In this study, to observe the behavior of chemical anchors embedded in concrete under the tensile effect, 108 different anchor specimens were prepared with different parameters as concrete compressive strength, reinforcement bar diameters, anchor depths, sizes of drilled holes, cleanliness of the drilled holes. Pull-out tests were conducted and obtained data were examined with the axial-load capacities and the failure situations. Finally, the depth of anchors, compressive strength and reinforcement diameter were observed to increase the axial-load-bearing capacity. The specimens cleaned with water could bear more axial loads than cleaned using air. For the anchors installed without cleaning the holes, a significant decrease was observed in the axial-load carrying capacities compared to the other two conditions. The ANN algorithm exhibited a 78.3% prediction success compared with other algorithms. The empirical relations in the literature were found to have limited level of prediction success rates according to the ANN's results.

在这项研究中，为观察化学锚在拉伸作用下的行为，制备了108个不同锚的标本，这些标本具有不同的参数，例如混凝土的抗压强度，钢筋直径，锚深度，钻孔的大小，钻孔的清洁度。进行了拉拔试验，并用轴向载荷能力和破坏情况检查了获得的数据。最后，观察锚的深度，抗压强度和钢筋直径以增加轴向承载能力。用水清洁的样品比承受空气清洁的样品承受的轴向载荷更大。对于安装时未清理孔的锚，与其他两种情况相比，轴向承载能力明显下降。ANN算法显示为78。与其他算法相比，预测成功率为3％。根据人工神经网络的结果，发现文献中的经验关系具有有限的预测成功率。