【作者】 李颖；

【导师】 刘小明；

【作者基本信息】 武汉大学 ， 工程力学， 2004， 硕士

【摘要】 本文是对自锁锚杆在拉伸和剪切荷载共同作用下的锚固机理进行研究。试验选用两种直径(即φ16、φ20)和四种加载角度(即0°、30°、45°、60°)，共24根试件。观察锚杆整个受力和破坏过程，测得锚杆轴向、横向变形和极限破坏荷载；通过对荷载—变形曲线的分析锚杆在拉剪组合荷载作用下的变形规律；选用三种方程(直线方程、三直线方程以及椭圆方程)对锚杆的极限承载力进行描述和比较；分析混凝土和钢筋的破坏模式；并在此基础上利用有限元方法对此试验进行数值模拟，将计算结果与试验结果做比较；最后采用神经网络的方法对自锁锚杆的极限承载力预测进行了尝试，对预测值和试验值进行比较。得出以下一些研究成果： (1) 自锁锚杆在拉剪组合荷载作用下的荷载—变形曲线分析表明，锚杆钢筋的变形大致上可以分为两个阶段，前一阶段基本满足线性规律，后一阶段锚杆每荷载步的变形较大，非线性特征明显；无论是横向变形还是轴向变形，当加载角度为45°时，非线性特征最显著；随着加载角度的增加，两种锚杆的极限承载力均增加，直径为20mm的锚杆增加幅度要大；采用a=2的椭圆方程能较好地拟合本文试验中得到的锚杆极限承载力，为锚杆设计提供了理论依据。 (2) 在拉剪组合荷载作用下，锚杆的失效包括两部分，一部分是混凝土的斜锥体破坏，一部分是锚杆钢筋的断裂；当加载角度为30°时，混凝土体的破坏形状最接近正锥体，且锥顶深度最大，锥底面积最小；随着加载角度的增大，锥顶深度变浅，锥底面积最大。锚杆断裂后，断口形状具有如下规律：加载角度为30°的锚杆断口为杯口状，加载角度为45°的断口受拉部位为接近平面，受压区接近30°的斜面，加载角度为60°的锚杆断口则为接近45°的斜面。 (3) 采用有限元数值分析方法，分析了锚杆在拉剪组合荷载作用下的极限承载力，结果表明，数值计算结果与实测结果具有可比性。受压区混凝土应力表明，混凝土的破坏面呈扇形弧状分布。这与实际测试结果是一致的。 (4) 运用现代信息技术中神经网络方法对拉剪组合荷载作用下锚杆的极限承载力进行了预测。通过预测值与实验值的比较表明，该方法的预测结果具有较高的精度。利用训练好的神经网络，进行指定的输出，对锚杆极限承载力的影响因素进行了探讨。更多还原

【Abstract】 The anchorage mechanism of the locking anchor under tension together shear loads is studied in the paper. Twenty-four specimens which differ in two kind of diameters (F 16,F20) and four kinds of load angles (00,30?45?600), are selected in the experiment. The load-deforming curve is got through the experiment of the each single anchor under tension together shear loads, then the axial and transverse deformations and ultimate bearing capacity of the anchor can be got too. The deforming rule of the anchor under the tension and shear loads is found by analysis of the load-deforming curve. Three sorts of functions (straight-line function, trilinear function and elliptical function) are used to illustrate and compare the ultimate bearing capacity of anchor; analyzing the demolish mode of the concrete and the locking anchor. Moreover, numerical simulate on this experimentation with finite element method is conducted. The experiment and numerical results are compared with each other in succession. In the end, the ultimate bearing capacity of the locking anchor is forecasted with artificial neural network (ANN) method and the forecasting value is compared with the experiment value. The following achievements are obtained:(1) The load- deformation curve of the locking anchor under tension together shear loads shows that the deformation of the anchor can be mainly divided into two stages: in the first stage it satisfies the linear law; in the latter, there is a big deformation per load pace, and behaves as nonlinear characteristic in evidence. No matter it is axial deformation or transverse one, it behaves the most nonlinear when the load angle is 45 degrees. The ultimate bearing capacity of two kinds of anchor increase with the enhancement of the load angle, and the anchor with diameter 20 millimeters increase more obviously than the other one. The elliptical function with the parameter a equals 2 can fit the ultimate bearing capacity in the experiment very well, which can provide a theory basis for the design of the anchor.(2) Under the tension together shear loads the invalidation of the anchor consists of two parts: one is declined cone failure of the concrete, the other is the failure of the anchor;When the load angle is 30 degrees, the form of the failure is almost the cone one, whichinduces the biggest depth of the cone peak and smallest area of the cone bottom. With the increase of the load angle, the cone peak become lower in depth and the cone bottom becomes smaller in area. After the failure of the anchor, the fracture surfaces have the following results: it is a rim of a cup when the load angle is 30 degrees, nearly a plane in the drew part and near 30 degrees inclined plane in the compression area when the load angle is 45 degrees and near 45 degrees inclined plane in the fracture when the load angle is 60 degrees.(3) The ultimate bearing capacity of anchor under tension together shear loads is analyzed using the finite element method in the paper. The result shows that the numerical results are comparable with the experimentation ones. The stress in concrete compressive zone shows that the fracture surfaces of concrete performs sector-arc distribution which is consistent with the experiment results.(4) The ultimate bearing capacity of anchor under tension together shear loads is forecasted by neural network method in the modem information technique. According to the compare between the forecasting value and the experiment one, we are convinced that this method holds a high accuracy. The influence factors of the ultimate bearing capacity of anchor are studied using the trained neural network together with the special output.更多还原