论文部分内容阅读
贝叶斯理论具有充分利用模型信息和数据信息且考虑先验分布等优点,已被广泛应用于各个领域。通过贝叶斯多元线性参数估计方法,建立基于影响参数的钢筋混凝土深受弯构件贝叶斯概率抗剪模型。基于该模型和271组钢筋混凝土深受弯构件试验结果,完成了模型参数计算及基于贝叶斯理论的受剪承载力计算,同时,利用贝叶斯参数剔除法对抗剪模型进行简化,并与我国混凝土结构设计规范(GB50010-2010)、ACI318-08、CSA和EC2等现有规范计算结果进行了对比分析。研究表明:利用贝叶斯方法建立的基于影响参数的深受弯构件抗剪模型计算结果与试验吻合良好,采用贝叶斯动态更新后结果较规范值更接近试验值,简化后模型能较为合理的进行深受弯构件受剪承载力计算。
Bayesian theory has the advantages of taking full advantage of model information and data information and considering a priori distribution, and has been widely used in various fields. Based on the Bayesian multiple linear parameter estimation method, a Bayesian probabilistic shear model based on influence parameters for deep flexural members of reinforced concrete is established. Based on the model and the experimental results of 271 deep flexural members, the calculation of model parameters and the calculation of shear capacity based on Bayesian theory were completed. At the same time, the shear model was simplified by Bayesian parameter culling It is compared with the calculation results of the existing codes such as Concrete Structure Design Code (GB50010-2010), ACI318-08, CSA and EC2 in our country. The results show that the results of Bayesian method based on the influence parameter are in good agreement with the experimental results. The result of Bayesian dynamic update is closer to the experimental value than the normalized value, and the simplified model can be more reasonable The calculation of the shear strength of deep flexural members.