Structure-preserving Galerkin POD reduced-order modeling of Hamiltonian systems

详细信息    查看全文

• 作者：Yuezheng Gong^a ; Qi Wang^b ; ^a ; ^c ; Zhu Wang^b ; ^{wangzhu@math.sc.edu}
• 关键词：37M25 ; 65M99 ; 65P10 ; 93A15
• 刊名：Computer Methods in Applied Mechanics and Engineering
• 出版年：2017
• 出版时间：1 March 2017
• 年：2017
• 卷：315
• 期：Complete
• 页码：780-798
• 全文大小：738 K
• 卷排序：315

文摘

We propose a new structure-preserving reduced-order model for the Hamiltonian system. A rigorous a priori error estimate is presented for the structure-preserving ROMs. The approach modifies the traditional Galerkin projection-based POD-ROM. Motivated by the estimate, weighted gradient information of Hamiltonian is used. POD basis from shifted snapshots is used to improve the approximation to Hamiltonian.