• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.31 no.2
• /
• pp.87-95
• /
• 2018

This paper introduces a new explicit spectral element method for the simulation of SH-waves in semi-infinite domains. To simulate the wave motion in unbounded domains, it is necessary to reduce the infinite extent to a finite computational domain of interest. To prevent the wave reflection from the trunctated boundaries, perfectly matched layer(PML) wave-absorbing boundary is introduced. The forward problem for simulating SH-waves in PML-truncated domains can be formulated as second-order PDEs. The second-order semi-discrete form of the governing PDEs is constructed by using a mixed spectral elements with Legendre-gauss-Lobatto quadrature method, which results in a diagonalized mass matrix. Then the second-order semi-discrete form is transformed to a first-order, whose solutions are calculated by the fourth-order Runge-Kutta method. Numerical examples showed that solutions of SH-wave in the two-dimensional analysis domain resulted in stable and accurate, and reflections from truncated boundaries could be reduced by using PML boundaries. Elastic wave propagation analysis using explicit time integration method may be apt for solving larger domain problems such as three-dimensional elastic wave problem more efficiently.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2013.04a
• /
• pp.89-90
• /
• 2013

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.24 no.6
• /
• pp.591-599
• /
• 2011

This study investigates influencing factors on numerical crash simulation between RC slab and soft projectile using explicit dynamic method. Considered experimental test is the MEPPEN II/4 test, which has been conducted at the end of the years 70' in Germany as one of the numerous experimental test related to design of nuclear power plants. LS-DYNA software is adopted for numerical study, and influencing factors such as constitutive model of concrete, strain rate effect of steel and concrete, support modeling method, etc. are investigated. More reasonable simulation results can be achieved through appropriate consideration of these factors, especially of constitutive model of concrete material since this factor affects most among the investigated factors.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.27 no.2
• /
• pp.79-86
• /
• 2014

This paper presents modelling approaches for the global and distortional buckling of cold-formed built-up steel sections using the finite element software packages, ANSYS and ABAQUS. Thin thickness of the cold-formed steel causes nonlinear behaviour due to local and distortional buckling, thus careful consideration is required in modelling for numerical analysis. Implicit static modelling using ANSYS provides unstable numerical results as the load approaches the limit point but explicit dyamic modelling with ABAQUS is able to display the behaviour even in post-buckling range. Meanwhile, axial load capacities obtained from the numerical analysis show higher values than the experimental axial capacities, due to eccentricity during the test. Axial capacities of the cold-formed steel obtained through numerical analysis requires reduction factor, and this paper suggests 0.88 for the factor.