A study on structural SEA model construction by base excitation

This paper describes a structure model construction for structural statistical energy analysis (SEA) using FEM. Identigying SEA parameter regardless of the excitation method is demanded effectively. The proposed method was based on a combination of SEA and vibration response by base excitation like...

Full description

Bibliographic Details
Main Authors: Katsuhiko KURODA, Shuhei NAMIKAWA
Format: Article
Language:Japanese
Published: The Japan Society of Mechanical Engineers 2019-10-01
Series:Nihon Kikai Gakkai ronbunshu
Subjects:
Online Access:https://www.jstage.jst.go.jp/article/transjsme/85/879/85_19-00043/_pdf/-char/en
Description
Summary:This paper describes a structure model construction for structural statistical energy analysis (SEA) using FEM. Identigying SEA parameter regardless of the excitation method is demanded effectively. The proposed method was based on a combination of SEA and vibration response by base excitation like large mass method, forced displacement vibration analysis and random vibration analysis. In order to evaluate the SEA parameters, it is necessary to calculate the subsystem energies and input power. However, the input power at the excitation point when the entire structure is oscilatted is unknown. So, it was assumed that the input power is proportional to the vibration energy of the excited subsystem. These method are effective on an analytical cost performance for the base excitation compared with the conventional force excitation method that is called rain-on-the-roof-exctation. In this study, firstly, the proposal method was validated through two oscillators coupled by a spring. Then it was also validated through L plate consiting of two subsystems. As a result, large mass method is valid for predicting the coupling loss factors between subsystems and in terms of analytical cost. Finally, the method was applied to two flat plates connected in 60, and 30 degrees configuration and a partial car model consisting of six subsystems, as a result, the method is shown to work quantitatively well to the internal loss factors and qualitatively well to the coupling loss factors.
ISSN:2187-9761