Experimental and numerical study on quantitative detection of gap defects in CFST members using temperature method

The gap defect has a negative impact on the bearing capacity of concrete-filled steel tubular (CFST) members. It is necessary to strengthen the detection of gap defects in CFST members. This paper presents experimental and numerical studies to investigate the feasibility of using the temperature method to detect gap defects in CFST members. In the experiment, CFST members with preset gap defects were fabricated. The defect detection was performed using the temperature method. The results indicate that the cooling rate can effectively identify the preset defects. A three-dimensional finite element model of heat transfer for CFST members with gap defects was developed. The correctness of the model was validated through temperature comparisons and defect identification. Parametric analysis is conducted to summarize the defect identification characterization index. The analysis demonstrates that the defect area can be quantitatively identified using the cooling rate difference (CRD). The average and variance of the ratio between the preset and identified defect areas are 0.951 and 0.013, respectively. This confirms the validity of using CRD as a boundary for defect identification. When the defect area is known, the thermal contrast-heat flux curve can be used to quantitatively identify the defect depth. The average and variance of the ratio between the preset and identified defect depths are 0.995 and 0.004, respectively. In conclusion, the temperature method enables the quantitative identification of both the defect area and depth. The predictions are generally conservative, which is beneficial for practical engineering detection.