Crack Identification by Source Inversion Analysis of AE Wave Motions

本研究的目的是开发一种技术，可以识别裂纹运动的声发射（AE）波动。从理论上讲，AE波形表示的积分公式的基础上的边界元法（BEM）。引入位错模型作为裂纹的数学模型，导出了由弹性常数、裂纹面法线方向矢量和裂纹运动方向矢量组成的矩张量表示。因此，裂纹运动学上的裂纹类型和裂纹取向的恢复，提供了六个独立的张量分量是完全确定的。为此，研制了一个计算机软件SiGMA（Simplified绿色函数用于矩张量分析）。通过矩张量分量的特征值分析和统一分解，将声发射源分为拉伸裂纹和剪切裂纹。由第一特征向量的方向确定了拉伸裂纹的张开方向。根据第一、三特征向量的方向反演剪切运动的方向，并将反演结果应用于现场水压致裂试验和混凝土断裂试验的声发射数据。水力压裂试验的数值研究证实了所提出的程序的合理精度。圆柱拉伸试验和锚栓拉拔试验的结果表明，声发射反演方法对裂纹运动学的识别具有很好的应用前景。

The aim of the present research is to develop a technique which can identify crack kinematics from acoustic emission (AE) wave motions. Theoretically, AE waveforms are represented by an integral formulation based on the boundary element method (BEM). Introducing a dislocation model as a mathematical model for a crack, the moment tensor representation is derived.The moment tensor consists of elastic constants, a direction vector of the unit normal to crack surface, and a direction vector of crack motion. Thus, crack kinematics on crack type and crack orientation are recovered, provided that six independent tensor components are fully determined. For this purpose, a computer software named SiGMA (Simplified Green's function for Moment tensor Analysis) is developed. Following the eigenvalue analysis of moment tensor components and a unified decomposition, AE sources are classified into tensile cracks and shear cracks. Crack opening direction of the tensile crack is obtained from the direction of the first eigenvector. The direction of shear motion is recovered from the directions of the first and the third eigenvectors.The inversion procedure developed is applied to AE data on in situ hydrofracturing tests and fracture tests in concrete. Numerical studies on hydrofracturing tests confirm the reasonable accuracy of the proposed procedure. Results of cylinder-tensile tests and pullout tests of anchor bolt show a great promise for identifying crack kinematics by AE inversion procedure.