• 한국정보통신학회논문지
• /
• 제12권7호
• /
• pp.1291-1300
• /
• 2008

본 논문에서는 패턴인식의 정확성을 향상시키기 위한 지능시스템을 제안한다. 제안한 지능시스템은 신경회로망의 무감독학습 방법인 SOPM(Self Organizing Feature Map), LVQ(Learning Vector Quantization), 그리고 퍼지이론의 FCM(Fuzzy C-means)을 이용하여 구성한다. 제안한 지능시스템의 유용성은 실험을 통해 확인한다. 실험은 Fisher의 Iris 데이터 분류, Cambridge 대학의 Olivetti 연구실(ORL; Olivetti Research Laboratory)에서 제공하는 얼굴 데이터베이스를 이용한 얼굴 영상 데이터 분류, 그리고 근전도(EMG, Electromyogram) 데이터를 분류하는 것이다. 제안한 지능시스템은 일반적인 LVQ와 비교한다. 실험을 통해 제안한 지능시스템이 일반적인 LVQ보다 패턴 인식의 정확성이 더 우수함을 알 수가 있었다.

• IEIE Transactions on Smart Processing and Computing
• /
• 제6권3호
• /
• pp.183-192
• /
• 2017

k-nearest neighbor (K-NN) is a well-known classification algorithm, being feature space-based on nearest-neighbor training examples in machine learning. However, K-NN, as we know, is a lazy learning method. Therefore, if a K-NN-based system very much depends on a huge amount of history data to achieve an accurate prediction result for a particular task, it gradually faces a processing-time performance-degradation problem. We have noticed that many researchers usually contemplate only classification accuracy. But estimation speed also plays an essential role in real-time prediction systems. To compensate for this weakness, this paper proposes correlation coefficient-based clustering (CCC) aimed at upgrading the performance of K-NN by leveraging processing-time speed and plurality rule-based density (PRD) to improve estimation accuracy. For experiments, we used real datasets (on breast cancer, breast tissue, heart, and the iris) from the University of California, Irvine (UCI) machine learning repository. Moreover, real traffic data collected from Ojana Junction, Route 58, Okinawa, Japan, was also utilized to lay bare the efficiency of this method. By using these datasets, we proved better processing-time performance with the new approach by comparing it with classical K-NN. Besides, via experiments on real-world datasets, we compared the prediction accuracy of our approach with density peaks clustering based on K-NN and principal component analysis (DPC-KNN-PCA).

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 2001년도 춘계학술대회 논문집(한국공작기계학회)
• /
• pp.263-268
• /
• 2001

The needs for precision machining of micro to milli parts have been increased as the industry require high quality products, especially for the micro-machining of IT products. The ultra-precision machining system is essential for the micro machining of fine structures, which insures machining accuracy, low systematic and random error and repeatability. In this study, we developed micro machining system, which is equipped with air bearing stage for ultra precision machining and also we present the results of V-grooving experiments, conducted by the developed system, to verify the performance of system. The results show that the machined V-grooving had good accuracy with repeatable stability.

• Journal of Magnetics
• /
• 제19권2호
• /
• pp.151-154
• /
• 2014

Recent years have seen an increasing range of planar Hall resistive (PHR) sensor applications in the field of magnetic sensing. This study describes a new application of the PHR sensor to monitor a current. Initially, thermal drift experiments of the PHR sensor are performed, to determine the accuracy of the PHR signal output. The results of the thermal drift experiments show that there is no considerable drift in the signals attained from 0.1, 0.5, 1 and 2 mA current. Consequently, the PHR sensor provides adequate accuracy of the signal output, to perform the current monitoring experiments. The performances of the PHR sensor with bilayer and trilayer structures are then tested. The minimum detectable currents of the PHR sensor using bilayer and trilayer structures are $0.51{\mu}A$ and 54 nA, respectively. Therefore, the PHR sensor having trilayer structure is the better choice to detect ultra low current of few tens nanoampere.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2002년도 춘계학술대회 논문집
• /
• pp.947-951
• /
• 2002

During the hotwire cutting of EPS foam sheet, the dimensional accuracy and part quality of the cut par are highly dependent upon the thermal field in the EPS. The thermal field is determined by operating parameters such as heat input, cutting speed and cutting angle. The objective of this study is to investigate into the influence of cutting angle on the kerfwidth and part quality of the cut part in hotwire cutting of EPS foam using the experiments and the numerical analysis in the case of a single sloped cutting. In order to estimate an accurate temperature field, the transient thermal analysis using a moving coordinate system and the sloped heat flux model is carried out. From the results of the experiments and the analysis, it has been found that the effect of cutting angle on the kerfwidth and the melted area at the edge are 0.1 mm and 0.11 m$m^2$ respectively. The results of the experiments show that the surface roughness is not appreciably influenced by the cutting angle.

• Journal of Information Processing Systems
• /
• 제13권2호
• /
• pp.305-320
• /
• 2017

The selection of subset size is of great importance to the accuracy of digital image correlation (DIC). In the traditional DIC, a constant subset size is used for computing the entire image, which overlooks the differences among local speckle patterns of the image. Besides, it is very laborious to find the optimal global subset size of a speckle image. In this paper, a self-adaptive and bidirectional dynamic subset selection (SBDSS) algorithm is proposed to make the subset sizes vary according to their local speckle patterns, which ensures that every subset size is suitable and optimal. The sum of subset intensity variation (${\eta}$) is defined as the assessment criterion to quantify the subset information. Both the threshold and initial guess of subset size in the SBDSS algorithm are self-adaptive to different images. To analyze the performance of the proposed algorithm, both numerical and laboratory experiments were performed. In the numerical experiments, images with different speckle distribution, different deformation and noise were calculated by both the traditional DIC and the proposed algorithm. The results demonstrate that the proposed algorithm achieves higher accuracy than the traditional DIC. Laboratory experiments performed on a substrate also demonstrate that the proposed algorithm is effective in selecting appropriate subset size for each point.

• 대한기계학회논문집B
• /
• 제24권9호
• /
• pp.1195-1209
• /
• 2000

The purpose of this study is to develop the non-gravity fluidized dryer. In this non-gravity fluidized dryer the fluidized zone is produced by two paddles in mixer, which maximizes the surface area of materials and then heated air through the guiding panels dehumidify them. This can conduct the drying process quickly and control moisture contents to lower limits. The ventilation system is closed loop system, which can be changeable to open system, and can be used as a multi-purposed dryer in which mixing, drying, granulating and cooling process is conducted. In order to develop the non-gravity fluidized dryer, in the first fundamental experiments were performed to mixing accuracy and then the other parts of dryer and control system were examined to check whether they were designed properly and operated harmoniously with mixer. Also the preparatory experiments were fulfilled to examine the efficiency and reliability of the dryer. Lastly, on the basis of preparatory experiments, performance test for the non-gravity fluidized dryer carried out for the variation of the initial moisture contents, desired moisture contents, heated air velocity and heating temperature.

• Journal of Electrochemical Science and Technology
• /
• 제15권2호
• /
• pp.299-313
• /
• 2024

Electrochemical machining (ECM) is an effective manufacturing method for difficult-to-machine materials and is widely used in the precision manufacturing of aerospace components. In recent years, the requirements for the machining accuracy and surface integrity of ECM have become increasingly stringent. To further improve the machining quality, this work investigated the intricate laws between electrolyte filtration accuracy and machining quality. Electrolytes with different filtration accuracies were compared, and a numerical simulation was used to evaluate the change in temperature and bubble rate of the flow field in the machining area. Experiments were conducted on ECM of Ti-6Al-4V (TC4) alloy workpieces using electrolytes with different filtration accuracy. The workpiece machining accuracy and surface quality were analyzed, and the repetition accuracy of the workpiece was evaluated. The intricate laws between electrolyte filtration accuracy and machining quality were explored. It was found that when the electrolyte filtration accuracy is improved, so too is the machining quality of the ECM. However, once the filtration accuracy has reached a certain value, the machining quality has extremely limited improvement. By evaluating the repetition accuracy of processed workpieces in electrolytes with different filtration accuracies, it was found that when the filtration accuracy reaches a certain value, there is no positive correlation between the repetition accuracy and filtration accuracy. The result shows that, for the workpiece material and conditions considered in this paper, an electrolyte with 0.5㎛ filtration accuracy is suitable for the wide application of precision ECM.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2000년도 춘계학술대회논문집B
• /
• pp.220-227
• /
• 2000

The purpose of this study is to develop the non-gravity fluidized dryer. In this non-gravity fluidized dryer the fluidized zone is produced by two paddles in mixer, which maximizes the surface area of materials and then heated air through the guiding panels dehumidify them. This can conduct the drying process quickly and control moisture contents to lower limits. The ventilation system is closed loop system, which can be changeable to open system, and can be used as a multi-purposed dryer in which mixing, drying, granulating and cooling process is conducted. In order to develop non-gravity fluidized dryer, in the first the fundamental experiments performed to mixing accuracy and then the other parts of dryer and control system were examined to check whether they were designed properly and operated harmoniously with mixer. Also the preparatory experiments were fulfilled to examine the efficiency and reliability of dryer. Lastly, on the basis of preparatory experiments in case the initial moisture contents, desired moisture contents, heated air velocity and heating temperature were vary, performance test for the non-gravity fluidized dryer carried out.

• 한국산학기술학회논문지
• /
• 제16권3호
• /
• pp.1645-1651
• /
• 2015

본 연구에서는 선삭 작업에서 가공 여유각 변경에 따른 피삭재의 형상정밀도가 어떻게 변화되는지 분석하고자 하였다. 피삭재는 3가지로 SM45C(기계구조용탄소강), SCM415(크롬몰리브덴강), STS303(스테인리스강)을 선택하여 정해진 가공조건인 회전속도 2,500 rpm으로 시작하여 이송속도 0.07 mm/rev와 0.10 mm/rev를 기준으로 가공깊이 0.1 mm, 0.2 mm, 0.3 mm 그리고, 절인 경사각인 네거티브 경사각 $0.0^{\circ}(-6.0^{\circ})$, $0.3^{\circ}(-6.3^{\circ})$, $0.9^{\circ}(-6.9^{\circ})$의 3가지 형태로 가공하였을 경우, 재료별 가공정밀도, 원통도, 동축도, 진원도 등의 결과를 비교 분석하였다. 피삭재의 재질별로 알아본 원통도의 정밀도는 $0.0^{\circ}{\rightarrow}0.3^{\circ}{\rightarrow}0.9^{\circ}$의 순으로 $0.9^{\circ}$일 경우가 가장 좋은 원통도 값을 나타냈다. 결과적으로 바이트의 네거티브 경사각이 커지면 정밀도가 특정 부분에서 재질에 관계없이 좋아져서 가공성이 향상됨을 알 수 있었다. 또한, 이송속도와 절삭 깊이에 따라 가공형상이 변화한다는 것과 재질에 따라 다르게 적용되어야 한다는 것을 확인하게 되었다.