Operational modal identification using variational Bayes

详细信息    查看全文

• 作者：Binbin Li^a ; ^{bbl@berkeley.edu} ; Armen Der Kiureghian^b ; ^c ; ^{adk@ce.berkeley.edu} ; ^{President@aua.am}
• 关键词：Operational modal analysis ; Variational Bayes ; System identification ; State space model ; Linear time-invariant system
• 刊名：Mechanical Systems and Signal Processing
• 出版年：2017
• 出版时间：1 May 2017
• 年：2017
• 卷：88
• 期：Complete
• 页码：377-398
• 全文大小：2530 K
• 卷排序：88

文摘

Posterior distributions of modal parameters are obtained using the VB approach. The VB approach tends to slightly underestimate uncertainties in modal parameters. The uncertainties in modal parameters can be used to remove spurious modes. The VB approach is recommended comparing to the EM algorithm and Gibbs sampler.