轻质混凝土与钢筋粘结强度概率模型分析

Probabilistic model for bond strength between lightweight concrete and steel bar

提出了一种新的轻质混凝土与钢筋粘结强度概率计算方法。结合676组粘结锚固试验数据和现有粘结强度计算公式,引入贝叶斯动态更新理论,完成了粘结强度影响因素的显著性分析,建立了轻质混凝土与钢筋粘结强度多参数概率预测模型,利用贝叶斯参数剔除法简化概率模型,采用分项系数量化分析预测模型的连续性和不确定性,提出了临界锚固长度建议公式。结果表明:混凝土抗拉强度、锚固长度、钢筋直径、保护层厚度、配箍率等因素对轻质混凝土与钢筋粘结性能的影响显著;概率模型预测值与试验值比值的均值、变异系数及分项系数分别为1.056、0.298和1.361,该概率模型具有较高预测精度和连续性,能够较准确地进行轻质混凝土与钢筋粘结强度的预测。

A new bond strength determination of lightweight concrete and rebar is proposed through a statistical approach.The Bayesian method establishes a multi-parameter bond prediction model based on a database of 676 experimental data and four existing deterministic models.Key explanatory terms are identified and the probability model is simplified in a systematic removal process.Partial factor method is used to evaluate model uncertainty.The formula of critical anchorage length is put forward.Some important factors affect the bond performance,including concrete tensile strength,anchorage length,reinforcement diameter,cover thickness,stirrup ratio.The proposed model can improve accuracy and great continuity with the Mean,CoV and partial factors of experimental-to-predicted results ratio being 1.056,0.298 and 1.361 respectively.