• Proceedings of the KIEE Conference
• /
• 1998.07a
• /
• pp.179-182
• /
• 1998

This paper is proposed a selection method of the major design dimension which constrain the maximum acceleration capability and minimum power loss of surface-mounted brushless do motor with NdFeB permanent magnet for servo drives. Expressions are derived from the air-gap flux density and the linear current density around the stator periphery and design dimensions. The linear current density is limited by the need to avoid demagnetization. In this paper, We compute the optimum design dimensions of 2KW BLDC motor with maximum acceleration capability and minimum power loss by using genetic algorithm.

• Proceedings of the KIEE Conference
• /
• 2000.07b
• /
• pp.1177-1181
• /
• 2000

Usually, in many applications. high frequency resonant inverters are used, and the PAM(Pulse Amplitude Modulation), PFM(Pulse Frequency Modulation) or PWM(Pulse Width Modulation) techniques are used to control the output power of resonant inverters. And the resonant inverters have to control the output frequency for the reliable operation under the variable load conditions. In this paper, a new switching scheme is proposed as a resonant frequency following control of resonant inverters. With the proposed method. it can be obtained that optimum resonant frequency and unity output displacement factor under the variable resonant frequency adaptively. The detail algorithm of the proposed switching scheme and its characteristics are discussed. And the veridity of the proposed method is confirmed with the experimental results.

• Proceedings of the KIEE Conference
• /
• 1995.07b
• /
• pp.1006-1010
• /
• 1995

This paper describes Sexual Reproduction Genetic Algorithm(SRGA) for function optimization. In SRGA, each individual utilize a diploid chromosome structure. Sex cells(gametes) are produced through artificial meiosis in which crossover and mutation occur. The proposed method has two selection operators, one, individual selection which selects the individual to fertilize, and the other, gamete selection which makes zygote for offspring production. We consider the effects of multi-selection and diploidy on search performance. SRGA improves local and global search(exploitation and exploration) and show optimum tracking performance in nonstationary environments. Gray coding is incorporated to transforming the search space and Genic uniform distribution method is proposed to alleviate the problem of premature convergence.

• Proceedings of the KIEE Conference
• /
• 1995.07b
• /
• pp.777-780
• /
• 1995

Accounting for actuator nonlinearities in control loops has often been perceived as an implementation issue and usually excluded in the design of controllers. Nonlinearities treated in this paper are saturation, and they are modelled as an inequality constraint. The CRHPC(Constrained Receding Horizon Predictive Control) with inequality constraints algorithm is used to handle actuator rate and amplitude limits simultaneously or respectively. Optimum values of future control signals are obtained by quadratic programming. Simulated examples show that predictive control law with inequality constraints offers good performance as compared with input clipping.

• Journal of computational fluids engineering
• /
• v.14 no.4
• /
• pp.31-40
• /
• 2009

Although optimization by sequentially refining metamodels is known to be computationally very efficient, the metamodel that can be used for this purpose is limited to Kriging method due to the difficulties related with sample points selections. The present study suggests a novel method for sequentially refining metamodels using Pareto Frontiers, which can be used independent of the type of metamodels. It is shown from the examples that the present method yields more accurate metamodels compared with full-factorial optimization and also guarantees global optimum irrespective of the initial conditions. Finally, in order to prove the generality of the present method, it is applied to a 2D transonic airfoil optimization problem, and the successful design results are obtained.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.3 no.2
• /
• pp.93-103
• /
• 1997

This paper develops a new heuristic, the Excursion Algorithm(EA), for constructing optimal designs for the experiments related with marine environment. The proposed EA consists of three parts: 1) construction of an initial feasible solution, 2) excursions over a bounded region, and 3) stopping rules. It is the second part that distinguishes the EA from the other existing heuristic methods. It turns out that excursions over a bounded feasible and/or infeasible region is effective in alleviating the risks of being trapped at a local optimum. Since this problem is formulated for the first time thesis, other heuristic algorithms do not exist. Therefore, global optimal solutions are obtained by complete enumeration for some cases, and the performance of the EA is evaluated in terms of solution quality. Computational results show that the proposed EA is effective in finding good(or, in many cases, global) solutions to the constrained optimal experimental design problems.

• Wind and Structures
• /
• v.18 no.1
• /
• pp.21-35
• /
• 2014

The latest developments in topology optimization are integrated with Computational Fluid Dynamics (CFD) for the conceptual design of building structures. The wind load on a building is simulated using CFD, and the structural response of the building is obtained from finite element analysis under the wind load obtained. Multiple wind directions are simulated within a single fluid domain by simply expanding the simulation domain. The bi-directional evolutionary structural optimization (BESO) algorithm with a scheme of material interpolation is extended for an automatic building topology optimization considering multiple wind loading cases. The proposed approach is demonstrated by a series of examples of optimum topology design of perimeter bracing systems of high-rise building structures.

• Journal of Mechanical Science and Technology
• /
• v.15 no.11
• /
• pp.1580-1587
• /
• 2001

In Part I, we extended the capability of the Goore Scheme for application to multi-dimensional problems and improved convergence performance. In this paper, we apply the improved Goore Scheme to th e control of turbulence for drag reduction. Direct numerical simulations combined with the control scheme are carried out to simulate a controlled turbulent channel flow at low Reynolds number. The wall blowing and suction is applied through the Goore algorithm using the total drag as feedback. An optimum distribution of the wall blowing and suction in terms of the wall-shear stresses in the spanwise and streamwise directions is sought. The best case reduces drag by more than 20 %.

• Journal of Advanced Marine Engineering and Technology
• /
• v.19 no.1
• /
• pp.54-59
• /
• 1995

In this paper, we propose unique CAM system for personal computer that can define the geometric shape in an ease manner and to machine the sculptured surfaces of a mold cavity. In this CAM system, if a user inputs simple initial information such as the control points for a shape definition and a radius of tool etc., all of the procedures for machining will be processed automatically by the CAM system as well as NC commands and simulations. In addition to this, the environment of the CAM system is composed of "C" language for an easy extention of aditional modules. Also, the CAM system with the following characteristics was developed. 1. The optimum tool path satisfying given tolerance limits reduces the time for the high precision machining of sculptured surface in a mold cavity. 2. The generated NC commands can be transmitted to NC directly by the CAM system through RS-232C from PC.C from PC.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.31 no.6 s.261
• /
• pp.651-658
• /
• 2007

A new tree-structured neural network classifier is proposed for the automatic real-time inspection of cold-rolled steel strip surface defects. The defects are classified into 3 groups such as area type, disk type, area & line type in the first stage of the tree-structured neural network. The defects are classified in more detail into 11 major defect types which are considered as serious defects in the second stage of neural network. The tree-structured neural network classifier consists of 4 different neural networks and optimum features are selected for each neural network classifier by using SFFS algorithm and correlation test. The developed classifier demonstrates very plausible result which is compatible with commercial products having high world-wide market shares.