Reviewed by Lexie CornerOct 16 2024
Jiali Wang and colleagues at Argonne National Laboratory, the National Science Foundation National Center of Atmospheric Research, the National Renewable Energy Laboratory, Michigan Technological University, and Pacific Northwest National Laboratory conducted a comprehensive review of tropical storm observation technology. Their findings were published in the Journal of Renewable and Sustainable Energy by AIP Publishing.
Storm observation technology has exploded with the rise of advanced tools like autonomous uncrewed systems and machine learning. To design storm-resistant wind turbines, AI-powered modeling can help study storms at multiple scales to better reflect the complexities of tropical storms. Image Credit: Jiali Wang
With plans to install 30 gigawatts of new offshore wind capacity by 2030 and 110 gigawatts by 2050, the U.S. is moving quickly to expand offshore wind production. However, this growth depends on designing turbines capable of withstanding the challenges of tropical storms.
Extreme weather impacts on offshore wind turbines are not fully understood by the industry. Manufacturers design wind turbines based on international design standards, but better models and data are needed to study the impacts of extreme weather to inform and revise design standards.
Jiali Wang, Study Author and Atmospheric Scientist, Environmental Science Division, Argonne National Laboratory
They also examined data-driven models that use AI and machine learning as well as sophisticated physics-based modeling.
Wang added, “The intensity of extreme weather events is not well predicted by traditional methods. After reviewing the state-of-the-science technologies and methods, we need to do the work to bridge between the scales of weather data, whole wind farms, and individual wind turbines.”
For instance, the International Electrotechnical Commission’s offshore wind turbine standards could benefit from solid data provided by emerging technologies and data-sharing collaborations, as these standards currently do not account for the full complexity of extreme weather impacts on turbines.
The authors also note rapid advancements in modeling techniques, such as deep neural networks, which downscale regional data to point-scale data using super-resolution methods. Additionally, machine learning methods for dynamic warm potential predictions are improving storm intensity forecasts.
“We need models that address problems at very small scales, such as understanding what happens from one turbine to another. Satellites and other remote sensing technologies that can scan a region autonomously are helpful during extreme weather conditions, but their accuracy may be affected by heavy rain, and they cannot provide wind information at multiple altitudes like rotor heights,” Wang stated.
The authors emphasize that to improve storm predictions and update models and turbine design standards, it is crucial to use data that capture the complex interactions of multiple storm effects at different scales, especially considering the impact of climate change.
Wang concluded, “Both high winds and waves are damaging because waves can create energy that can drive ocean currents. These three components of wind, waves, and ocean currents can come from and go in different directions. This is known as misalignment and makes the turbine more vulnerable.”
Eric Hendricks, Christopher M. Rozoff, Matt Churchfield, Longhuan Zhu, Sha Feng, William J. Pringle, Mrinal Biswas, Sue Ellen Haupt, Georgios Deskos, Chunyong Jung, Pengfei Xue, Larry K. Berg, George Bryan, Branko Kosovic, and Rao Kotamarthi co-authored the study.
Journal Reference:
Wang, J. et. al. (2024) Modeling and observations of North Atlantic cyclones: Implications for U.S. Offshore wind energy. Journal of Renewable and Sustainable Energy. doi.org/10.1063/5.0214806