• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.7
• /
• pp.970-976
• /
• 2007

In this study, Polynomial Network Pattern Classifier(PNC) based on Fuzzy Inference Mechanism is designed and its parameters such as learning rate, momentum coefficient and fuzzification coefficient are optimized by means of Particle Swarm Optimization. The proposed PNC employes a partition function created by Fuzzy C-means(FCM) clustering as an activation function in hidden layer and polynomials weights between hidden layer and output layer. Using polynomials weights can help to improve the characteristic of the linear classification of basic neural networks classifier. In the viewpoint of linguistic analysis, the proposed classifier is expressed as a collection of "If-then" fuzzy rules. Namely, architecture of networks is constructed by three functional modules that are condition part, conclusion part and inference part. The condition part relates to the partition function of input space using FCM clustering. In the conclusion part, a polynomial function caries out the presentation of a partitioned local space. Lastly, the output of networks is gotten by fuzzy inference in the inference part. The proposed PNC generates a nonlinear discernment function in the output space and has the better performance of pattern classification as a classifier, because of the characteristic of polynomial based fuzzy inference of PNC.

• Proceedings of the KSME Conference
• /
• 2007.05b
• /
• pp.2603-2608
• /
• 2007

Flows in the centrifugal compressor volute with circular cross section are numerically investigated. The computational grid for the calculation utilized a multi-block arrangement to form a butterfly grid and flow calculations are performed using commercial CFD software, CFX-TASCflow. The centrifugal compressor of this study has axial diffuser after radial diffuser because of the shape of inlet duct and installation constraints. Due to this feature the swirling flow pattern is different from the other investigations. The flow inside volute is very complex and three dimensional with strong vortex and recirculation through volute tongue. The calculation results show circumferential variations of the swirl and through flow velocity and pressure distribution. The mechanism deciding flow structure is explained by considering the force balance in volute cross section. And static pressure recovery and total pressure loss are estimated from the calculated results and compared with Japikse model.

• ALGAE
• /
• v.19 no.3
• /
• pp.277-282
• /
• 2004

Prymnesium patelliferum was collected from Hwajinpo, South Korea in November 2002. This is the first record of P. patelliferum in Korea waters. The characteristic shape of P. patelliferum was that of a rice grain. Cells were sub spherical or elongate, and sometimes slightly compressed. The motile cells of P. patelliferum (8-10 ㎛ long, 5-6 ㎛ broad) prossessed two equal or sub equal flagella(10.5-13 ㎛) and a short non-coiling haptonema (3-5 ㎛). The cell was covered by two layers of organic scales, which were important taxonomic characteristics. Internal scales were a narrow inflexed rim on the distal face, a thickening central and radial fibrillar pattern. External scales were very similar to internal scales with relatively tall upright rims. The ridges were in quadrants with 11-12 ridges per quadrant.

• ETRI Journal
• /
• v.31 no.5
• /
• pp.565-575
• /
• 2009

This paper concerns the use of support vector regression (SVR), which is based on the kernel method for learning from examples, in identification of walking robots. To handle complex dynamics in humanoid robot and realize stable walking, this paper develops and implements two types of reference natural motions for a humanoid, namely, walking trajectories on a flat floor and on an ascending slope. Next, SVR is applied to model stable walking motions by considering these actual motions. Three kinds of kernels, namely, linear, polynomial, and radial basis function (RBF), are considered, and the results from these kernels are compared and evaluated. The results show that the SVR approach works well, and SVR with the RBF kernel function provides the best performance. Plus, it can be effectively applied to model and control a practical biped walking robot.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.48 no.12
• /
• pp.1486-1491
• /
• 1999

This paper presents a method of recognizing high impedance fault(HIF) of electrical power systems and classifying fault patterns based on chaos attractors. Two dimensional chaos attractors are reconstructed from neutral point current waveforms. Reliable features for HIF pattern classification are obtained from the chaos attractors. Radial basis function network, trained with two types of HIF data generated by the electromagnetic transient program and measured form actual faults. The RBFN successfully classifies normal and the three types of fault patterns according to the features generated from the chaos attractors.

• Proceedings of the KIEE Conference
• /
• 2011.07a
• /
• pp.1960-1961
• /
• 2011

얼굴 인식에서 가장 많이 사용되고 있는 PCA(Principal Component Analysis)는 고차원의 얼굴 데이터를 낮은 차원으로 표현할 수 있다는 장점이 있다. LDA(Linear Discriminant Analysis)는 서로 다른 데이터를 잘 분리할 수 있으며, 얼굴 인식에서 우수한 성능을 보인다. 본 연구에서는 서로의 장점을 결합하여 PCA와 LDA를 혼합, 적용하였다. 고차원의 얼굴데이터를 PCA로 차원 축소한 후 LDA를 이용해 더욱 효과적인 분류가 되어 얼굴 인식률을 향상시킨다. 인식 모듈로는 pRBFNN(Polynomial Based Radial Basis Function Neural Networks) 모델을 구축하여 고차원 패턴인식 문제에 대한 해결책을 제시하고자 한다. 그리고 제안된 패턴분류기는 얼굴 데이터를 사용하여 성능을 확인한다.

• Steel and Composite Structures
• /
• v.24 no.3
• /
• pp.359-367
• /
• 2017

In this work, transient heat transfer analysis of functionally graded (FG) carbon nanotube reinforced nanocomposite (CNTRC) cylinders with various essential and natural boundary conditions is investigated by a mesh-free method. The cylinders are subjected to thermal flux, convection environments and constant temperature faces. The material properties of the nanocomposite are estimated by an extended micro mechanical model in volume fraction form. The distribution of carbon nanotube (CNT) has a linear variation along the radial direction of axisymmetric cylinder. In the mesh-free analysis, moving least squares shape functions are used for approximation of temperature field in the weak form of heat transform equation and the transformation method is used for the imposition of essential boundary conditions. Newmark method is applied for solution time depended problem. The effects of CNT distribution pattern and volume fraction, cylinder thickness and boundary conditions are investigated on the transient temperature field of the nanocomposite cylinders.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1987.11a
• /
• pp.39-39
• /
• 1987

자동차에서 발생되는 소음은 여러 가지 발생원으로부터의 복합적인 것으로서 차량의 속도가 고속화하면서 타이어 소음의 기여도가 매우 커지는 것으로 알려져 있다. 타이어 소음은 근본적으로 노면과 타이어의 상호작용에 의하여 발생되는 데 타이어/노면의 상호작용에 영향을 미치는 인자로는 마모상태, 차량속도, 하중, 공기압, carcass 구조, 타이어 온도등 여러 가지가 있으나 tread 모양과 노면의 상태에 가장 크게 영향을 받는다. 본 연구에서는 승용차용 155SR13 radial 타이어를 대상으로 하여 실내에 설치된 dynamometer를 이용하여 groove 의 개수, groove 길이, groove 폭, groove 깊이, groove 방향 등 트레드 패턴 인자가 발생소음에 미치는 기여도를 실험적으로 측정하고 트레드 패턴형상에 따른 소음도를 예측할 수 있는 실험식을 구했다. 또한 단일 groove 내에서 발생되는 소음의 시간신호를 측정하여 모델화하고 차량속도와 groove 사이의 간격에 따른 시간신호를 합성하고 이 신호로부터 FFT 알고리듬을 통해 소음 spectrum을 구하는 소음 예측 프로그램을 개발하였다. 비교적 단순한 tread 패턴에 대해 이를 적용한 결과 실험적으로 구한 spectrum과 상당히 유사함을 볼 수 있었다.

• Journal of the Korean Society of Combustion
• /
• v.7 no.1
• /
• pp.44-51
• /
• 2002

This paper presents an atomization model for sprays under flash boiling conditions. The atomization is represented by the secondary breakup of a bubble/droplet system, and the breakup is considered as the results of two competing mechanisms, aerodynamic force and bubble growth. The model was applied to predict the atomization of a hollow-cone spray from pintle injector under flash boiling conditions. In the regimes this study considered, sprays are atomized by bubble growth, which produces smaller SMD#s than aerodynamic forces alone. With decreasing ambient pressures, the spray thickness, fuel vaporization rate and vapor radial penetration increases, and the drop size decreases. With increasing the fuel and ambient temperatures to some extent, the effect of flash boiling and air entrainment completely change the spray pattern.

• Transactions of Materials Processing
• /
• v.10 no.6
• /
• pp.465-470
• /
• 2001

Hole flanging experiments are performed on flat circular plates with a hole in the center and the flangeability and fracture behaviors of TRIP steels and ferrite-Bainite duplex steels were examined. In the hole flanging, deformation by lip and petalling occurs when plates are struck by punches of various shapes and high circumferential strains induced in the target material cause radial cracking and the subsequent rotation of the affected plate material in a number of symmetric petals. In all cases, failure of the plate was due to lip fracture that results from multiple localized neckings that take place around the hole periphery where straining is most severe and a somewhat regular pattern was observed in a fracture shape. The neck characteristics in flange formation and the transition from the lip to petal mode at which fracture occurs were compared with two materials.