• ETRI Journal
• /
• 제45권1호
• /
• pp.119-130
• /
• 2023

We propose a quantized gradient search algorithm that can achieve global optimization by monotonically reducing the quantization step with respect to time when quantization is composed of integer or fixed-point fractional values applied to an optimization algorithm. According to the white noise hypothesis states, a quantization step is sufficiently small and the quantization is well defined, the round-off error caused by quantization can be regarded as a random variable with identically independent distribution. Thus, we rewrite the searching equation based on a gradient descent as a stochastic differential equation and obtain the monotonically decreasing rate of the quantization step, enabling the global optimization by stochastic analysis for deriving an objective function. Consequently, when the search equation is quantized by a monotonically decreasing quantization step, which suitably reduces the round-off error, we can derive the searching algorithm evolving from an optimization algorithm. Numerical simulations indicate that due to the property of quantization-based global optimization, the proposed algorithm shows better optimization performance on a search space to each iteration than the conventional algorithm with a higher success rate and fewer iterations.

• 한국자동차공학회논문집
• /
• 제6권3호
• /
• pp.143-153
• /
• 1998

In this reaserch, a program for real time simulation of a vehicle is developed. This program uses relative coordinates to save the computation time and BDF(Backward Difference Formula) to integrate system variables. Numerical tests were performed for J-turn and Lane change steering, respectively. The validity of the program is proved by the ADAMS package. Numerical results showed that the proposed implicit method is more stable in carrying out the numerical integration for vehicle dynamics than the explicit method. Hardware requirements for real time simulation are suggested.

• 동력기계공학회지
• /
• 제2권3호
• /
• pp.21-26
• /
• 1998

Numerical simulations for various fluid flows require enormous computing time during iterations. In order to solve this problem, several techniques have been proposed. A SOR method is one of the effective methods for solving elliptic equations. However, it is very difficult to find the optimum relaxation factor, the value of this factor for practical problems used to be estimated on the basis of expertise. In this paper, the implication of the relaxation factor are translated into fuzzy control rules on the basis of the expertise of numerical analysers, and fuzzy controller incorporated into a numerical algorithm. From two cases of study, Poisson equation and cavity flow problem, we confirmed the possibility of computational acceleration with fuzzy logic and qualitative reasoning in numerical simulations. Numerical experiments with the fuzzy controller resulted in generating a good performance.

• Journal of Mechanical Science and Technology
• /
• 제17권7호
• /
• pp.986-998
• /
• 2003

In this paper, an adaptive numerical integration scheme, which does not need non-overlapping and contiguous integration meshes, is proposed for the MLPG (Meshless Local Petrov-Galerkin) method. In the proposed algorithm, the integration points are located between the neighboring nodes to properly consider the irregular nodal distribution, and the nodal points are also included as integration points. For numerical integration without well-defined meshes, the Shepard shape function is adopted to approximate the integrand in the local symmetric weak form, by the values of the integrand at the integration points. This procedure makes it possible to integrate the local symmetric weak form without any integration meshes (non-overlapping and contiguous integration domains). The convergence tests are performed, to investigate the present scheme and several numerical examples are analyzed by using the proposed scheme.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2002년도 학술대회지
• /
• pp.103-106
• /
• 2002

This paper presents an effective numerical method for analyzing three-dimensional unsteady conjugate heat transfer problems of a curved pipe subjected to infernally thermal stratification. In the present numerical analyses, the thermally stratified flows in the pipe are simulated using the standard $k-{\varepsilon}$turbulent model and the unsteady conjugate heat transfer is treated numerically with a simple and convenient numerical technique. The unsteady conjugate heat transfer analysis method is implemented in a finite volume thermal-hydraulic computer code based on a non-staggered grid arrangement, SIMPLEC algorithm and higher-order bounded convection scheme. Numerical calculations have been performed far the two cases of thermally stratified pipe flows where the surging directions are opposite each other i.e. In-surge and out-surge. The results show that the present numerical analysis method is effective to solve the unsteady flow and conjugate heat transfer in a curved pipe subjected to infernally thermal stratification.

• Environmental Engineering Research
• /
• 제11권3호
• /
• pp.164-171
• /
• 2006

Numerical modeling of the flow velocity fields for the near corona wire electrohydrodynamic (EHD) flow was conducted. The steady, two-dimensional momentum equations have been computed for a wire-plate type electrostatic precipitator (ESP). The equations were solved in the conservative finite-difference form on a fine uniform rectilinear grid of sufficient resolution to accurately capture the momentum boundary layers. The numerical procedure for the differential equations was used by SIMPLEST algorithm. The Phoenics (Version 3.5.1) CFD code, coupled with Poisson's electric field, ion transport equations and the momentum equation with electric body force were used for the numerical simulation and the Chen-Kim ${\kappa}-{\varepsilon}$ turbulent model numerical results that an EHD secondary flow was clearly visible in the downstream regions of the corona wire despite the low Reynolds number for the electrode ($Re_{cw}=12.4$). Secondary flow vortices caused by the EHD increases with increasing discharge current or EHD number, hence pressure drop of ESP increases.

• 융합신호처리학회논문지
• /
• 제9권1호
• /
• pp.82-88
• /
• 2008

본 논문은 주기적 진동 문제를 연구하기 위해 Runge-Kutta(RK)-Butcher 알고리듬이 소개되었다. RK-Butcher 알고리듬을 사용하여 얻어진 시뮬레이션 결과와 고전적인 4차 RK(4) 방법을 통해 얻은 결과들을 제안한 알고리듬의 성능을 확인하기 위하여 몇몇 주기적 진동 문제들의 정확한 해와 비교하였다. RK-Butcher 알고리듬의 시뮬레이션 결과는 항상 문제의 정확한 해 RK(4) 방법보다 더 근접한 결과를 줌이 확인되었다. 정확도 측면에서 RK-Butcher 알고리듬이 RK(4) 방법과 비교해볼 때 우수함을 알 수 있다. 제안한 RK-Butcher 알고리듬은 프로그램 언어로 쉽게 구현할 수 있으며 임의 시간에 종료해도 훌륭한 근사적인 해를 얻을 수 있다. RK-Butcher 알고리듬은 짧은 시간내에 이상적인 정확한 해에 근접한 결과를 주기 때문에 궤도 와 두 물체의 문제를 연구하는데 훌륭한 수치 알고리듬으로 적용 가능하다.

• Structural Engineering and Mechanics
• /
• 제14권2호
• /
• pp.171-190
• /
• 2002

Two cases of design are performed for the hyperbolic paraboloid saddle shell (Lin-Scordelis saddle shell) and the hyperbolic cooling tower (Grand Gulf cooling tower) to check the design strength against a consistent design load, therefore to verify the adequacy of the design algorithm. An iterative numerical computational algorithm is developed for combined membrane and flexural forces, which is based on equilibrium consideration for the limit state of reinforcement and cracked concrete. The design algorithm is implemented in a finite element analysis computer program developed by Mahmoud and Gupta. The amount of reinforcement is then determined at the center of each element by an elastic finite element analysis with the design ultimate load. Based on ultimate nonlinear analyses performed with designed saddle shell, the analytically calculated ultimate load exceeded the design ultimate load from 7% to 34% for analyses with various magnitude of tension stiffening. For the cooling tower problem the calculated ultimate load exceeded the design ultimate load from 26% to 63% with similar types of analyses. Since the effective tension stiffening would vary over the life of the shells due to environmental factors, a degree of uncertainty seems inevitable in calculating the actual failure load by means of numerical analysis. Even though the ultimate loads are strongly dependent on the tensile properties of concrete, the calculated ultimate loads are higher than the design ultimate loads for both design cases. For the cases designed, the design algorithm gives a lower bound on the design ultimate load with respect to the lower bound theorem. This shows the adequacy of the design algorithm developed, at least for the shells studied. The presented design algorithm for the combined membrane and flexural forces can be evolved as a general design method for reinforced concrete plates and shells through further studies involving the performance of multiple designs and the analyses of differing shell configurations.

• Smart Structures and Systems
• /
• 제26권3호
• /
• pp.345-360
• /
• 2020

Modern swarm intelligence heuristic search methods are widely applied in the field of structural health monitoring due to their advantages of excellent global search capacity, loose requirement of initial guess and ease of computational implementation etc. To this end, a hybrid strategy is proposed based on butterfly optimization algorithm (BOA) and differential evolution (DE) with purpose of effective combination of their merits. In the proposed identification strategy, two improvements including mutation and crossover operations of DE, and dynamic adaptive operators are introduced into original BOA to reduce the risk to be trapped in local optimum and increase global search capability. The performance of the proposed algorithm, hybrid butterfly optimization and differential evolution algorithm (HBODEA) is evaluated by two numerical examples of a simply supported beam and a 37-bar truss structure, as well as an experimental test of 8-story shear-type steel frame structure in the laboratory. Compared with BOA and DE, the numerical and experimental results show that the proposed HBODEA is more robust to detect the reduction of stiffness with limited sensors and contaminated measurements. In addition, the effect of search space, two dynamic operators, population size on identification accuracy and efficiency of the proposed identification strategy are further investigated.

• 산업공학
• /
• 제22권1호
• /
• pp.10-16
• /
• 2009

In this paper, we propose an efficient search algorithm for finding an optimal schedule to minimize makespan, while avoiding deadlock situation in Flexible Manufacturing Systems (FMS) with finite capacity, in which each job needs to be processed in several job stages for completion. The proposed algorithm uses a modeling and control method based on Petri-net. Especially, we improve the efficiency of the search algorithm by using a priority rule and an efficient bounding function during the search procedure. The performance of the proposed algorithm is evaluated through a numerical experiment, showing that it holds considerable promise for providing an optimal solution efficiently comparing to past work.