A comparative study of dynamic mode decomposition methods for mode identification in a cryogenic swirl injector. (7th July 2021)

• Record Type: Journal Article
• Authors: Luong, Hung Truyen; Wang, Yuangang; Sung, Hong-Gye; Sohn, Chae Hoon
• Is Part Of: Journal of sound and vibration. Volume 503(2021)
• Publication Date: 2021-07-07
• Subjects: Dynamic mode decomposition -- Resonant frequencies -- Damping coefficient -- Acoustic-mode shapes -- Cryogenic swirl injector; Sound -- Periodicals; Vibration -- Periodicals; Son -- Périodiques; Vibration -- Périodiques; Sound; Vibration; Periodicals; Electronic journals; 620.205
• DOI: 10.1016/j.jsv.2021.116108
• Languages: English
• ISSNs: 0022-460X

Abstract

Abstract: The dynamic mode decomposition (DMD) is one of the post-processing methods used in analyzing dynamic or transient flow field data from experiments or CFD analyses. This method extracts resonant frequencies, damping coefficients, and spatial structures of mode shapes without a priori physical knowledge of the specific form of structure or power spectrum density analysis. The optimized DMD method is applied to improve the accuracy of DMD results in finding the eigen-components by fitting the nonlinear dynamic system from initial eigenvalues and modes determined by applying the exact DMD. A comparison is made between the optimized DMD and the exact DMD method. Two DMD methods are applied to the numerical data calculated with a cryogenic swirl injector to show validity of their algorithms in identifying acoustic modes of a dynamic system. Compared with the exact DMD, the optimized DMD can extract feasibly the acoustic modes of the dynamic system with higher accuracy. Furthermore, the optimized DMD can find the resonant modes of the system more completely.