English：

Aiming at the problem of difficult non⁃destructive inspection of internal defects in steelreinforced concrete structures under complexworking conditions, this paper selects the piezoelectric wavemethod and proposes a comprehensive correction coefficient method based on the frequency band energyratio and entropyvalue normalization to process the data of the original detection signal energy. In view ofthe different variation patterns of vibration response spectrum characteristics caused by different damagefactors, a linear regression model based on the influence coefficient of damage factor weights is proposedas a defect evaluation index to quantitatively analyze the degree of damage. Through the defect detectiontest of the steel reinforced concrete model column, the accuracy of damage identification of thepiezoelectric sensor under the composite working conditions inside the structure is verified. The testresults show that the defect detection method for steel reinforced concrete based on piezoelectrictechnology is applicable to concrete of different strength grades. With the increase of detection distanceand damage degree, the signal energy decreases, and the defect evaluation index DI gradually increases.The wavelet packet energyvalue under the three⁃obstacle condition decreases by 85.42% compared withthe barrier⁃free condition, and the damage index DI gradually increases from 0 to 0.86. Finally, theproposed defect detection method is used to conduct defect detection and evaluation on the structurecolumns of an actual engineering project,verifying the applicability of the proposed defect detectionmethod in the actual structure.