• Nuclear Engineering and Technology
• /
• 제53권6호
• /
• pp.1769-1785
• /
• 2021

The Integral Pressurized Water Reactors (IPWRs) as the innovative advanced and generation-III + reactors are under study and developments in a lot of countries. This paper is aimed at the thermal hydraulic study of the hot and average fuel sub-channel in a Generation III + IPWR by loose external coupling to the neutronic simulation. The power produced in fuel pins is calculated by the neutronic simulation via MCNPX2.6 then fuel and coolant temperature changes along fuel sub-channels evaluated by computational fluid dynamic thermal hydraulic calculation through an iterative coupling. The relative power densities along the fuel pin in hot and average fuel sub-channel are calculated in sixteen equal divisions. The highest centerline temperature of the hottest and the average fuel pin are calculated as 633 K (359.85 ℃) and 596 K (322.85 ℃), respectively. The coolant enters the sub-channel with a temperature of 557.15 K (284 ℃) and leaves the hot sub-channel and the average sub-channel with a temperature of 596 K (322.85 ℃) and 579 K (305.85 ℃), respectively. It is shown that the spacer grids result in the enhancement of turbulence kinetic energy, convection heat transfer coefficient along the fuel sub-channels so that there is an increase in heat transfer coefficient about 40%. The local fuel pin temperature reduction in the place and downstream the space grids due to heat transfer coefficient enhancement is depicted via a graph through six iterations of neutronic and thermal hydraulic coupling calculations. Working in a low fuel temperature and keeping a significant gap below the melting point of fuel, make the IPWR as a safe type of generation -III + nuclear reactor.

• 정보처리학회논문지:컴퓨터 및 통신 시스템
• /
• 제2권12호
• /
• pp.533-540
• /
• 2013

인터넷의 발달의 주요 요인 중 하나인 익명성은 다수 사용자들의 자유로운 개인 의사 표현을 도와 다양한 인터넷 기반 분산시스템을 활성화 하는데 있어 큰 도움이 되어 왔다. 하지만, 익명성은 개인의 정보가 외부로 알려지지 않는 다는 점 때문에 악용될 소지도 다분하다. 특히 분산시스템은 한 명의 악의적인 사용자가 다수의 가짜 신분을 생성하고 조정하는 시빌 어택(Sybil Attack)에 매우 취약하게 된다. 시빌 어택을 막기 위해서 분산시스템 상에서 신분 생성 작업의 복잡도를 높이는 방식이나 시스템상의 신분과 현실상의 신분의 연결 고리를 만드는 방법을 생각 할 수 있다. 하지만 복잡도를 높이는 방식은 가짜 신분이 만들어지는 시간을 늘리는 효과만 있을 뿐, 일단 가짜 신분이 만들어진 이후에 대한 대응법이 부족하다. 또한, 현실상의 신분과의 연결을 사용할 경우 온라인 사용자의 익명성이 훼손당할 우려가 있다. 최근 온라인 소셜 네트워크의 대중화와 함께 소셜 네트워크 그래프 정보를 사용해 시빌 어택에 대응하기 위한 기법들이 연구되고 있다. 이 논문에서는 온라인 소셜 네트워크 그래프에 내포된 특성을 이용해 임의의 사용자에 대한 시스템 차원 시빌-저항 신뢰도(System-wide Sybil-resistant trust value) 추출 방법을 제안한다. 제안하는 기법은 온라인 소셜 네트워크 전체 그래프를 이해 할 수 있는 서비스 제공자들을 위한 방법으로, 샘플링 및 판단방법에 따라 3가지 종류의 세부 기법들을 제안한다. Facebook에서 추출한 온라인 소셜 네트워크 샘플 그래프를 이용하여 제안된 기법들의 성능을 분석 및 비교한다. 또한 시빌 어택의 특성을 이해하기 위해 서로 다른 노드 특성을 가지는 노드들이 시빌 어택에 의해 받는 영향을 분석한다.

• 한국측량학회지
• /
• 제38권6호
• /
• pp.623-633
• /
• 2020

사람들이 실내공간에서 소비하는 시간이 증가함에 따라, 실내공간을 대상으로 하는 서비스들에 대한 수요가 지속적으로 증가하고 있다. 실내공간정보 서비스를 제공하기 위해서는 실내공간정보를 구축하는 것이 우선적으로 이루어져야 하며 본 연구에서는 기 구축되어 있는 BIM (Building Information Management) 데이터로부터 실내공간정보 국제표준 데이터인 IndoorGML을 생성하는 방법을 제안하였는데 대표적인 BIM 데이터 자료인 IFC (Industry Foundation Class) 데이터를 IndoorGML로 변환하기 위하여 IFC 객체들의 특성을 조사하고, IndoorGML로 변환이 필요한 객체들을 선별하였으며 실내공간의 표현과 내부통행에 제약을 주는 객체로 구분하였다. 또한 개발된 변환도구를 이용하여 빔서버(BIMserver) 깃헙(github)에서 제공하는 데이터 셋 일부와 서울시립대학교 21세기관 IFC모델을 IndoorGML의 PrimalSpaceFeatures 데이터로 구축하는 실험을 수행하였다. 각각의 IFC 데이터를 IndoorGML로 변환한 결과 기하정보를 손실 없이 IndoorGML문서가 생성되었으며, IndoorGML이 가지고 있는 특징인 NavigableBoundary가 정확하게 생성되는 결과를 보였다. 추후, 다양한 형태의 IfcStair 객체를 변환하는 방법에 대한 추가적인 연구가 필요하며, 프라이멀 스페이스(PrimalSpace) 데이터를 이용하여 듀얼 스페이스(DualSpace) 데이터인 MultiLayeredGraph를 자동 생성하는 방법에 대한 개발이 필요하다.

• 한국멀티미디어학회논문지
• /
• 제23권8호
• /
• pp.965-976
• /
• 2020

We present in this work a segmentation method of E. coli bacterial images generated via phase contrast microscopy using a deep learning based hybrid feature generation. Unlike conventional machine learning methods that use the hand-crafted features, we adopt the denoising autoencoder in order to generate a precise and accurate representation of the pixels. We first construct a hybrid vector that combines original image, difference of Gaussians and image gradients. The created hybrid features are then given to a deep autoencoder that learns the pixels' internal dependencies and the cells' shape and boundary information. The latent representations learned by the autoencoder are used as the inputs of a softmax classification layer and the direct outputs from the classifier represent the coarse segmentation mask. Finally, the classifier's outputs are used as prior information for a graph partitioning based fine segmentation. We demonstrate that the proposed hybrid vector representation manages to preserve the global shape and boundary information of the cells, allowing to retrieve the majority of the cellular patterns without the need of any post-processing.

• 한국정밀공학회지
• /
• 제26권11호
• /
• pp.138-145
• /
• 2009

A suspended membrane micro fluidic heat flux sensor that is able to measure the heat flow rate was designed and fabricated by a complementary-metal-oxide-semiconductor-compatible process. The combination of a thirty-junction gold and nickel thermoelectric sensor with an ultralow noise preamplifier, low pass filter, and lock-in amp has enabled the resolution of 50 nW power and provides the sensitivity of $11.4\;mV/{\mu}W$. The heater modulation method was used to eliminate low frequency noises from sensor output. It is measured with various heat flux fluid of DI-water to test as micro fluidic application. In order to estimate the heat generation of samples from the output measurement of a micro fluidic heat-flux sensor, a methodology for modeling and simulating electro-thermal behavior in the micro fluidic heat-flux sensor with integrated electronic circuit is presented and validated. The electro-thermal model was constructed by using system dynamics, particularly the bond graph. The electro-thermal system model in which the thermal and the electrical domain are coupled expresses the heat generation of samples converts thermal input to electrical output. The proposed electro-thermal system model shows good agreement with measured output voltage response in transient state and steady-state.

• 대한전기학회논문지:전력기술부문A
• /
• 제55권9호
• /
• pp.351-358
• /
• 2006

In this paper, DSPOCS(Digital Switchgear-Panel Operation and Control Solution) is designed, which is the intelligent inference based operation and control solution to obtain the safety and reliability of electric power supply in substation based on IED. DSPOCS is designed as a scheduled monitoring and control task and a real-time alarm inference task, and is interlinked with BRES(Bus Reconfiguration Expert System) in the required case. The intelligent alarm inference task consists of the alarm knowledge generation part and the real-time pattern matching part. The alarm knowledge generation part generates automatically alarm knowledge from DB saves it in alarm knowledge base. On the other hand, the pattern matching part inferences the real-time event by comparing the real-time event information furnished from IEDs of substation with the patterns of the saved alarm knowledge base.; Especially, alarm knowledge base includes the knowledge patterns related with fault alarm, the overload alarm and the diagnosis alarm. In order to design the database independently in substation structure, busbar is represented as a connectivity node which makes the more generalized graph theory possible. Finally, DSPOCS is implemented in MS Visual $C^{++}$, MFC, the effectiveness and accuracy of the design is verified by simulation study to the typical distribution substation.

• 대한기계학회논문집
• /
• 제17권6호
• /
• pp.1330-1341
• /
• 1993

본 연구에서는 회전대우(revolute joint)들로 이루어진 1자유도 평면기구들의 특성을 파악하고 그래이프이론(graph theory)을 이용하여 기구의 형태를 행렬로 표시 하여 체이터베이스에 저장하였다. 또한 기구의 특성중의 하나인 작업조건(function specification)에 따른 무오차점(precision point)의 최대 개수를 알아내는 프로그램 을 개발하여 이로써 얻은 정보를 또 다른 데이타베이스에 저장하였다. 이와같이 구축 된 데이터베이스들로부터 설계자의 요구에 적합한 기구를 선택하여 형태합성 문제를 해결하였다.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2003년도 가을 학술발표논문집 Vol.30 No.2 (1)
• /
• pp.196-198
• /
• 2003

Virtually every digital signal processors(DSPs) support on-chip multi- memory banks that allow the processor to access multiple words of data from memory in a single instruction cycle. Also, all existing fixed-point DSPs have irregular architecture of heterogeneous register which contains multiple register files that are distributed and dedicated to different sets of instructions. Although there have been several studies conducted to efficiently assign data to multi-memory banks, most of them assumed processors with relatively simple, homogeneous general-purpose resisters. Therefore, several vendor-provided compilers fer DSPs were unable to efficiently assign data to multiple data memory banks. thereby often failing to generate highly optimized code fer their machines. This paper presents an algorithm that helps the compiler to efficiently assign data to multi- memory banks. Our algorithm differs from previous work in that it assigns variables to memory banks in separate, decoupled code generation phases, instead of a single, tightly-coupled phase. The experimental results have revealed that our decoupled algorithm greatly simplifies our code generation process; thus our compiler runs extremely fast, yet generates target code that is comparable In quality to the code generated by a coupled approach

• 산업공학
• /
• 제21권1호
• /
• pp.1-7
• /
• 2008

In this paper we study the Minimum Edge Addition Problem(MEAP) to decrease the diameter of a graph. MEAP can be used for improving the serviceability of telecommunication networks with a minimum investment. MEAP is an NP-hard optimization problem. We present two mathematical programming models : One is a multi-commodity flow formulation and the other is a path partition formulation. We propose a branch-and-price algorithm to solve the path partition formulation to the optimality. We develop a polynomial time column generation sub-routine conserving the mathematical structure of a sub problem for the path partition formulation. Computational experiments show that the path partition formulation is better than the multi-commodity flow formulation. The branch-and-price algorithm can find the optimal solutions for the immediate size graphs within reasonable time.

• 전기학회논문지
• /
• 제59권4호
• /
• pp.803-808
• /
• 2010

We propose a smooth path planning method for autonomous mobile robots. Due to nonholonomic constraints by obstacle avoidance, the smooth path planning is a complicated one. We generate smooth path that is considered orientation of robot under nonholonomic constraints. The proposed smooth planning method consists of two steps. Firstly, the initial path composed of straight lines is obtained from V-graph by Dijkstra's algorithm. Then the initial path is transformed by changing the curve. We apply the cardinal spline into the stage of curve generation. Simulation results show a performance of proposed smooth path planning method.