Open AccessCorrelation of Simulated Results with Experimental Modal Analysis for an Asymmetric Structure
Author(s) -
Mitul Shah,
Madhuri Karnik
Publication year - 2019
Publication title -
international journal of engineering sciences
Language(s) - English
Resource type - Journals
ISSN - 0976-6693
DOI - 10.36224/ijes.120104
Subject(s) - welding , finite element method , modal analysis , stiffness , modal , structural engineering , materials science , microstructure , modal testing , composite material , engineering
The aim of this study is to investigate the reasons of the discrepancies in results of modal analysis due to Finite Element (FE) simulation and inaccuracies in Experimental Modal Analysis (EMA) for an asymmetric structure. An asymmetric structure with 87 stitch welds is simulated, tested and correlated. As the initial FE-EMA correlation showed frequency differences of over 24%, an optimization model is developed for the model to bring FE results close to EMA. FE optimization is performed with each weld modeled as a separate component which led to FE-EMA frequency difference reduced from over 24% to about 10%. The sources of error in FEA were identified as non-incorporation of manufacturing deviations such as efficacy of the welds, changes in stiffness due to changes in microstructure of Heat Affected Zone (HAZ), dimensional tolerances of componentsin assembly and variation of sheet thickness across the model.