提出一种用于评价锚杆锚固质量的定量分析方法。基于小波分析和人工神经网络建立了用于识别锚杆杆侧刚度系数的神经网络模型,利用样本对所建立的神经网络进行训练,为锚杆系统的锚固质量评价提供了一个有效的智能化的手段;通过对不同围岩下完整锚杆的数值模拟结果进行动力参数识别,得到了完整锚杆杆侧刚度系数与围岩弹性模量的关系曲线及相应的二次拟合公式,以此可作为锚杆锚固质量的衡量基准;提出了用于描述锚杆锚固状态的锚杆锚固度概念,建立了锚杆锚固质量定量分析方法并进行了工程应用。
A method is developed to quantitatively assess the quality of anchors. An analytical model is combined with the wavelet analysis and artificial neural network is used to estimate the values of bolt-side stiffness coefficients for the anchor system. The samples of bolt's anchorage system are fed into the artificial neural network for training. It's a useable intelligent mean to assess the quality of bolt's anchor system. The dynamic parameters of integrity bolt with different physical and mechanical parameters of surrounding rocks are acquired. A relationship curve between bolt-side stiffness and surrounding rock's modulus of deformation is erected, and its quadratic fitting formula is acquired. The curve and the formula can be used as the assessment standards for the anchor quality. The anchor degree is used to judge the anchor quality. The quantitative assessment method is applied to the engineering field.