Please use this identifier to cite or link to this item: https://idr.l4.nitk.ac.in/jspui/handle/123456789/10467
Title: Assessment of Cohesive Parameters Using High Dimensional Model Representation for Mixed Mode Cohesive Zone Model
Authors: Kesava, Rao, B.
Balu, A.S.
Issue Date: 2019
Citation: Structures, 2019, Vol.19, , pp.156-160
Abstract: Simulation of the mechanical behavior of bonded joints using a cohesive zone model (CZM) is the most common technique to characterize the delamination process. It is usually dependent on the calculation of cohesive parameters of the traction-separation law, and the parameters are iteratively obtained with the help of simulation and experimental results. The non-availability of standard methods to obtain the parameters necessitates the iterative adjustments of simulation results to the experimental results. However, the calculations based on all individuals for the simulation are not effective as it demands high computational effort. To overcome this issue, this paper proposes a computationally efficient method using high dimensional model representation (HDMR). The cohesive parameters are determined by adopting an efficient sampling scheme within the limits of the parameters. Single leg bending (SLB) joint is tested under the influence of dominant conditions such as mode-I and mode-II to determine the equivalent parameters. The errors resulted from the comparison between the simulation, and experimental values are minimized in order to determine the optimal values. The mixed mode (MM) CZM is then established by pure mode cohesive parameters, and the same is implemented to the SLB joint under various mode mixities for analyzing the fracture process. Comparison between the numerical analysis and the experimental study proves that the proposed HDMR based approach estimates the failure mechanism exactly. 2019 Institution of Structural Engineers
URI: http://idr.nitk.ac.in/jspui/handle/123456789/10467
Appears in Collections:1. Journal Articles

Files in This Item:
There are no files associated with this item.


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.