Open AccessThe improvement of bolted joints model via finite element model updating method
Author(s) -
M. A. Yunus,
M. N. Abdul Rani,
M. A. S. Aziz Shah,
M. S. M. Sani,
Z. Yahya
Publication year - 2021
Publication title -
journal of mechanical engineering and sciences
Language(s) - English
Resource type - Journals
eISSN - 2231-8380
pISSN - 2289-4659
DOI - 10.15282/jmes.15.4.2021.16.0682
Subject(s) - fastener , structural engineering , finite element method , bolted joint , deformation (meteorology) , sheet metal , joint (building) , engineering , modal , computer science , materials science , composite material
Efficient schemes to represent mathematical model of thin-sheet metal structures jointed by bolted joints for accurately predict the structure dynamic behaviour has been a significant unresolved issue in structural dynamics community. The biggest challenge is to efficiently incorporate the joints local deformation effects on the developed mathematical model via finite element (FE) method. Generally, the joints local deformation typically exerts on the joints mating area. To solve this issue, this paper proposes efficient schemes to represent mathematical model of thin-sheet metal structures jointed by bolted joints with application to accurately calculate the structure dynamic behaviour using FE model updating method. The initial FE model of the assembled structure was developed by employed Fastener Connector (CFAST) in MSC NASTRAN software to represent the bolted joints while, the inclusion of the local deformation effects at the bolted joints mating area was represented by contact elements. Then, the responses obtained from the FE model was evaluated by weight up with experimental data. FE model updating (FEMU) method then was utilised for minimising prediction discrepancies originated from the initial FE model based on the experimental data. The proposed scheme shows the accuracy of the initial prediction was improved from 25.03 % to 14.65 % while the accuracy of the predicted mode shapes via modal assurance criterion (MAC) analysis were above 0.8. Therefore, the findings offer useful schemes for improving the quality of predicted dynamic behaviour, particularly in the thin-sheet metal jointed structure and the developed model can be used with confident for any subsequence dynamic analyses.