Stability behaviour and dynamic response of cooling towers subjected to wind loading

• Main Author: Kucherera, Grant Tarwirei
• Other Authors: Zingoni, Alphose
• Format: Dissertation
• Language: English
• Published: University of Cape Town 2017
• Subjects: Civil Engineering
• Online Access: http://hdl.handle.net/11427/24305

In this study, a linear eigenvalue buckling parametric analysis is presented for various cooling tower shell geometries. The shells are subjected to increasing wind pressures (speeds) to observe the trends in the critical buckling pressures/speeds at which the shell first buckles and the corresponding buckling modes. The cooling tower's geometry is changed in a systematic manner to obtain the relationship between critical wind speeds associated with the first mode of buckling and the cooling tower's geometry. Geometrical parameter ratios of the cooling tower's dimensions are considered in order to cover a wider spectrum of the cooling tower's geometry. The critical wind speed versus height curve is observed to be similar to the Euler buckling curve. There appears to be a certain optimum throat height to total height ratio of about 0.75 for any cooling tower at which the critical wind speed is maximum. The critical wind speed varies linearly with the cooling tower thickness and non-linearly with all diameter ratios. A linear eigenvalue vibration parametric analysis is presented for various cooling tower shell geometries to observe trends in the free vibration response (natural frequencies and mode shapes). The forced response of the cooling tower to various forcing frequencies of wind gusts is analysed using the mode superposition method. The shells are subjected to increasing wind gust periods of the same speed to obtain the trends in the forced vibration response (response frequencies and modes). The cooling tower's geometry is changed in a systematic manner to obtain the free and forced vibration behaviour. The natural frequencies and their corresponding bandwidths for the first ten different modes reduce with increasing height. They are generally invariant with the height to top diameter ratio, but the bandwidth increases with increasing height to top diameter ratio. The response frequencies and their corresponding bandwidths generally decrease with increasing height as well as the height to top diameter ratios. The response frequency generally decreases with decreasing forcing frequency, but not for all the cooling tower geometries. The findings can be used as a basis for further research and establishment of conceptual design guidelines when considering stability, free and forced vibration cooling tower behaviour.