• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.25 no.6
• /
• pp.36-41
• /
• 2002

We consider the Fuzzy clustering which is devised for partitioning a set of objects into a certain number of groups by assigning the membership probabilities to each object. The researches carried out in this field before show that the Fuzzy clustering concept is involved so much that for a certain set of data, the main purpose of the clustering cannot be attained as desired. Thus we propose a new objective function, named as Fuzzy-Entroppy Function in order to satisfy the main motivation of the clustering which is classifying the data clearly. Also we suggest Mean Field Annealing Algorithm as an optimization algorithm rather than the ISODATA used traditionally in this field since the objective function is changed. we show the Mean Field Annealing Algorithm works pretty well not only for the new objective function but also for the classical Fuzzy objective function by indicating that the local minimum problem resulted from the ISODATA can be improved.

• Management & Information Systems Review
• /
• v.14
• /
• pp.51-65
• /
• 2004

We consider the Fuzzy clustering which is devised for partitioning a set of objects into a certain number of groups by assigning the membership probabilities to each object. The researches carried out in this field before show that the Fuzzy clustering concept is involved so much that for a certain set of data, the main purpose of the clustering cannot be attained as desired. Thus we Propose a new objective function, named as Fuzzy-Entroppy Function in order to satisfy the main motivation of the clustering which is classifying the data clearly. Also we suggest Mean Field Annealing Algorithm as an optimization algorithm rather than the ISODATA used traditionally in this field since the objective function is changed. We show the Mean Field Annealing Algorithm works pretty well not only for the new objective function but also for the classical Fuzzy objective function by indicating that the local minimum problem resulted from the ISODATA can be improved.

• Applied Microscopy
• /
• v.47 no.1
• /
• pp.29-35
• /
• 2017

Transverse magnetic field annealing (TFA) was carried out on $Fe_{73.5}Cu_1Nb_3Si_{15.5}B_7$ nano-crystalline magnetic core with the aim at decreasing coercivity ($H_c$) while keeping high inductance ($L_s$). The magnetic field generated by direct current (DC) was applied on the magnetic core during different selected annealing stages and it was proved that the nanocrystalline magnetic core achieved lowest $H_c$ when applying transverse field during the whole annealing process (TFA1). Although the microstructure and crystallization degree of the nanocrystalline magnetic core exhibited no obvious difference after TFA1 compared to no field annealing, the TFA1 sample showed a more uniform nanostructure with a smaller mean square deviation of grain size distribution. $H_c$ of the nanocrystalline magnetic core annealed under TFA1 decreased along with the increasing magnetic field. As a result, the certain size nanocrystalline magnetic core with low $H_c$ of 0.6 A/m, low core loss (W at 20 kHz) of 1.6 W/kg under flux density of 0.2 T and high $L_s$ of $13.8{\mu}H$ were obtained after TFA1 with the DC intensity of 140 A. The combination of high $L_s$ with excellent magnetic properties promised this nanocrystalline alloy an outstanding economical application in high frequency transformers.

• Journal of the Korea Convergence Society
• /
• v.8 no.11
• /
• pp.103-108
• /
• 2017

In the ubiquitous environment, there has been a problem of transmitting data from various sensors at a long distance. Especially, in the process of integrating data arriving at different locations, data having different property values of data or having some loss in data had to be processed. This paper present a method to analyze such data. The core of this method is to define an objective function suitable for the problem and to develop an algorithm that can optimize this objective function. The objective function is used by modifying the OCS function. MFA (Mean Field Annealing), which was able to process only binary data, is extended to be applicable to fields with continuous values. It is called CMFA and used as an optimization algorithm.

• Journal of Digital Contents Society
• /
• v.19 no.2
• /
• pp.343-349
• /
• 2018

The proposed hybrid algorithm combines the benefits of rapid convergence property of mean filed annealing(MFA) and the effective genetic operations of simulated annealing-like genetic algorithm(SGA). This algorithm is applied to the isotropic material stock cutting problem, especially to glass cutting in distributed computing environments base on MPI called message passing interface. The glass cutting is to place the required rectangular patterns to the given large glass sheets resulting in reducing the wasted scrap area. Our experimental results show that the heuristic method improves the performance over the conventional ones by decreasing the scrap area and maximum execution time. It is also proved that the proposed distributed algorithm maintains the convergence properties of sequential one while it achieves almost linear speedup as the problem size increases.

• Communications for Statistical Applications and Methods
• /
• v.3 no.1
• /
• pp.219-227
• /
• 1996

역추정 문제에서 독립변수가 하나 더 추가되었을 때 나타나는 효과가 각 변수간의 상관 계수들 및 이들의 함수로 계산되는 결정계수에 따라 어떻게 달라지는지를 살펴보았다. 연구방법으로는 베이지안 방법을 택하였고, 계산상 어려움을 극복하기 위해 Gibbs sampling 방법 및 Mean Field Annealing 방법을 도입하였다.

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.3
• /
• pp.417-424
• /
• 1995

This paper describes MFT(Mean Field Theory) neural network with continuous with continuous variables is applied to quantification analysis problem. A quantification analysis problem, one of the important problems in statistics, is NP complete and arises in the optimal location of objects in the design space according to the given similarities only. This paper presents a MFT neural network with continuous variables for the quantification problem. Starting with reformulation of the quantification problem to the penalty problem, this paper propose a "one-variable stochastic simulated annealing(one-variable SSA)" based on the mean field approximation. This makes it possible to evaluate of the spin average faster than real value calculating in the MFT neural network with continuous variables. Consequently, some experimental results show the feasibility of this approach to overcome the difficulties to evaluate the spin average value expressed by the integral in such models.ch models.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10a
• /
• pp.364-366
• /
• 2003

본 논문에서는 병렬처리에서 중요한 부하균형 문제에 대한 새로운 솔루션을 소개한다. 제안하는 매핑 알고리즘은 평균장 어닐링과 유전자 알고리즘을 합성한 휴리스틱 부하균형 기법이다. 합성된 알고리즘을 세 개의 다른 알고리즘들과의 성능향상비를 측정하는 성능평가 시뮬레이션을 개발하였고 솔루션 품질과 수행시간 면에서 우지의 방법은 기존의 것들 보다 개선된 실험결과를 얻었다.

• Journal of Magnetics
• /
• v.7 no.3
• /
• pp.72-79
• /
• 2002

Local M-H loops have been measured on the free layer of a tunneling magnetoresistance (TMR) junction using the magneto-optical Kerr effect (MOKE) system, with an optical beam size of about 2 $\mu$m diameter. Tunnel junctions were deposited using the DC magnetron sputtering method in a chamber with a base pressure of 3$\times$10$^{-9}$ Torr. The relatively irregular variations of coercive force H$_c$(∼17.5 Oe) and unidirectional anisotropy field H$_{ua}$(∼7.5 Oe) in the as-deposited sample are revealed. After $200{^{\circ}C}$ annealing, He decreases to 15 Oe but H$_{ua}$ increases to 20 Oe with smooth local variations. Two-dimensional plots of H$_c$ and H$_{ua}$ show the symmetric saddle shapes with their axes aligned with the pinned layer, irrespective of the annealing field angle. This is thought to be caused by geometric effects during deposition, together with a minor annealing effect. In addition, the variation of root mean square (RMS) surface roughness reveals it to be symmetric with respect to the center of the pinned-layer axis, with the roughness of 2.5 $\AA$ near the edge and 5.8 $\AA$ at the junction center. Comparison of surface roughness with the variation of H$_{ua}$ suggests that the H$_{ua}$ variation of the free layer is well described by dipole interactions related to surface roughness. As a whole, the reversal magnetization is not uniform over the entire junction area and the macroscopic properties are governed by the average sum of local distributions.

• Journal of Mechanical Science and Technology
• /
• v.20 no.8
• /
• pp.1169-1182
• /
• 2006

Robust design technology has been applied to versatile engineering problems to ensure consistency in product performance. Since 1980s, the concept of robust design has been introduced to numerical optimization field, which is called the robust optimization. The robustness in the robust optimization is determined by a measure of insensitiveness with respect to the variation of a response. However, there are significant difficulties associated with the calculation of variations represented as its mean and variance. To overcome the current limitation, this research presents an implementation of the approximate statistical moment method based on kriging metamodel. Two sampling methods are simultaneously utilized to obtain the sequential surrogate model of a response. The statistics such as mean and variance are obtained based on the reliable kriging model and the second-order statistical approximation method. Then, the simulated annealing algorithm of global optimization methods is adopted to find the global robust optimum. The mathematical problem and the two-bar design problem are investigated to show the validity of the proposed method.