2행정 디젤기관의 연소과정에 미치는 소기압력의 영향

한종호,최익수,방중철 國立金烏工科大學校 産業技術開發硏究院 2002 産業技術開發硏究 Vol.18 No.-

The process of complete combustion of diesel engine is influenced by many factors. Several parameters such as flow system, injection system, shape of combustion chamber and fuel property affect the complete combustion state of diesel engine. Especially, the fuel-air mixture of diesel engine combustion has predominant effects on engine performance and pollutant emission. For the experiment, three kinds of the combustion chamber were designed and manufactured to consider the cavity shape and the air flow. In addition we changed four-cycles diesel engine into two-cycles visualization of diesel engine. And then we analyzed the effect of shape of combustion chamber type, scavenging pressure(106.4kPa, 111.5kPa) and scavenging temperature(2p℃) on combustion flame in order to find optimal combustion chamber condition for better combustion and better performance. As a result the combustion process of the cylindrical type in the state of scavenging pressure of 111.5kPa and scavenging temperature of 20℃ showed the best effects.

디지털 경북 농촌건설을 위한 정보화 추진전략

신호균 國立金烏工科大學校 産業技術開發硏究院 2002 産業技術開發硏究 Vol.18 No.-

This paper reviews rural informatization strategies for building digital Gyeongbuk by informatization of rural administration, society, industry, and life sectors. As activating policy is transformed into economizing policy for successful implementation of these strategies by utilizing rural resources, the concepts such as agri-business, precision farming, and knowledge management farming are emerging as major competitive forces for building digital Gyeongbuk in agricultural area. Also, it is necessary to practically support the linkage of rural and urban policy for reducing digital divide and encouraging diffusion of informatization mind

SOFM 알고리듬의 개선에 관한 연구

최광영,김상희,박원우 國立金烏工科大學校 産業技術開發硏究院 2002 産業技術開發硏究 Vol.18 No.-

The SOFM(Self Organize Feature Map) is used many fields on pattern recognition and image compression because of it's simple structure. But it's learning process is complex and takes a long time for searching whole weights. This paper propose an improved SOFM algorithm to decrease learning period and to increase PSNR. The proposed SOFM has two processes : appling hierarchical structure, and learning to set sub-level structure and initializing sub-level weights. The improved SOFM algorism is applied to image compression and restoration.

면방전 AC PDP에서 콘트라스트 개선을 위한 구동 파형과 회로설계

안양기,윤동한,김태형 國立金烏工科大學校 産業技術開發硏究院 2002 産業技術開發硏究 Vol.18 No.-

This paper proposes a method to drive an AC plasma display panel(PDP) with a significantly improved contrast ratio. In the proposed method, during the first sub-field of one frame, all PDP cells are reset by the ramp waveform, and during the other sub-fields, only the cells turned on in the previous sub-field are reset. No light is emitted during the reset period of every sub-field except the first sub-field. For a 10-bit picture, the luminance of the dark level for the proposed method is 10 times lower than that for the conventional method, in which the ramp waveform for the reset is used in every sub-field. Accordingly, the contrast ratio for the proposed method is 10 times higher than that for the conventional method. For the 10-bit picture, the measured contrast ratio was about 3080:1 for the proposed method and about 185:1 for the conventional method, resulting in 10.8 times increase in the contrast ratio. This result shows that the proposed method can realize an image with high contrast ratio.

복합 지능 제어기를 이용한 3축 푸마 로봇의 경로제어

박창범,남영재,채창현 國立金烏工科大學校 産業技術開發硏究院 2002 産業技術開發硏究 Vol.18 No.-

This paper describes the Hybrid Intelligent controller using Modular Recurrent Neural Network(MRNN) controller and Simplified Indirect Inference Method(SIIM) Fuzzy PD controller, which is used to control a three axis PUMA robot. First, the SIIM Fuzzy PD controller is derived from the conventional continuous time linear PD controller. Then the fuzzyfication, control-rule base, and defuzzification using SIIM in the design of the fuzzy controller are discussed in detail. The proposed controller enhances the self-tuning control capability. Second, the modular recurrent neural network(MRNN) controller has additional feedback neurons in the conventional multi-layer neural network. the learning law based on gradient search techniques has been used. The simulation results of a three-axis PUMA robot shows that comparing to the MRNN controller plus fixed PD controller, the proposed method gives the better precision on the tracking of the given circular trajectory and the better approximation for three-axis respectively.

플레이트 거더사교의 둔각부에서 전단력 산정을 위한 윤하중분배계수

강창조,송재호,장일영 國立金烏工科大學校 産業技術開發硏究院 2002 産業技術開發硏究 Vol.18 No.-

The wheel load distribution factors specified by Korean Bridge Design Code and AASHTO Specifications don't account for the effect of skew. Therefore use of inappropriate wheel load distribution factors can mislead to design shear in skewed bridges. This study is making understanding of property of skew bridges and concept of distribution factor for a shear in skew bridge obtuse and gives overview of problems of existing formulas, and suggesting a finite element model using results from field tests on two plate girder bridges. The considered parameters are girder spacing, span length, slab width, skew angle. and the regression analysis is performed to develop the distribution formulas which include many parameters which have an effect on distribution factors. To confirm the validity of the proposed formulas, its results are compared with AASHTO, AASHTO LRFD. At the result, this suggesting formula is showing that increasing rate of wheel load distribution factor concerned with parameters is decreasing and this is showing definite difference with existing specifications.

중소하천유역에서 HEC-1 모형을 이용한 강우-유출 해석

이정식,이재준,박종영 國立金烏工科大學校 産業技術開發硏究院 2002 産業技術開發硏究 Vol.18 No.-

The objective of this study is to examine the characteristics of HEC-1 model, and to evaluate the rainfall loss methods and the synthetic unit hydrographs of HEC-1 model in the small to medium sizes watershed. In order to forecast the rainfall-runoff relation in ungauged watershed, the regression analysis is conducted between the parameters of Clark method and watershed characteristics. Three synthetic unit hydrograph methods of Clark, Snyder, and 'SCS method incorporated in the HEC-1 model are used to simulate the rainfall runoff process in Wi stream watershed. Clark method produces better results for peak discharge and time distribution than other methods. Four loss rate methods of initial and uniform loss rates, exponential loss function, Holtan loss function and SCS loss method can be evaluated to select the appropriate method on basis of Clark method. It is shown that the initial and uniform loss rates is significantly better than other methods. As the test of the ability of the HEC-1 model to synthesize hydrograph for ungauged watershed, the parameters of Clark method are estimated the regression equation between the parameters and watershed characteristics. From the result of the test, the simulated hydrograph has good agreement with observed one in the peak discharge and time to peak discharge.

분포형 광섬유센서를 활용한 지표이동 측정에 관한 연구

장기태 國立金烏工科大學校 産業技術開發硏究院 2002 産業技術開發硏究 Vol.18 No.-

Optical fibre sensors have shown a potential to serve real time health monitoring of the structures. They can be easily embedded or attached to the structures and are not affected by the electro-magnetic field. Furthermore, they have the flexibility of the sensor size and very highly sensitive. In this study, we conducted several laboratory and field tests using a novel optical sensor based on Brillouin scattering. One of the advantages of this technique is that the bare fibre itself acts as sensing element without any special fibre processing or preparation. Test results have shown that BOTDR can be a great solution for sensor systems of Civil Engineering Smart Structures.

3GPP W-CDMA 시스템 개요 및 물리계층

송영준 國立金烏工科大學校 産業技術開發硏究院 2002 産業技術開發硏究 Vol.18 No.-

본 논문은 3세대 이동통신 3GPP (3rd Generation Partnership Project) W-CDMA (Wideband Code Division Multiple Access) 시스템 개요 및 물리계층에 관하여 논한다. W-CDMA 방식은 한국의 TTA(Telecommunication Technology Association), 일본의 ARIB (Association for Radio Industry and Business)/ TTC (Telecommunications Technology Committee), 유럽의 ETSI (European Telecommunications Standards Institute), 미국의 T1 (Standardization Committee T1 - Telecommunications), 중국의 CWTS (China Wireless Telecommunication Standard) 등의 그룹이 공동으로 규격작업을 진행중이다. 2GHz 대역에서 하나의 단말기로 전세계 어디서나 누구와 어떤 형태의 통신을 가능하게 하기 위한 IMT-2000 (International Mobile Telecommunication-2000) 기술의 하나인 W-CDMA 시스템은 현존하는 2세대 이동통신 시스템인 GSM(Global System for Mobile communications)을 핵심 망으로 하며 진화한다. 다양한 무선 통신 환경에서 144kbps-2Mbps(bit per second)의 데이터 전송속도와 양질의 음성통화를 동시에 보장하며, 논리 채널에서의 다양한 멀티미디어 서비스를 한 개 또는 그이상의 물리채널에 전송하는 것을 특징으로 하고 있다. 최대의 전송속도와 개선된 주파수 효율을 가져오기 위하여, 데이터 전송용 터보 코딩 (turbo coding), 고속 전력제어, 다이버시티 기법, 다른 시스템간의 핸드오버를 위한 압축모드, 다양한 서비스를 제공하기 위한 TFCI(Transport Format Combination Indicator) 코딩기법등이 사용된다. This paper discusses overview of 3GPP(3rd Generation Partnership Project) W-CDMA(Wideband Code Division Multiple Access) system and physical layer. The specification of W-CDMA system is under progress with international coorporation from TTA(Telecommunication Technology Association) of Korea, ARIB(Association for Radio Industry and Business) of Japan, ETSI(European Telecommunications Standards Institute) of Eruope, T1(Standardization Committee T1-Telecommunications) of USA, and CWTS(China Wireless Telecommunication Standard) of China. W-CDMA system based on GSM(Blogal System for Mobile Telecommunications) is one of IMT-2000(International Mobile Telecommunication-2000) technology which enables global communication with single terminal in 2GHz frequency band. The wireless system is designed to have voice services of wireline quality and to provide high bit rate data service of 144kbps to 2Mbps depending on the radio environment for various multimedia services. To accomplish this purposes, W-CDMA system employes advanced technologies such as turbo coding, fast power control, various diversity methods, compressed mode for hard handover, and TFCI (Transport Format Combination Indicator) coding.

Microstructural Evolution in Binary Forsterite(Mg_2SiO_4)-Spinel(MgAl_2O_4) System

Han, Young-Hwan,Kim, Sung-jin 國立金烏工科大學校 産業技術開發硏究院 2002 産業技術開發硏究 Vol.18 No.-

Spinel(MgAl_2O_4) additions to the pure forsterite(Mg_2SiO_4) (forsterite/spinel binary system) require extensive study relative to the spinel effects on the grain growth of the forsterite for an possible attractive refractory for industrial applications. Therefore, several binary compositions in the spinel(MgAl_2O_4)/forsterite(Mg_2SiO_4) system were investigated for their densities and microstructural evolution between 1400℃ and 170℃(1650℃). Utilizing the purse synthetic powder and a commercial fine spinel, specimens were pressed and then sintered for several time/temperature conditions between 1400℃ and 1700℃(1650℃). The fired densities of the compositions range from about 70% to 90% of the theoretical values. Generally, the spinel additions have inhibited the grain growth of the forsterite.