Wind Pressures on a Large Cooling Tower

Abstract

Mean/fluctuating wind pressure distributions on an actual large cooling tower (167-meter high) have been obtained and compared to previous results of other cooling towers. The agreement of mean wind pressure distributions is good. But the fluctuating wind pressure coefficients on the large cooling tower are much smaller than those on cooling towers of smaller sizes over the full range. It is assumed that the differences have resulted from the discrepancy of the turbulence intensity of the incoming flow, which is proven true according to wind tunnel model tests. Besides, it is found that most wind pressure signals produced by transducers around the throat section of the actual large cooling tower are of non-Gaussian distributions and they are not likely to be caused by the organized large-scale vortices according to the spatial correlation analyses. The upcoming flow of unsteady speed and direction in the engineering field might be an explanation for the phenomena. Two different methodologies (the traditional Davenport methodology and the complete probability methodology) are adopted to calculate the peak factors, in which the complete probability methodology is appropriate for non-Gaussian distributions. Studies about the spectral characteristics of wind-induced pressures on the actual large cooling tower are also included in this paper, and results show some differences from the previous study.