Study On Ultrasonic Multi-parameter Joint Inversion Of Concrete Strength

The microstructure of concrete is complex,so it is very important to describe and analyze the microstructure of concrete by nondestructive testing.Traditional ultrasonic testing of concrete has the advantages of high efficiency and stability.The ultrasonic strength measurement USES the single parameter of wave velocity to estimate the concrete strength,which is obviously not sensitive enough to the microstructure of materials.In view of this,nonlinear acoustic parameters are introduced and linear and nonlinear ultrasonic multi-parameters are combined to realize the prediction of concrete strength.The research work and main research results of this paper are as follows:(1)Considering the influence of concrete strength,concrete specimens with different mix ratios were designed and prepared,the ultrasonic testing signal and strength of concrete are obtained through ultrasonic testing and uniaxial compression testing,the ultrasonic signal is analyzed and processed by spectral analysis method,and the ultrasonic parameters of concrete are extracted.(2)The variation law of ultrasonic parameters when ultrasonic wave penetrates concrete of different strength is analyzed,the linear parameters with good correlation with concrete strength were extracted;Using the nonlinear wave theory to solve the one-dimensional nonlinear wave equation,the relative nonlinear coefficient is defined and the expression is given,and through the correlation analysis of nonlinear coefficient with aggregate particle size and water-cement ratio,the variation rule of nonlinear ultrasonic parameter is obtained.(3)Through the correlation analysis between concrete strength and ultrasonic parameters,the influence of concrete strength on ultrasonic parameters such as wave velocity,amplitude and nonlinear coefficient is obtained,the feasibility of using nonlinear parameters to predict concrete strength is proved;By analyzing the influence of concrete strength and material on acoustic parameters,the characteristic parameters required for the prediction model are obtained.(4)Based on the correlation analysis of concrete strength and material to acoustic parameters,the characteristic variables of the prediction model are extracted,the regression processing of data is completed by using random forest algorithm,and the prediction model of concrete strength is established.Through the optimization of the model and the calculation of the importance of the characteristics and the analysis of the evaluation index,it is proved that the model has a high accuracy.Finally,the prediction results of the model on the strength of concrete prove the feasibility of the model.