基于贝叶斯方法的砌体结构自振频率推断

发布时间：2018-03-14 09:45

本文选题：砌体结构　切入点：自振频率　出处：《建筑结构学报》2014年10期 　论文类型：期刊论文

【摘要】：自振频率是既有砌体结构振动特性的重要指标,但其现场实测结果数据量有限且较为离散,难以对结构振动特性进行评估。为解决此问题,采用数理统计方法处理实测自振频率的不确定性。首先通过环境激励和人工激励,由动力测试获得试件自振频率的实测数据。然后基于贝叶斯方法,通过马尔科夫蒙特卡洛抽样,在实测数据中发掘总体信息;再结合对自振频率的先验判断,构建自振频率的样本后验分布。基于所获得的样本后验分布推断砌体试件的自振频率。结果表明,以上方法在实测数据量少、噪音影响较大的情况下具有很好的鲁棒性,能够较快收敛于稳定、合理的样本后验分布,为推断自振频率提供可靠的基础。结合某复杂既有砌体结构的自振频率监测过程,说明所提方法可用于工程检测。
[Abstract]:The natural vibration frequency is an important index of the vibration characteristics of existing masonry structures, but the measured results are limited and discrete, so it is difficult to evaluate the vibration characteristics of masonry structures. The uncertainty of measured natural vibration frequency is dealt with by mathematical statistical method. First, the measured data of natural vibration frequency of specimen are obtained by dynamic test through environmental excitation and artificial excitation, and then based on Bayesian method, Through Markov Monte Carlo sampling, the general information is excavated from the measured data, and a priori judgment of the natural vibration frequency is made. The natural vibration frequency of masonry specimens is inferred based on the sample posteriori distribution. The results show that the proposed method is robust in the case of less measured data and greater noise impact. It can quickly converge to a stable and reasonable sample posterior distribution, which provides a reliable basis for inferring the natural vibration frequency. Combined with the monitoring process of the natural vibration frequency of a complex existing masonry structure, it shows that the proposed method can be used for engineering detection.
【作者单位】：上海理工大学环境与建筑学院;上海大学土木工程系;
【基金】：国家自然科学基金项目(51208300)
【分类号】：TU364;TU311.3

【参考文献】