will harm the global economy.

Renewable energy advance
In order to identify materials that can improve storage technologies for fuel cells and batteries, you need to be able to visualize the actual three-dimensional structure of a particular material up close and in context.

Illuminating the future of renewable energy
A new chemical compound created by researchers at West Virginia University is lighting the way for renewable energy.

Using fiber optics to advance safe and renewable energy
Fiber optic cables, it turns out, can be incredibly useful scientific sensors.

Renewable energy developments threaten biodiverse areas
More than 2000 renewable energy facilities are built in areas of environmental significance and threaten the natural habitats of plant and animal species across the globe.

Could water solve the renewable energy storage challenge?
Seasonally pumped hydropower storage could provide an affordable way to store renewable energy over the long-term, filling a much needed gap to support the transition to renewable energy, according to a new study from IIASA scientists.

Scientists take strides towards entirely renewable energy
Researchers have made a major discovery that will make it immeasurably easier for people (or super-computers) to search for an elusive 'green bullet' catalyst that could ultimately provide entirely renewable energy.

Read More: Renewable Energy News and Renewable Energy Current Events
Brightsurf.com is a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to Amazon.com.