Identifying the dynamic characteristics of super tall buildings by multivariate empirical mode decomposition

Abstract

In this study, multivariate empirical mode decomposition (MEMD) is used to evaluate the dynamic characteristics of super‐tall buildings. Two super‐tall buildings, including Milad Tower, which is located in Tehran, Iran, and Canton Tower, which is located in Guangzhou, China, are used as examples to estimate the capability of multivariate empirical mode decomposition for recognizing the dynamic characteristics of buildings. A method is suggested to extract the frequency of structures automatically. First, the best segment of required data, including acceleration response and wind speed is found, and then wavelet transform is used to eliminate the noise and find proper and wanted natural frequency. Finally, to investigate all signals, that is, acceleration responses of all channels simultaneously, MEMD is applied to identify the frequency of the filtered signals. The extracted frequencies are selected in the order of amplitude power of each mode for each intrinsic mode function (IMF). The obtained results are appropriate, corresponding to other studies. Hence, the proposed method can automatically select the accurate frequency of super‐tall buildings in less time duration by considering all required data simultaneously.