• Proceedings of the Korean Nuclear Society Conference
• /
• 2004.05a
• /
• pp.180.2-180.2
• /
• 2004

• Journal of Electrical Engineering and Technology
• /
• v.12 no.2
• /
• pp.911-917
• /
• 2017

There are generally three folds when developing neural network classifiers. They are as follows: 1) discriminant function; 2) lots of parameters in the design of classifier; and 3) high dimensional training data. Along with this viewpoint, we propose space search optimized polynomial neural network classifier (PNNC) with the aid of data preprocessing technique and simultaneous tuning strategy, which is a balance optimization strategy used in the design of PNNC when running space search optimization. Unlike the conventional probabilistic neural network classifier, the proposed neural network classifier adopts two type of polynomials for developing discriminant functions. The overall optimization of PNNC is realized with the aid of so-called structure optimization and parameter optimization with the use of simultaneous tuning strategy. Space search optimization algorithm is considered as a optimize vehicle to help the implement both structure and parameter optimization in the construction of PNNC. Furthermore, principal component analysis and linear discriminate analysis are selected as the data preprocessing techniques for PNNC. Experimental results show that the proposed neural network classifier obtains better performance in comparison with some other well-known classifiers in terms of accuracy classification rate.

• Journal of the Society of Naval Architects of Korea
• /
• v.55 no.4
• /
• pp.289-296
• /
• 2018

The offshore subsea platforms are connected to subsea pipelines to transport gas/oil from wells. The pipe is a multilayered structure of polymer and steel for compensating both flexibility and strength. The pipe also requires reinforcement structures to endure the extreme environmental conditions. A vertebrae structure of bend restrictors is one of the reinforcement structures installed to protect the subsea pipe from excessive bending deformations. In this study, structural behaviors of the subsea pipeline with bend restrictors are investigated by the multi-body contact analysis in Abaqus 6.14-2. Contact forces of each bend restrictor extracted from the multi-body contact analysis can be boundary conditions for topology design optimization in Altair Hyperworks 13.0 Hypermesh-Optistruct. Multiple design constraints are considered to obtain a manufacturable design with efficient material usage. Through the multi-body contact analysis with optimized bend restrictors, it is confirmed that the bending performance of the optimized design is enhanced.

• Bulletin of the Korean Space Science Society
• /
• 2008.10a
• /
• pp.28.2-28.2
• /
• 2008

Recently, star sensors have been successfully used as main attitude sensors for attitude control in many satellites. This research presents the star visibility analysis for star trackers and the goal of this analysis is to make sure that the star tracker implementation is suitable to the mission profile and scenario and satisfies the requirement of attitude orbit control system. As a main optical attitude sensor imaging stars, accomodations of a star tracker should be optimized in order to improve the probability of the usage by avoiding the blinding (the unavailability) by the Sun and the Earth. For the analysis, a statistical approach and a time simulation approach are used. The statistical approach is based on the generation of numerous cases, to derive relevant statistics about Earth and Sun proximity probabilites for different lines of sight. The time simulation approach is performed for one orbit to check the statistical result and to refine the statistical result and accomodations of star trackers. In order to perform simulations first of all, an orbit and specific mission profiles of a satellite are set, next the earth proximity probability and the sun proximity probability are calculated by considering the attitude maneuvers and the geometry of the orbit, and then finally the unavailability positions are estimated. As a result, the optimized accomodations of two star trackers are suggested for the low earth orbit satellite.

• Transactions of Materials Processing
• /
• v.27 no.2
• /
• pp.81-86
• /
• 2018

In the cold forming process, it is not easy to fabricate a asymmetric type nut, due to the difficulty in the exact prediction of metal-flow. As we have identified, in that case, it often results in the various forging defects such as burrs, and an incomplete shape, as well as other problems because of this issue. In the current study, we introduce the development of an unparalleled shape Weld Nut by using a forging analysis tool (AFDEX). For the multi-forming machine, the optimized shapes of each intermediate product (step product) could be found by the use of a model for the prediction and analysis of various types, sizes and heights. Chiefly, forging tools were prepared based on the simulation results and an unparalleled shape could be prepared at one time without any burrs, incomplete shape and size.

• Wind and Structures
• /
• v.18 no.3
• /
• pp.253-265
• /
• 2014

The platform and floating structure of spar type offshore wind turbine systems should be designed in order for the 6-DOF motions to be minimized, considering diverse loading environments such as the ocean wave, wind, and current conditions. The objective of this study is to optimally design the platform and substructure of a 3MW spar type wind turbine system with the maximum postural stability in 6-DOF motions as well as the minimum material cost. Therefore, design variables of the platform and substructure were first determined and then optimized by a hydrodynamic analysis. For the hydrodynamic analysis, the body weight of the system was considered, and the ocean wave conditions were quantified to the wave forces using the Morison's equation. Moreover, the minimal number of computation analysis models was generated by the Design of Experiments (DOE), and the design variables of the platform and substructure were finally optimized by using a genetic algorithm with a neural network approximation.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2012.04a
• /
• pp.361-366
• /
• 2012

In this paper, we propose the efficient technique that reduces the number of spot-welds and increases the structural rigidity by using the topology optimization technique. Eigen value analysis is used to evaluate the rigidity of the optimized model. As a first step, the topology optimization is performed to find optimal spot-weld distributions. In this study, the design objective is to maximize the weighted frequencies. The volume fractions of the weld components are used as design constraints, and also the densities of each element in the individual design space are used as design variables. And then, to consider the possibility of spot-weld failure, the contribution rate analysis was performed by using the orthogonal array method of DOE. The spot-welds in the rear panel part are reinforced according to estimation results of the contribution rate analysis. Finally, we obtained optimized spot-weld layout model which has the reduced number of spot-welds and the improved dynamic stiffness.

• Proceedings of the KSME Conference
• /
• 2008.11a
• /
• pp.544-547
• /
• 2008

As the mobile electronic product is getting slimmer and smaller, the necessity of keypad is being increased. But the possibility of mis-typing keypad is increased rapidly due to the integrated keypad in the small mobile product. The business division has not considered the methodology of keypad design essentially. In this paper, analysis method and design evaluation standard to reduce the mis-typing of UMPC(Ultra Mobile Personal Computer) is suggested. First, the finite element analysis model and the biomechanical human body model are implemented in order to simulate the exact contact characteristic between finger and keypad. The reliability of analysis model is guaranteed by the comparison of the contact pressure between analysis result and experiment result of the pressure sensor. The design optimization of key shape and layout is derived through the response surface method. The prototype model is produced with the optimized design of keypad, and then it verified the advanced function with user mis-typing detection test. The optimized keypad design reduced the mis-typing ratio from 35% of existing model to 75 of proposed model. If this paper is widely applied to not only UMPC but also the other electronic products, the emotional quality of all products could be improved considerably.

• The KSFM Journal of Fluid Machinery
• /
• v.15 no.5
• /
• pp.37-41
• /
• 2012

High-efficiency and low-noise axial flow fan is developed by combining the FanDAS, a computerized axial fan design/performance analysis system, and CFD software(CFX). Based on fan design requirements, FanDAS conducts 3-D blade geometry design, quasi-3D flow/ performance analyses and noise evaluation by using through-flow analysis method and noise models for discrete frequency and broadband noise sources. Through the parametric studies of fan design variables using FandDAS, preliminary and baseline design is achieved for high efficiency and low noise fan, and then can be coupled with a CFD technique such as the CFX code for constructing final and optimized fan design. The FanDAS-CFX coupled system and its design procedure are applied to actual fan development practice. The FanDAS provides an optimized 3-D fan blade geometry, and its predictions on the performance and the noise level of designed fan are well agreed with actual test results.

• Journal of Environmental Science International
• /
• v.4 no.1
• /
• pp.91-103
• /
• 1995

The reaction rate, equilibrium, and flow injection analysis methods were fundamentally evaluated for the determination of aqueous ammonia. The selected indophenol blue method was based on the formation of indophenol blue in which ammonium ion reacted with hypochlorite and phenol in alkaline solution. In the optimized reaction condition, the reaction followed 1st order reaction kinetics and the final product was stable. The absorbance measurements before and after the equilibrium were utilized for the reaction rate and equilibrium methods. The reaction rate methods, based on the relative analytical signals for the possibility of eliminating interferents, were shown to have good linear calibration curves but the detection limit and the calibration sensitivity were poorer than those in the equilibrium method. The detection limits were 32-49 pub and 24 pub for the reaction rate and equilibrium methods, respectively In the flow injection analysis, the absorbance was measured before the equilibrium reached and thus resulted in 30% reduction of calibration sensitivity. However, the detection limit was 11 ppb, indicating that the peak-to-peak noise for the blank was remarkably improved. Compared to the manual methods, the optimized experimental condition in a closed reaction system reduced the blank absorbance and the inclusion of ammonia from the atmosphere was prevented. In addition, highly reproducible mixing of sample and reagents and analytical data extracted from continuous recording showed excellent reproducibility.