• Computers and Concrete
• /
• v.18 no.5
• /
• pp.1019-1039
• /
• 2016

In this paper, a parallel algorithm of nonlinear dynamic analysis of three-dimensional (3D) reinforced concrete (RC) frame structures based on the platform of graphics processing unit (GPU) is proposed. Time integration is performed using Newmark method for nonlinear implicit dynamic analysis and parallelization strategies are presented. Correspondingly, a parallel Preconditioned Conjugate Gradients (PCG) solver on GPU is introduced for repeating solution of the equilibrium equations for each time step. The RC frames were simulated using fiber beam model to capture nonlinear behaviors of concrete and reinforcing bars. The parallel finite element program is developed utilizing Compute Unified Device Architecture (CUDA). The accuracy of the GPU-based parallel program including single precision and double precision was verified in comparison with ABAQUS. The numerical results demonstrated that the proposed algorithm can take full advantage of the parallel architecture of the GPU, and achieve the goal of speeding up the computation compared with CPU.

• Structural Engineering and Mechanics
• /
• v.74 no.4
• /
• pp.547-557
• /
• 2020

In this paper, an automatic adaptive mesh refinement procedure is presented for two-dimensional problems on the basis of a new probabilistic error estimator. First-order perturbation theory is employed to determine the lower and upper bounds of the structural displacements and stresses considering uncertainties in geometric sizes, material properties and loading conditions. A new probabilistic error estimator is proposed to reduce the mesh dependency of the responses dispersion. The suggested error estimator neglects the refinement at the critical points with stress concentration. Therefore, the proposed strategy is combined with the classic adaptive mesh refinement to achieve an optimal mesh refined properly in regions with either high gradients or high dispersion of the responses. Several numerical examples are illustrated to demonstrate the efficiency, accuracy and robustness of the proposed computational algorithm and the results are compared with the classic adaptive mesh refinement strategy described in the literature.

• Structural Engineering and Mechanics
• /
• v.17 no.3_4
• /
• pp.593-609
• /
• 2004

Peak response is a more suitable index than mean response in the light of structural safety. In this study, a controller optimization method is proposed to restrict peak responses of building structures subject to earthquake excitations, which are modeled as partially stationary stochastic process. The constraints are given with specified failure probabilities of peak responses. LQR is chosen to assure stability in numerical process of optimization. Optimization problem is formulated with weightings on controlled outputs as design variables and gradients of objective and constraint functions are derived. Full state feedback controllers designed by the proposed method satisfy various design objectives and output feedback controllers using LQG also yield similar results without significant performance deterioration.

• Interaction and multiscale mechanics
• /
• v.1 no.4
• /
• pp.423-435
• /
• 2008

This paper employs the reproducing kernel (RK) approximation for evaluation of field theory-based incompatibility tensor in a polycrystalline plasticity simulation. The modulation patterns, which is interpreted as mimicking geometrical-type dislocation substructures, are obtained based on the proposed method. Comparisons are made using FEM and RK based approximation methods among different support sizes and other evaluation conditions of the strain gradients. It is demonstrated that the evolution of the modulation patterns needs to be accurately calculated at each time step to yield a correct physical interpretation. The effect of the higher order strain derivative processing zone on the predicted modulation patterns is also discussed.

• Structural Engineering and Mechanics
• /
• v.61 no.2
• /
• pp.275-282
• /
• 2017

The Fleck-Hutchinson theory on strain gradient plasticity (SGP), proposed in Adv. Appl Mech 33 (1997) 295, has recently been reformulated by adopting the strategy of decomposing the second order strain presented by Lam et al. in J Mech Pays Solids 51 (2003) 1477. The newly built SGP satisfies the non negativity of plastic dissipation, which is still an outstanding issue in other SGP theories. Furthermore, it explicitly shows how elastic strain gradients and corresponding elastic characteristic length scales come into play in general elastic-plastic loading histories. In this study, the relation between elastic length scales and plastic length scales is investigated by taking wire torsion as an example. It is concluded that the size effects arising when two sets of length scales are of the same order are essentially elastic instead of plastic.

• 한국연소학회:학술대회논문집
• /
• 2004.06a
• /
• pp.7-14
• /
• 2004

The lift-off characteristics of the triple flame have been studied experimentally with various mean velocities and concentration gradients using a multi-slot burner, which can control the concentration gradient and the mean velocity independently, Lift-off height, axial maximum velocity, flame temperature, and some other characteristics were examined for methane and propane flame, It was found that minimum values of the lift-off heights exist at a certain concentration gradient for constant mean velocity, and this result implies that the propagation velocity has a maximum value at this condition, OH radical distribution was measured with LIF method and velocity variation along streamline was measured with PlV system. In addition maximum temperature along streamline was measured with CARS system. The intensity of the diffusion flame affects on the propagation velocity of triple flame in the region of very weak concentration gradient.

• Bulletin of the Korean Mathematical Society
• /
• v.51 no.1
• /
• pp.287-301
• /
• 2014

In this paper, we prove a necessary optimality theorem for a nonsmooth optimization problem in the face of data uncertainty, which is called a robust optimization problem. Recently, the robust optimization problems have been intensively studied by many authors. Moreover, we give examples showing that the convexity of the uncertain sets and the concavity of the constraint functions are essential in the optimality theorem. We present an example illustrating that our main assumptions in the optimality theorem can be weakened.

• Journal of Multimedia Information System
• /
• v.1 no.2
• /
• pp.119-126
• /
• 2014

We propose a bicycle detection system riding on people based on modified projected local binary pattern(PLBP) for vision based intelligent vehicles. Projection method has robustness for rotation invariant and reducing dimensionality for original image. The features of Local binary pattern(LBP) are fast to compute and simple to implement for object recognition and texture classification area. Moreover, We use uniform pattern to remove the noise. This paper suggests that modified LBP method and projection vector having different weighting values according to the local shape and area in the image. Also our system maintains the simplicity of evaluation of traditional formulation while being more discriminative. Our experimental results show that a bicycle and motorcycle riding on people detection system based on proposed PLBP features achieve higher detection accuracy rate than traditional features.

• 한국연소학회:학술대회논문집
• /
• 2012.11a
• /
• pp.355-357
• /
• 2012

A probability density function (PDF) approach to account for turbulence-chemistry interaction in the context of large eddy simulation (LES) based simulation of scramjets is developed. To solve the high-dimensional joint-composition PDF transport equation robustly, the semi-discrete quadrature method of moments (SeQMOM) is recently proposed [1]. The SeQMOM approach addresses key numerical issues in LES related to the inaccuracies in computing filter-scale gradients, enabling an efficient and numerically consistent solution of the PDF transport equation. The computational tool is used to simulate a cavity-stabilized Mach 2.2 supersonic combustor.

• Nuclear Engineering and Technology
• /
• v.30 no.1
• /
• pp.17-25
• /
• 1998

Transport of carbon and boron impurity ions parallel to magnetic field lines in the tokamak SOL (scrape-off layer) is numerically investigated for a one-dimensional steady state. The spatial distributions of density and velocity of the impurity ions in a steady state are calculated by finite difference method for a single-fluid model. The calculated results show that among forces acting on SOL particles thermal force produced tv plasma temperature gradient is a principal force determining the feature of impurity distribution profiles in the tokamak edge. However, strong collisional friction forces appearing dominant in front of the diverter plate restrain impurity ion flows due to temperature gradients from moving toward the midplane. Consequently, the stagnation point develops in the impurity flow by these two forces near the diverter region, in which ion flows change their directions. Impurity ions turn out to be accumulated at the stagnation points, where peaked profiles of highly-ionized state ions are relatively predominant over those of low-ionized state ions.