Abstract Structural health monitoring of super high‐rise buildings is of great importance for ensuring safe operation. GNSS‐RTK technology is widely used for dynamic monitoring, but its measurement accuracy is often influenced by environmental and instrumental noise. To eliminate such noise and obtain high‐precision structural dynamic deformation data along with accurate natural frequencies identification, this study proposes the improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN)‐independent component analysis (ICA)/Hilbert Transform (HT)‐Fast Fourier Transform (FFT) method, namely ICEEMDAN‐ICA/HT‐FFT. The performance of the proposed method was first validated using simulated signals and compared against the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN)‐ICA/HT‐FFT, ICEEMDAN, CEEMDAN approach. Results indicate that the proposed method yields lower mean absolute error (MAE) and root mean square (RMS) values, and its output more accurately reproduces both the fluctuations and amplitudes of the original signal. Furthermore, the ICEEMDAN‐ICA filtering successfully identified all five preset natural frequencies within the simulated signal. Subsequently, the method was applied to GNSS‐RTK measurements collected from the super high‐rise building (i.e., Tianjin Tower). The natural frequencies identified in the east–west (E‐W) and north–south (N‐S) directions were 0.1562 and 0.1567 Hz, respectively, with a maximum deviation of only 0.0024 Hz from the finite element analysis result of 0.1586 Hz, demonstrating high accuracy and reliability. In conclusion, GNSS‐RTK monitoring combined with the ICEEMDAN‐ICA/HT‐FFT method proposed can be effectively employed to evaluate the dynamic response of super high‐rise buildings under ambient excitement, yielding reliable results.
Cao et al. (Tue,) studied this question.