Computational fluid dynamics modeling of the structural influences on offshore wind measurements at the Cleveland water intake crib in Lake Erie

Abstract

Installation of dedicated offshore meteorological tower is very expensive due to the water depth and in some locations, icing conditions. However, many port cities have existing offshore structures such as water intake facilities or historical coastal defense structures that could be utilized as a low-cost alternative for wind resource assessment studies. Such a study was conducted using the city of Cleveland’s Water Division water intake facility located 3.5 miles offshore of downtown Cleveland in Lake Erie. Lake Erie is the smallest of the Great Lakes and as such can completely ice over during the winter. The water intake crib was installed in 1904 and has a small building on top of the water intake pipe. In 2005, a meteorological tower was installed on top of this building and has been recording wind measurements at 30, 40, and 50 m above the nominal water level. The influence of this structure on the recorded wind measurements was assessed using a computational fluid dynamics model. The assessment was conducted across the eight cardinal and ordinal directions and evaluated as a function of incoming wind profile. The model assumed a neutral boundary condition and the standard power law equation was used to predict the true mean wind speeds and wind shear coefficients. Input wind profiles with programmed wind shear coefficients of 0.0, 0.1, and 0.2 were found to be reduced above the crib structure and resulted in wind shear coefficients of −0.07, 0.05, and 0.17, respectively, at the locations of the cup anemometer measurements. This reduction of wind shear coefficient values was found to be caused by an increase in the wind speed above the building, which affected the lower height measurements more than the upper height measurements. The relationship between the known programmed wind shear coefficients and the lower resultant wind shear coefficients predicted from the computational fluid dynamics model was then used to adjust the experimental measurements to correct historical data. The corrected mean wind speeds occurring during a 2-year period beginning in October 2005 at 30, 40, and 50 m are 6.79, 7.02, and 7.16 m s⁻¹, respectively, as compared to the previously reported values of 7.14, 7.25, and 7.34 at 30, 40, and 50 m, respectively. These values represented the wind speeds which would have been measured using an isolated 50 m meteorological mast positioned directly on the water’s surface rather than atop the offshore water intake building. The wind shear coefficient averaged over the corrected wind speeds was 0.1038 as compared to the previously reported value of 0.0541.