Comparative assessment of simplified methods for hydrodynamic force on cylinder under earthquakes

Abstract

The effect of hydrodynamic force on the seismic response of structure in water cannot be ignored. Various methods have been developed to describe the effect of incompressible water, but the relative performance of some practical simplified methods is largely unknown. In this paper a comprehensive comparison of three representative simplified methods for assessing cylindrical hydrodynamic force is presented. Relative errors of structural elastic peak responses are constructed to measure the accuracy of the simplified method. The errors are tested on 154 circular cylinders subjected to 100 ground motions. The errors were first obtained by averaging the results of ground motions. Then the comparison between methods and the influence of parameters were studied based on the mean errors. Furthermore, by analyzing the statistical characteristic values of the errors, the kernel density estimation method was adopted to obtain the probability density of the errors. Finally, the probabilities of cylindrical cases and methods were investigated. The results show that if the added mass of hydrodynamic force depends on more structural parameters, the accuracy of the method may higher. The rigid-structure method has the highest accuracy and the most complicated expression. The accuracy and complexity of the method based on approximation of fundamental frequency are moderate. The Morison method has the simplest form and the maximum error. The results indicate that the added mass should relate to several structural parameters, ignoring the change of the added mass along the height will bring error to method and assuming the structure to be rigid is an effective way to simplify the analytical hydrodynamics.