Synchro-Reassigning Transform: A Novel Approach to Earthquake Damage Identification in Structures

Structures may lose rigidity and exhibit a reduction in their fundamental natural frequency following a strong earthquake. Analysing the time–frequency representation of instrumented structure’s responses reveals key information about the changes in their frequency characteristics over time. The short-time Fourier transform (STFT), Wigner–Ville distribution, reassigned smoothed pseudo-Wigner–Ville distribution, wavelet transform, and Hilbert–Huang transform are among the well-known time–frequency techniques that are commonly employed for building damage detection. This is the first application of synchro-reassigning transform (SRT) for improved frequency-shift detection of structural damage after an earthquake, with greater robustness and resolution compared to other time–frequency transforms. This study thoroughly examines its performance by analysing two synthetic signals and three real earthquake signals: two from the 1994 Northridge earthquake and one from the 1971 San Fernando earthquake. The simulation results show that the SRT method provides improved time–frequency resolution and fewer distortions and is less sensitive to noise compared to well-established time–frequency techniques. Owing to these advantages, SRT could be more effective and promising in detecting frequency shifts for postearthquake structural assessment.