Proper Orthogonal Decomposition and its Use in the Analysis of Fluid Structure Interaction Noise

Author(s):  
Stewart A. Glegg ◽  
William J. Devenport ◽  
Nicholas J. Molinaro ◽  
William N. Alexander
Author(s):  
Erwan Liberge ◽  
Mustapha Benaouicha ◽  
Aziz Hamdouni

This paper describes Reduced Order Modeling (ROM) in Fluid Structure Interaction (FSI) and discusses Proper Orthogonal Decomposition (POD) utilization. The ROM method was selected because its performance in fluid mechanics. The principal problems of its application in FSI are due the space character of the modes resulting from the POD whereas domains are mobile. To use POD in moving domain, a charateristic function of fluid is introduced in order to work on a fixed rigid domain, and the global velocity (fluid and structure) is studied. The POD modes efficiency is tested to reconstruct velocity field in one and two-dimensional FSI case. Then reducing dynamic system using POD is introduced in moving boundaries problem. In addition, the one dimensional case of Burgers equation coupled with spring equation is tested.


Author(s):  
Wentao Hao ◽  
Ling Tian ◽  
Bingshu Tong

Because of their good performance to speed up MEMS system simulation processes, macromodels have aroused lots of attentions of scientists in the last decades. However, studies on FSI (Fluid-Structure Interaction) MEMS devices still can not satisfy the macromodeling requests because of the high complexity of fluid fields. A new method based on modal analysis and POD (Proper Orthogonal Decomposition) is tentatively put forward to reduce the order of FSI MEMS models. The structure macromodeling theory is firstly reviewed. Then the fluid field macromodeling approach is discussed in detail. At last, a 2D fixed-fixed micro-beam is analyzed and the results show that the macromodel extracted in this method can highly decrease the system degrees of freedom, while its precision is still comparable with that of detailed models.


Sign in / Sign up

Export Citation Format

Share Document