Tall Column Reliability under Nonstationary Loads: Model Application

Abstract

A nonstationary stochastic model is used to determine the reliability of tall industrial columns. A typical distilling column from an oil refinery is defined for the analysis and a standard set of ground motion parameters is selected. The importance of each of the model parameters is then determined by changing the parameter and repeating the analysis. As a result of these analyses, the duration of ground motions is found to have a significant effect on the reliability of the column. The frequency content of the ground motion becomes important when the fundamental frequency of the ground motion resonates with the natural frequency of the first mode of vibration of the column. The importance of including several modes of vibration in the column response is observed to be small for elastic deformations. Vertical acceleration is found to affect the reliability only slightly. Higher modes and vertical acceleration become more important as the slenderness of the column increases.