The reflection of the HF radar signal from the spinning blades of an offshore wind turbine has been observed to have an impact on HF radar systems in Europe. With the first U.S. offshore wind turbine facility going operational in 2016 in the monitoring area of the Mid-Atlantic Regional Association Coastal Ocean Observing System (MARACOOS) SeaSonde network, the proximity of radars and turbines prompted a study to better understand the potential effects on the U.S. National HF Radar Network. Construction of several more offshore wind turbines is planned to begin in the coming years, begging the question: what can be done to minimize the impact on the U.S. National HF Radar Network?
|
Unlike fixed objects, the periodic change in radar cross-section (RCS) of a turbine creates a signal spike in the Doppler cross spectra, and since the turbine blades are longer than a wavelength, those peaks appear in more than one range/Doppler bin and will vary in range and Doppler as the turbine rotation rates change.
This presents a challenge for processing software. CODAR is now almost half way into a two-year study funded by the Bureau of Ocean Energy Management (BOEM) to assess, classify, and mitigate the turbine-induced interference. With six radar sites, including both 25 MHz and 5MHz systems on Block Island, this first offshore wind farm is proving to be an excellent testbed.
Due to the complex nature of the interference, the initial turbine study relies on the comparison between numerical simulations and radar data. For simulations, we use the Numerical Electro Magnetic Code (NEC), developed at Lawrence Livermore National Laboratory, to simulate the RCS of the turbines with their fan blades in different positions. The RCSs are then put into a time series corresponding with a given rotation rate. A simulated cross spectra of expected interference is then produced by running the time series through CODAR’s dual FFT FMCW processing. By using this process with the fan blades spinning at different RPM, we can see the expected location of the turbine interference in the SeaSonde cross spectra. Ultimately, this helps us distinguish the difference between peaks in SeaSonde cross spectra from vessels and turbine interference as well as if the turbine contamination falls in the Bragg region. By comparing the simulated data to radar data, we can also estimate the rotation rates of the turbines and use this information to design noise mitigation techniques. Now that we have been able to classify the structure of the interference we are moving forward with our efforts into removal methods.
Study funding was provided by the U.S. Department of the Interior, Bureau of Ocean Energy Management, Environmental Studies Program, Washington, D.C, under Contract Number M16PC00017. |
A comparison between SeaSonde cross spectra (Top) when the turbines were rotating at 11.4 rpm, and NEC simulated cross spectra (Middle) showing turbine interference. The turbine interference is located in the first range cell (Bottom). The strongest peaks in the simulated data are all present in the SeaSonde data. |
