• Structural Engineering and Mechanics
• /
• v.20 no.2
• /
• pp.191-207
• /
• 2005

Time delay exists inevitably in active control, which may not only degrade the system performance but also render instability to the dynamic system. In this paper, a novel active controller is developed to solve the time delay problem in flexible structures. By using the independent modal space control method, the differential equation of the controlled mode with time delay is obtained from the time-delay system dynamics. Then it is discretized and changed into a first-order difference equation without any explicit time delay by augmenting the state variables. The modal controller is derived based on the augmented system using the discrete variable structure control method. The switching surface is determined by minimizing a discrete quadratic performance index. The modal coordinate is extracted from sensor measurements and the actuator control force is converted from the modal one. Since the time delay is explicitly included throughout the entire controller design without any approximation, the system performance and stability are guaranteed. Numerical simulations show that the proposed controller is feasible and effective in active vibration control of dynamic systems with time delay. If the time delay is not explicitly included in the controller design, instability may occur.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.20 no.4
• /
• pp.211-222
• /
• 2016

Reinforced concrete shear walls with deficient reinforcement details are tested under cyclic loading. The deficiency of reinforcement details includes insufficient splice length in U-stirrups at the ends of horizontal reinforcement and boundary column dowel bars found in existing low- to mid-rise Korean buildings designed non-seismically. Three test specimens have rectangular, babel and flanged sections, respectively. Flexure- and shear-controlled models for reinforced concrete shear walls specified in ASCE/SEI 41-13 are compared with the flexural and shear components of force-displacement relation extracted separately from the top displacement of the specimen based on the displacement data measured at diverse locations. Modification of the shear wall models in ASCE/SEI 41-13 is proposed in order to account for the effect of bar slip, cracking loads in flexure and shear. The proposed modification shows better approximation of the test results compared to the original models.

• Journal of Biomedical Engineering Research
• /
• v.25 no.3
• /
• pp.207-215
• /
• 2004

A finite element analysis is used in this study to model 6 kinds of electrode by a genetic program in order to evaluate the mechanical effect on the 3D human cochlea model and the behavior of electrode. Human cochlea is modeled by the spiral-approximation method and the shape of scala tympani is extracted from the mid section of the human cochlea. Contact pressure at the tip and the insertion force are found to be highest when the wires stack horizontally. Axial rotation of electrode is minimal comparing with the stimulating current spread. The results indicate that the electrode stiffness is important to minimize the trauma.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.23 no.8
• /
• pp.945-956
• /
• 1999

This work reports a set of approximate analytical solutions describing the initial transient process of close-contact melting between a rectangular parallelepiped solid and a flat plate on which either constant temperature or constant heat flux is imposed. Not only relative motion of the solid block tangential to the heating plate, but also the density difference between the solid and liquid phase is incorporated in the model. The thin film approximation reduces the force balance between the solid weight and liquid pressure, and the energy balance at the melting front into a simultaneous ordinary differential equation system. The normalized model equations admit compactly expressed analytical solutions which include the already approved two-dimensional solutions as a subset. In particular, the normalized liquid film thickness is independent of all pertinent parameters, thereby facilitating to define the transition period of close-contact melting. A unique behavior of the solid descending velocity due to the density difference is also resolved by the present solution. A new geometric function which alone represents the three-dimensional effect is introduced, and its properties are clarified. One of the representative results is that heat transfer is at least enhanced at the expense of the increase in friction as the cross-sectional shape deviates from the square under the same contact area.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.7 no.2
• /
• pp.193-203
• /
• 1983

The mixed convective heat transfer problems are characterized by the relatively significant contribution of buoyancy force to the transport processes of momentum and heat. Past analytical studies on this kind of problems have been carried out by employing either the conventional R-.epsilon. turbulence model which includes constant turbulent Prandtl number .sigma.$_{+}$ 1 or an extended R-.epsilon. turbulence model which takes account of the buoyancy effect in appropriate length scale equations. But in the latter case, the temperature variance .the+a.$^{2}$ over bar is approximated by a model under local equilibrium condition and the time scale ratio between velocity and temperature is assumed to be constant. These approximation is known to break down when the buoyancy effect is dominant. The present study is aimed at development of new computational turbulence closure level which can be applied to this rather complex turbulent process. The temperature variance is obtained directly by solving its dynamic transport equation and the time scale ratio which is variable in space is computed by a solution of a dynamic equation for the rate of scalar dissipation .epsilon.$_{\thetod}$ It was found that the computational results are in good agreement with available experimental data of wide range of unstable conditions.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.19 no.1
• /
• pp.1-21
• /
• 2015

This paper presents two algorithms based on the Jamshidian equation which is from the Black-Scholes partial differential equation. The first algorithm is for American call options and the second one is for American put options. They compute numerically free boundary and then option price, iteratively, because the free boundary and the option price are coupled implicitly. By the upwind finite-difference scheme, we discretize the Jamshidian equation with respect to asset variable s and set up a linear system whose solution is an approximation to the option value. Using the property that the coefficient matrix of this linear system is an M-matrix, we prove several theorems in order to formulate a bisection method, which generates a sequence of intervals converging to the fixed interval containing the free boundary value with error bound h. These algorithms have the accuracy of O(k + h), where k and h are step sizes of variables t and s, respectively. We prove that they are unconditionally stable. We applied our algorithms for a series of numerical experiments and compared them with other algorithms. Our algorithms are efficient and applicable to options with such constraints as r > d, $r{\leq}d$, long-time or short-time maturity T.

• Proceedings of the Korea Association of Crystal Growth Conference
• /
• 1999.06a
• /
• pp.45-49
• /
• 1999

When concentration of vacancies in a CZ silicon crystal is defined by molar fraction XB, the degree of super-saturation $\sigma$ is given by [XB-XBS]/XBS=XB/XBS-1=ln (XB/XBS) because XB/XBS is nearly equal to unity. Here, XBS is the saturated concentration of vacancies in a silicon crystal and XB is a little larger than XBS. According to Bragg-Williams approximation, the chemical potential of the vacancies in the crystal is given by ${\mu}$B=${\mu}$0+RT ln XB+RT ln ${\gamma}$, where R is the gas constant, T is temperature, ${\mu}$0 is an ideal chemical potential of the vacancies and ${\gamma}$ is an adjustable parameter similar to the activity of solute in a solution. Thus, $\sigma$(T) is equal to (${\mu}$B-${\mu}$BS)/RT. Driving force of nucleation of the vacancy agglomeration will be proportional to the chemical potential difference (${\mu}$B-${\mu}$BS) or $\sigma$(T), while growth of the vacancy agglomeration is proportional to diffusion of the vacancies and grad ${\mu}$B.

• Proceedings of the KSME Conference
• /
• 2002.08a
• /
• pp.365-368
• /
• 2002

The design of industrial arc plasma systems is still largely based on trial and error although the situation is rapidly improving because of the available computational power at a cost which is still fast coming down. The desire to predict the behavior of arc plasma system, thus reducing the development cost, has been the motivation of arc research. To interrupt fault current, the most enormous duty of a circuit breaker, is achieved by separating two contacts in a interruption medium, $SF_{6}$ gas or air etc., and arc plasma is inevitably established between the contacts. The arc must be controlled and interrupted at an appropriate current zero. In order to analyze arc behavior in $SF_{6}$ gas circuit breakers, a numerical calculation method combined with flow field and electromagnetic field has been developed. The method has been applied to model arc generated in the Aachen nozzle and compared the results with the experimental results. Next, we have simulated the unsteady flow characteristics to be induced by arcing of AC cycle, and conformed that the method can predict arc behavior in account of thermal transport to $SF_{6}$ gas around the arc, such as increase of arc voltage near current zero and dependency of arc radius on arc current to maintain constant arc current density.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.24 no.3
• /
• pp.450-458
• /
• 2000

This study deals with the unsteady close-contact melting of solid blocks on a flat surface subject to convective heating. Normalizing the model equations in reference to the steady solution successfully leads them to cover constant heat flux and isothermal limits at small and large extremes of the Biot number, respectively. The resulting equations admit a compactly expressed analytical solution, which includes the previous solutions as a subset. Based on the steady solution, the characteristics of close-contact melting can be categorized into constant heat flux, transition, and isothermal regimes, the boundaries of which appear to be nearly independent of the contact force. The unsteady solutions corresponding to Biot numbers in the transition regime show intermediate behaviors between those of the two limits. With a proper approximation, the present solution procedure can cope with the case of variable fluid temperature and heat transfer coefficient. Regardless of imposed conditions, the mean normalized Nusselt number during the unsteady process asymptotically approaches to a constant value as the Biot number comes close to each limit.

• Computational Structural Engineering
• /
• v.7 no.1
• /
• pp.117-124
• /
• 1994

Multi-level Multi-objective optimization(MLMO) for reinforced concrete framed structure is performed, and compared with the results of single-level single-objective optimization. MLMO method allows flexibility to meet the design needs such as deflection and cost of structures using weighting factors. Using Multi-level formulation, the numbers of constraints and variables are reduced at each levels, and the optimization formulation becomes simplified. The force approximation method is used to reflect the variation in design variables between the substructures, and thus coupling is maintained. And the linear approximated constraints and objective function are used to reduce the number of structural analysis in optimization process. It is shown that the developed algorithm with move limit can converge effectively to optimal solution.