Classification and Noise Correlation of Local Wind Turbine Signals in Grant County Oklahoma

Session: Waveform Cross-Correlation-Based Methods in Observational Seismology [Poster]
Type: Poster
Date: 4/30/2020
Time: 08:00 AM
Room: Ballroom
Description: 

Due to the increasing demands of highly dense seismic networks and the shift towards renewable energy sources such a wind energy, it is common that both seismic networks and wind farms will share locations across the United States in regions such as Oklahoma. Seismic noise generated by wind turbines poses an interesting scenario where both natural wind source coupling and resonance frequencies of the turbine blades and towers are transmitted through the subsurface. Understanding the characteristics of the noise field associated with wind turbines is important to develop better techniques in noise suppression, identify near-surface resonance and improve signal-to-noise ratio. We investigate the wind and wind turbine generated noise within the seismic field through the application of the power density function and noise correlation on 3-component waveforms collected from a temporary array of 5 Hz geophone sensors. The temporary array consists of 8 Fairfield nodes that were active for one-month with varying distances (10 meters - 2000 meters) from wind turbine towers located in Grant County, Oklahoma. We also investigate and compare waveform signals measured from another 64 Fairfield node temporary array that was deployed temporally concurrent, 70km west in Alfalfa County and lacks proximity to any wind turbines. The spectral amplitudes and peak frequencies of the power spectrum show spatial temporal variations in noise levels in respect to the location of the wind turbine towers. Noise amplitude decreases with distance from the wind turbine and are not detectable in Alfalfa County. Wind speed correlates power spectrum peak frequencies observed at multiple frequencies along with a constant peak at approximately 0.27 Hz which is explained as the resonance of the wind turbine tower. Directivity is characterized using cross-correlation function where time lag indicates signal propagation. We observe the wind turbines to be a very clear source of seismic noise with power degradation at distance and can better identify these signals through comparison with the distant array.

Presenting Author: Raymond Ng

Authors