• Interaction and multiscale mechanics
• /
• v.1 no.3
• /
• pp.369-379
• /
• 2008

A combined stochastic diffusion and mean-field model is developed for a systematic study of the grain growth in a pure single-phase polycrystalline material. A corresponding Fokker-Planck continuity equation is formulated, and the interplay/competition of stochastic and curvature-driven mechanisms is investigated. Finite difference results show that the stochastic diffusion coefficient has a strong effect on the growth of small grains in the early stage in both two-dimensional columnar and three-dimensional grain systems, and the corresponding growth exponents are ~0.33 and ~0.25, respectively. With the increase in grain size, the deterministic curvature-driven mechanism becomes dominant and the growth exponent is close to 0.5. The transition ranges between these two mechanisms are about 2-26 and 2-15 nm with boundary energy of 0.01-1 J $m^{-2}$ in two- and three-dimensional systems, respectively. The grain size distribution of a three-dimensional system changes dramatically with increasing time, while it changes a little in a two-dimensional system. The grain size distribution from the combined model is consistent with experimental data available.

• Structural Engineering and Mechanics
• /
• v.33 no.4
• /
• pp.407-428
• /
• 2009

The effects of the uniform and spatially varying ground motions on the stochastic response of fluid-structure interaction system during an earthquake are investigated by using the displacement based fluid finite elements in this paper. For this purpose, variable-number-nodes two-dimensional fluid finite elements based on the Lagrangian approach is programmed in FORTRAN language and incorporated into a general-purpose computer program SVEM, which is used for stochastic dynamic analysis of solid systems under spatially varying earthquake ground motion. The spatially varying earthquake ground motion model includes wave-passage, incoherence and site-response effects. The effect of the wave-passage is considered by using various wave velocities. The incoherence effect is examined by considering the Harichandran-Vanmarcke and Luco-Wong coherency models. Homogeneous medium and firm soil types are selected for considering the site-response effect where the foundation supports are constructed. A concrete gravity dam is selected for numerical example. The S16E component recorded at Pacoima dam during the San Fernando Earthquake in 1971 is used as a ground motion. Three different analysis cases are considered for spatially varying ground motion. Displacements, stresses and hydrodynamic pressures occurring on the upstream face of the dam are calculated for each case and compare with those of uniform ground motion. It is concluded that spatially varying earthquake ground motions have important effects on the stochastic response of fluid-structure interaction systems.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2006.05a
• /
• pp.354-359
• /
• 2006

The analysis of large-scale water resources systems is often complicated by the presence of multiple reservoirs and diversions, the uncertainty of unregulated inflows and demands, and conflicting objectives. Reinforcement learning is presented herein as a new approach to solving the challenging problem of stochastic optimization of multi-reservoir systems. The Q-Learning method, one of the reinforcement learning algorithms, is used for generating integrated monthly operation rules for the Keum River basin in Korea. The Q-Learning model is evaluated by comparing with implicit stochastic dynamic programming and sampling stochastic dynamic programming approaches. Evaluation of the stochastic basin-wide operational models considered several options relating to the choice of hydrologic state and discount factors as well as various stochastic dynamic programming models. The performance of Q-Learning model outperforms the other models in handling of uncertainty of inflows.

• Structural Engineering and Mechanics
• /
• v.39 no.4
• /
• pp.541-558
• /
• 2011

This paper presents an analysis of stochastic eigenvalue problem of plane bar structures. Particular attention is paid to the effect of spatial variations of the flexural properties of the structure on the first four eigenvalues. The problem of spatial variations of the structure properties and their effect on the first four eigenvalues is analyzed in detail. The stochastic eigenvalue problem was solved independently by stochastic finite element method (stochastic FEM) and Monte Carlo techniques. It was revealed that the spatial variations of the structural parameters along the structure may substantially affect the eigenvalues with quite wide gap between the two extreme cases of zero- and full-correlation. This is particularly evident for the multi-segment structures for which technology may dictate natural bounds of zero- and full-correlation cases.

• Journal of the Korean Institute of Telematics and Electronics A
• /
• v.31A no.6
• /
• pp.159-168
• /
• 1994

Stochastic computation is adopted to reduce the silicon area of the multipliers in implementing neural network in VLSI. In addition to this advantage, the stochastic computation has inherent random errors which is required for implementing Boltzmann machine. This random noise is useful for the simulated annealing which is employed to achieve the global minimum for the Boltzmann Machine. In this paper, we propose a method to implement the Boltzmann machine with stochastic computation and discuss the addition problem in stochastic computation and its simulated annealing in detail. According to this analysis Boltzmann machine using stochastic computation is suitable for the pattern recognition/completion problems. We have verified these results through the simulations for XOR, full adder and digit recognition problems, which are typical of the pattern recognition/completion problems.

• Smart Structures and Systems
• /
• v.9 no.3
• /
• pp.231-251
• /
• 2012

The stochastic optimal control for a piezoelectric spherically symmetric shell subjected to stochastic boundary perturbations is constructed, analyzed and evaluated. The stochastic optimal control problem on the boundary stress output reduction of the piezoelectric shell subjected to stochastic boundary displacement perturbations is presented. The electric potential integral as a function of displacement is obtained to convert the differential equations for the piezoelectric shell with electrical and mechanical coupling into the equation only for displacement. The displacement transformation is constructed to convert the stochastic boundary conditions into homogeneous ones, and the transformed displacement is expanded in space to convert further the partial differential equation for displacement into ordinary differential equations by using the Galerkin method. Then the stochastic optimal control problem of the piezoelectric shell in partial differential equations is transformed into that of the multi-degree-of-freedom system. The optimal control law for electric potential is determined according to the stochastic dynamical programming principle. The frequency-response function matrix, power spectral density matrix and correlation function matrix of the controlled system response are derived based on the theory of random vibration. The expressions of mean-square stress, displacement and electric potential of the controlled piezoelectric shell are finally obtained to evaluate the control effectiveness. Numerical results are given to illustrate the high relative reduction in the root-mean-square boundary stress of the piezoelectric shell subjected to stochastic boundary displacement perturbations by the optimal electric potential control.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.13 no.5
• /
• pp.29-37
• /
• 1993

The extension of the weighted integral method in the area of stochastic finite element analysis is presented. The use of weighted integral method in numerical analysis was extended to CST(constant strain triangle) element by Deodatis to calculate the response variability of 2D stochastic systems. In this paper, the extension of the weighted integral method for general plane-elements is represented. It has been shown that the same mesh used in the deterministic FE analysis can be used in the stochastic FE analysis. Furthermore, because the CST element is a special case which has constant strain-displacement matrix the mingling of CST elements with the other quadrilateral elements in the analysis may also be possible.

• Structural Engineering and Mechanics
• /
• v.43 no.5
• /
• pp.637-655
• /
• 2012

In this study, stochastic responses of a cable-stayed bridge subjected to the spatially varying earthquake ground motion are investigated by the finite element method taking into account soil-structure interaction (SSI) effects. The considered bridge in the analysis is Quincy Bay-view Bridge built on the Mississippi River in between 1983-1987 in Illinois, USA. The bridge is composed of two H-shaped concrete towers, double plane fan type cables and a composite concrete-steel girder deck. In order to determine the stochastic response of the bridge, a two-dimensional lumped masses model is considered. Incoherence, wave-passage and site response effects are taken into account for the spatially varying earthquake ground motion. Depending on variation in the earthquake motion, the response values of the cable-stayed bridge supported on firm, medium and soft foundation soil are obtained, separately. The effects of SSI on the stochastic response of the cable-stayed bridge are also investigated including foundation as a rigidly capped vertical pile groups. In this approach, piles closely grouped together beneath the towers are viewed as a single equivalent upright beam. The soil-pile interaction is linearly idealized as an upright beam on Winkler foundation model which is commonly used to study the response of single piles. A sufficient number of springs on the beam should be used along the length of the piles. The springs near the surface are usually the most important to characterize the response of the piles surrounded by the soil; thus a closer spacing may be used in that region. However, in generally springs are evenly spaced at about half the diameter of the pile. The results of the stochastic analysis with and without the SSI are compared each other while the bridge is under the sway of the spatially varying earthquake ground motion. Specifically, in case of rigid towers and soft soil condition, it is pointed out that the SSI should be significantly taken into account for the design of such bridges.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.2A
• /
• pp.383-390
• /
• 2006

In this paper, a stochastic field that is compatible with Monte Carlo simulation is suggested for an expansion-based stochastic analysis scheme of weighted integral method. Through investigation on the way of affection of stochastic field function on the displacement vector in the series expansion scheme, it is noticed that the stochastic field adopted in the weighted integral method is not compatible with that appears in the Monte Carlo simulation. As generally recognized in the field of stochastic mechanics, the response variability is not a linear function of the coefficient of variation of stochastic field but a nonlinear function with increasing variability as the intensity of uncertainty is increased. Employing the stochastic field suggested in this study, the response variability evaluated by means of the weighted integral scheme is reproduced with high precision even for uncertain fields with moderately large coefficient of variation. Besides, despite the fact that only the first-order expansion is employed, an outstanding agreement between the results of expansion-based weighted integral method and Monte Carlo simulation is achieved.

• Computers and Concrete
• /
• v.15 no.2
• /
• pp.259-277
• /
• 2015

The lack of experimental studies on the mechanical behavior of reinforced concrete (RC) haunched beams leads to difficulties in statistical and reliability analyses. This study performs stochastic and reliability analyses of the ultimate shear capacity of RC haunched beams based on nonlinear finite element analysis. The main aim of this study is to investigate the influence of uncertainty in material properties and geometry parameters on the mechanical performance and shear capacity of RC haunched beams. Firstly, 65 experimentally tested RC haunched beams and prismatic beams are analyzed via deterministic nonlinear finite element method by a special program (ATENA) to verify the efficiency of utilized numerical models, the shear capacity and the crack pattern. The accuracy of nonlinear finite element analyses is verified by comparing the results of nonlinear finite element and experiments and both results are found to be in a good agreement. Afterwards, stochastic analyses are performed for each beam where the RC material properties and geometry parameters are assigned to take probabilistic values using an advanced simulating procedure. As a result of stochastic analysis, statistical parameters are determined. The statistical parameters are obtained for resistance bias factor and the coefficient of variation which were found to be equal to 1.053 and 0.137 respectively. Finally, reliability analyses are accomplished using the limit state functions of ACI-318 and ASCE-7 depending on the calculated statistical parameters. The results show that the RC haunched beams have higher sensitivity and riskiness than the RC prismatic beams.