Detection of Debonding Damage Location of the Concrete Core from the Steel Tube of Concrete-Filled Steel Tube (CFST) Columns Using Wavelet Analysis Analytical Method

Structures get local damages by passing time during the service period under environmental conditions and loads, although insignificant. It is essential and important to maintain the health, durability and proper performance of structures and their various parts and lack of proper recognition of the behavior of structures may cause spontaneity damages and consequently, high social and economic costs may occur. According to the proper performance of CFST columns, using this type of columns in high-rise buildings and bridge structures has expanded especially in seismic areas. Steel and concrete can cover each otherchr('39')s weaknesses by simultaneously using concrete and steel in CFST columns. The weakness of concrete against tensile and the weakness of steel against pressure has compensated by the combination of steel and concrete in this type of columns. Also these columns may be damaged during construction or after experiencing load periods (earthquake, wind, etc.), because getting structures damage is inevitable. One of the primary goals of Structural Health Monitoring (SHM) is damages detection of the structure in the early stages of formation. If the damage locations in the structure can be determined and its gradual course can be observed, the damaged members can be repaired or replaced before reaching the critical condition and occurring complete breakdown. Among the methods of damage detection, many researchers consider the methods based on signal processing. One of the methods of signal processing is the mathematical method of wavelet analysis. By using wavelet analysis, more information can be obtained from the intended signal based on its ability to localize the signal in both time and frequency domains. One of the most probable damages in CFST columns is the debonding of the concrete core from the steel tube. In this paper, the CFST column element was modeled and frequency analyzed in ABAQUS finite element software in two conditions including damage and no-damage. The effect of the debonding was considered by decreasing the modulus of elasticity of the concrete in the damage places with depth of 3 mm. The results of the analysis have shown that the information of the mode shapes of the damage and no-damage conditions (angle between the mode shape vectors and the frequency values) changes due to the effect of the damage. In order to identify the debonding damage locations, in the Continuous Wavelet Transform (CWT) detection algorithm, the input signal was defined as the sum or difference of the mode shape of the damage condition and the mode shape of the no-damage condition based on the angle between the damaged and no-damaged mode shape vectors. The results showed that the output signals obtaining from the details of input signal wavelet analysis have useful information to identify the debonding locations of the concrete core from the steel tube and at high scales, the locations of the debonding damage identify easily, and at low scales, more convergence of wavelet coefficients is observed in the locations of the damage. According to the results, the proposed method was introduced as an effective detection method of debonding damage in CFST columns.