Leveraging weather forecasts for a more efficient electric grid

Learn how GSL's High-Resolution Rapid Refresh weather model helps refine Dynamic Line Ratings for power lines, optimizing energy capacity and distribution (story map).

Power line conditions can be affected by factors both inside and outside of the physical wire. The line rating is affected by air temperature around the line, wind speed and direction across the line, and solar radiation incident on the line. GSL’s HRRR weather models are becoming an important piece in planning Dynamic Line Ratings, maximizing energy distribution along power lines.

With highly accurate observations, forecasts, and an understanding of how wind and solar resources fluctuate across time and space, the electric grid will better accommodate the variable nature of wind and solar energy. In particular, accurate day-ahead to week-two forecasts of hub-height winds and solar radiation at the surface allow energy providers to integrate renewable energy into the electrical grid more efficiently, making the grid more resilient. This will yield greater production of carbon-free renewable energy while also reducing air pollutant emissions. In addition, seasonal-to-decadal predictions from NOAA models are used for planning the development of new generation, transmission, and energy storage infrastructure, extreme event mitigation, and evaluating the impact of climate change on clean energy resources.

A closer look:

How NOAA's HIgh-Resolution Rapid Refresh wind forecasts result in $150 million in energy savings each year and pave the way for other renewables (story map)

Applications

The private sector uses predictions from NOAA weather prediction models as foundational input to their tailored forecast products. These models can also help the energy industry plan for the variable nature of wind and solar energy and can help target the most beneficial locations for clean energy development.

NOAA's Atmospheric Science for Renewable Energy (ASRE) program

ASRE works with the Department of Energy (DOE) and the wind and solar energy community to improve existing meteorological observing networks and weather forecast models. NOAA has conducted several field campaigns to collect data to study atmospheric processes associated with wind shear, turbulence, low-level jets, clouds, aerosols, and other aspects of the weather. These datasets are being analyzed to better understand atmospheric processes that affect winds at the height of wind turbines and the amount of sunshine hitting the solar panels.

Recent field campaigns

• The Wind Forecast Improvement Projects (WFIP) were a collaboration with the DOE’s Wind Energy Technologies Office (WETO). Two field campaigns have been conducted: one on initializing forecast models with observations to improve wind forecasts, and one in complex terrain. A third WFIP campaign is being organized with WETO to collect a dataset to improve off-shore wind energy predictions.
• The Position of Offshore Wind Energy Resources (POWER) campaign was conducted to enable better predictions of offshore wind resources to inform site selection for offshore wind farms. Lessons from POWER are being used in the design of the WFIP-3 campaign
• DOE’s Solar Energy Technologies Office (SETO) collaborated with ASRE, the National Center for Atmospheric Research, and IBM to conduct a field campaign to develop more accurate methods for solar forecasts using their state-of-the-art weather models.

In addition to these field campaigns, long-term observational datasets collected by NOAA (such as the SURFRAD surface radiation network stations) and the DOE’s Atmospheric Radiation Measurement (ARM) program are also used heavily by ASRE researchers.

Transitions

ASRE-supported research results in improvements to NOAA’s numerical weather prediction models, which are moved to operations within the National Weather Service approximately every two years.

Metrics

The HRRR parent model has improved its short-term forecasts of clouds and winds by 25% and 15% over the last decade. In a recent study, researchers found that better 12-hour wind forecasts from the HRRR model between versions 2 and 3 would have resulted in cost savings to the wind energy industry of over $200M per year. This result assumes that the wind industry used only the HRRR forecasts for their day-ahead decisions regarding the mix of wind-produced vs fossil fuel-produced energy. The improvement is due to both reduced overprediction and underprediction errors by the improved version 3 of the HRRR, relative to version 2. These savings would have a direct impact on household incomes through reduced energy costs.