Open Access

ARTICLE

Damage Detection in CFST Column by Simulation of Ultrasonic Waves Using STFT-Based Spectrogram and Welch Power Spectral Density Estimate

by Nadom K. Mutlib^1,*, Muna N. Ismael¹, Shahrizan Baharom²

1 Anbar Technical Institute, Middle Technical University, Baghdad, 10013, Iraq
2 Department of Civil and Structural Engineering, National University of Malaysia, Selangor, 43600, Malaysia

* Corresponding Author: Nadom K. Mutlib. Email:

Structural Durability & Health Monitoring 2021, 15(3), 227-246. https://doi.org/10.32604/sdhm.2021.010725

Received 23 March 2020; Accepted 20 November 2020; Issue published 07 September 2021

Abstract

Structural health monitoring employs different tools and techniques to provide a prediction for damages that occur in various structures. Damages such as debond and cracks in concrete-filled steel tube column (CFST) are serious defects that threaten the integrity of the structural members. Ultrasonic waves monitoring applied to the CFST column is necessary to detect damages and quantify their size. However, without appropriate signal processing tools, the results of the monitoring process could not be crucial. In this research, a monitoring process based on a Multiphysics numerical simulation study was carried out. Two signal processing tools: short time Fourier transform (STFT) and Welch Power Spectral Density Estimate (PSD) were used to analyse the captured raw signals. The STFT spectrogram was effective in identifying the different size of damage based on a graphical interpretation. The results show that the increasing of frequency of the excited signal give a better results. The increase in peak magnitude values in Welch PSD was found to be proportionate to the change in damage length whereas the damage depth has a less effect. The results for the crack size identification were less promising than those of debond damage because of the different type of the signal’s propagation path. Simulation process conducted by COMSOL software has proved the validity of the adopted signal processing techniques in detecting such damages in CFST columns.

---

Keywords

---