结构系统识别不确定性分析的Bayes方法及其进展

发布时间：2018-05-14 12:26

本文选题：Bayes理论 + 系统识别　；参考：《应用数学和力学》2017年01期

【摘要】：受测试误差、建模误差、数值离散化以及环境变异等因素的影响,结构系统识别过程不可避免地存在不确定性,因此有必要引入概率统计方法来提高其鲁棒性,为工程结构安全监测提供更为可靠的结果.近年来,Bayes(贝叶斯)方法因为其诸多优势在系统识别领域受到了广泛关注.该文梳理了Bayes系统识别的历史脉络和研究进展.从Bayes系统识别的理论框架出发,分析了量化系统识别不确定性两类方法的适用条件与局限性.此外,文章综述了Bayes方法在模态参数识别、有限元模型修正以及结构损伤识别方面进行不确定性分析的理论、实现及其应用.最后对基于Bayes方法进行系统识别研究的发展趋势做出了展望.
[Abstract]:Due to the influence of test error, modeling error, numerical discretization and environmental variation, the process of structural system identification inevitably exists uncertainty. Therefore, it is necessary to introduce probability and statistics method to improve its robustness and provide more reliable results for engineering structure safety monitoring. In recent years, Bayes (Bias) method is due to its various methods. Many advantages have attracted wide attention in the field of system recognition. This paper combs the history and research progress of Bayes system recognition. Starting from the theoretical framework of Bayes system recognition, this paper analyzes the applicable conditions and limitations of the two methods of quantifying system identification uncertainty. In addition, the article reviews the identification of modal parameters and limited Bayes method. The theory, implementation and application of the uncertainty analysis of the metamodel and structural damage identification are carried out. Finally, the trend of the development of the system recognition based on the Bayes method is prospected.

【作者单位】：合肥工业大学土木与水利工程学院;
【基金】：国家自然科学基金(51408176;51278163) 国家重点研发计划(2016YFE0113400)~~
【分类号】：O212.8

【相似文献】