• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.5 no.3
• /
• pp.137-144
• /
• 1985

This paper deals with the adjoint variable method in design sensitivity analysis that is essential to the structure optimization. The method is shown to be much simpler than the conventional method in structure optimization by applying it to the optimal design of finite dimensional structures. Design sensitivity analyses and their numerical solutions for the principal constraints, i.e., displacement and stress constraints under static loads are obtained. Furthermore, it is proved that optimization can be carried out efficiently by applying the optimization algorithm. Structure optimization problems of minimizing the volumes of the truss structures(finite dimensional structures) under the appropriate boundary conditions, loading conditions and constraints are considered.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2007.04a
• /
• pp.321-326
• /
• 2007

The joined-wing is a new concept of the airplane wing. The fore-wing and the aft-wing arc joined together in the joined-wing. The range and loiter are longer than those of a conventional wing. The joined-wing can lead to increased aerodynamic performances and reduction of the structural weight. The structural behavior of the joined-wing has a high geometric nonlinearity according to the external loads. The gust loads are the most critical loading conditions in the structural design of the joined-wing. The nonlinear behavior should be considered in the optimization of the joined-wing. It is well known that conventional nonlinear response optimization is extremely expensive: therefore, the conventional method is almost impossible to use in large scale structures such as the joined-wing. In this research, geometric nonlinear response structural optimization is carried out using equivalent loads. Equivalent loads are the load sets which generate the same response field in linear analysis as that from nonlinear analysis. In the equivalent loads method, the external loads are transformed to the equivalent loads (EL) for linear static analysis, and linear response optimization is carried out based on the EL.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.14 no.6
• /
• pp.122-128
• /
• 2015

Recently, as the demand for high efficiency, multi-function machine tools has increased, domestic machine tool industries are investing in research and development for Gantry-Type Machining centers. In this thesis, for the purpose of evaluating machining accuracy and designing a machine tool structure, a simplified model of the main frame is suggested. The results show the general characteristics of the optimum design, and the approach is shown as practicable for the preliminary design analysis and improvement of a conceptual design of a Gantry-Type Machining center. This paper's results are expected to improve the static characteristics of Gantry-Type Machine centers. The three-dimensional finite element models proved that the modeling method might be applied to real machine tool structures.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2000.06a
• /
• pp.109-115
• /
• 2000

The spindle unit is core parts in high precision machine tools. Diverse static, dynamic and thermal charateristics of spindle unit are needed for special purpose of machine tools. Compromise between those charateristics will be done in concept design phase. High static stiffness at spindle nose may be very important performance for heavy cutting work. High dynamic stiffness is also useful to high precision and high speed machine tools. Improvement of thermal charateristics in spindle lead to high reliability of positioning accuracy. For high speed spindle structure, the design parameter such as, bearing span, diameter, bearing type and arrangement, preload, cooling and lubrication method should be in harmony.

• Journal of Electrical Engineering and Technology
• /
• v.4 no.2
• /
• pp.175-184
• /
• 2009

In this paper, a method for optimal measurement placement in the problem of static harmonic state estimation in power systems is proposed. At first, for achieving to a suitable method by considering the precision factor of the estimation, a procedure based on Genetic Algorithm (GA) for optimal placement is suggested. Optimal placement by regarding the precision factor has an evident solution, and the proposed method is successful in achieving the mentioned solution. But, the previous applied method, which is called the Sequential Elimination (SE) algorithm, can not achieve to the evident solution of the mentioned problem. Finally, considering both precision and economic factors together in solving the optimal placement problem, a practical method based on GA is proposed. The simulation results are shown an improvement in the precision of the estimation by using the proposed method.

• Proceedings of the KIEE Conference
• /
• 2005.07d
• /
• pp.2564-2566
• /
• 2005

This paper presents an iterative linear matrix inequality (LMI) method for $H_2$ and $H_{\infty}$ optimal static output feedback (SOF) control, which is expressed in terms of LMIs subject to an additional rank condition. We propose a linear Penalty function to penalize the rank constraint so that static $H_2$ and $H_{\infty}$ synthesis results in solving a series of convex LMI optimization problems. Numerical experiments for various $H_2$ and $H_{\infty}$ SOF synthesis were performed to demonstrate the effectiveness of the proposed algorithm.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 1999.10a
• /
• pp.135-142
• /
• 1999

Plate girder bridges for tile Korean high-speed railway are optimally designed. Static and dynamic constraints are all considered. The design variables are the thicknesses and the lengths of the plates that are used to form I-shaped main girders with variable cross-sections. And the objective function is tile steel weight of a main girder. A C++ based design program is developed; this program interfaces with a FORTRAN based optimization program ADS. From the results of optimal design for various span lengths, it is observed that the deck vertical acceleration is one of the most important constraints in a special range of tile span length. Front a parametric study, sensitivity of the optimal design to static as well as dynamic constraints are presented.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.32 no.5
• /
• pp.392-399
• /
• 2008

Accurate measurement of speed and altitude of flying vehicles in air data system remains a critical technical issue. A highly reliable Pitot-static probe is required to obtain air data such as total pressure and static pressure. In this study, an analysis of the characteristics of flowfield around the Pitot-static probe was performed by using a Navier-Stokes CFD code. In addition, for the purpose of finding an optimal configuration, a technique based on the response surface method is applied to the problem with design parameters including shape of the nose section and cone angle. It is shown that the optimal configuration fulfills the MIL specification in wider range of high angles of attack.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.44 no.1
• /
• pp.21-24
• /
• 1995

The larger and the more complicated the system size and configuration grow, the more serious the system loss problem becomes. Exessive system loss causes severs system voltage depression, which even may result in system voltage collapse. This paper proposes an effective tool for minimizing the system power loss by optimal re-location of the static condenser based on the system loss sensitivity index .lambda.$_{Q}$. It is possible to determine the optimal location and amount of VAR investment for minimizing the system loss by priority of .lambda.$_{Q}$ index given for each bus. Several computational techniques for avoiding divergency of the load flow solution are proposed. The loss sensitivity index .lambda.$_{Q}$ uses information of normal power flow equations and their Jacobians. Two case studies proved the effectiveness of the algorithm proposed.posed.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.43 no.5
• /
• pp.753-760
• /
• 1994

This paper proposes a new method to compute the static thrust of a hybrid-type LPM by the coenergy method, which can be applied to a nonlinear programming of the optimization techniques for the optimal design. In the process of the computation, the magnetic nonlinearities of the permanent magnet, the primary and the secondary core are interpolated by the cubic spline method. And then, the equivalent magnetic circuit including the airgap reluctance which is a function of the displacement is modelled by the permeance method. The static thrust which is a derivative of the coenergy is computed by the Newton Raphson method at each displacement. To verify this proposed method, the results of the compjutation are compaction with those of the experiment obtained from the DLPM with 2 phase and 4 poles.