基金项目:河南省科技厅攻关项目(182107000007); 河南工业大学横向项目(ZZDT/SMWH-B/20171024)
第一作者:陈桂香(1976-),博士,教授,博士生导师,主要从事土木工程现代建造技术与管理.E-mail:chen-guixiang@163.com
DOI: 10.15986/j.1006-7930.2020.02.006
承插型盘扣式钢管支架盘扣节点半刚性明显,是架体力学性能的重要影响因素之一.因此对架体进行精确数值分析时,需要合理选取盘扣节点扭矩 - 转角数学模型.在盘扣节点抗扭试验的基础上,采用数理统计方法获取盘扣节点扭矩 - 转角的二参数对数模型和三线性模型.通过架体荷载试验和有限元数值计算,对比分析盘扣节点单线性模型、三线性模型和二参数对数模型的适用性.三种盘扣节点模型参数选取合理时,架体有限元数值计算结果均可满足实际工程对稳定承载力的精度要求,其中盘扣节点采用二参数对数模型时,架体极限稳定承载力有限元数值计算值与试验值吻合性最好.由于立杆荷载分配不均、上下立杆连接半刚性、构件间隙和过盈装配等因素,架体侧移和盘扣节点扭矩的有限元数值计算值与试验值相差明显,结合试验数据规律,采用系数修正法对有限元数值计算结果进行了修正.
The semi-rigidity of the disk-pin joint node of the disk lock steel tubular scaffold is obvious, which is one of the important factors affecting the mechanical properties of the frame. Therefore the reasonable selection of the torque-rotation angle mathematical model of the disk-pin joint node is the precondition of accurate numerical analysis of the frame. On the basis of the torsion test of the disk-pin joint node, the two-parameter logarithmic model and the trilinear model of the torque-rotation angle of the disk-pin joint node were obtained by mathematical statistics method. The applicability of single linear model, trilinear model and two-parameter logarithmic model was compared and analyzed by frame load test and finite element numerical calculation. The three kinds of disk-pin joint node models with reasonable parameters can satisfy the accuracy requirements of the stability bearing capacity of the practical engineering. When the two-parameter logarithmic model was adopted for the disk-pin joint node, the numerical calculation value of the ultimate stability bearing capacity of the frame was in the best agreement with the experimental value. Due to the uneven distribution of the load on the standing tube, the semi-rigid connection of the upper and lower standing tubes, the clearance and the interference fit, the calculated values of the lateral displacement of the frame and the torque of the disk-pin joint nodes are obviously different from the experimental values. The finite element calculation results were corrected by the coefficient correction method combined with the experimental data.