• 한국수자원학회논문집
• /
• 제53권2호
• /
• pp.97-105
• /
• 2020

최근 데이터 예측 방법으로 인공신경망(Artificial Neural Network, ANN)분야에 대한 관심이 높아졌으며, 그 중 시계열 데이터 예측에 특화된 LSTM(Long Short-Term Memory)모형은 수문 시계열자료의 예측방법으로도 활용되고 있다. 본 연구에서는 구글에서 제공하는 딥러닝 오픈소스 라이브러리인 텐서플로우(TensorFlow)를 활용하여 LSTM모형을 구축하고 금강 상류에 위치한 용담다목적댐의 유입량을 예측하였다. 분석 자료로는 WAMIS에서 제공하는 용담댐의 2006년부터 2018년까지의 시간당 유입량 자료를 사용하였으며, 예측된 유입량과 관측 유입량의 비교를 통하여 평균제곱오차(RMSE), 평균절대오차(MAE), 용적오차(VE)를 계산하고 모형의 학습변수에 따른 정확도를 평가하였다. 분석결과, 모든 모형이 고유량에서의 정확도가 낮은 것으로 나타났으며, 이와 같은 문제를 해결하기 위하여 용담댐 유역의 시간당 강수량 자료를 추가 학습 자료로 활용하여 분석한 결과, 고유량에 대한 예측의 정확도가 높아지는 것을 알 수 있었다.

• 한국전자거래학회지
• /
• 제1권1호
• /
• pp.27-50
• /
• 1996

In this paper. we present a definition of the term virtual enterprise (VE) and discuss some related terms and concepts. The definition was given based on Porter's Value Chain Theory. Then we discuss some positioning strategies for entering or joining a virtual enterprise from an executive's viewpoint. which show a framework for making decisions on for what. when and how a (real) enterprise need to participate or form a virtual enterprise and whom the partners should be. The major pert of the paper is on underlying information technologies and architectures for building virtual enterprises. Requirements for modeling virtual enterprises and core technologies needed for implementation are discussed. A virtual enterprise reference architecture based on NIIIP's model is presented. Finally we discuss some points that should. addressed from a macroeconomic and public policies viewpoints. These are the issues of IT standardization and investments in IT infrastructure.

• 한국전산유체공학회지
• /
• 제10권1호
• /
• pp.80-86
• /
• 2005

In order to control the transonic flow field with a shock wave, a condensing flow was produced by an expansion of moist air on a circular bump model and shock waves were occurred in the supersonic parts of the fields. Furthermore, the additional passive technique of shock-boundary layer interaction using the porous wall with a cavity underneath was adopted in this flow field. The effects of these methods on the shock wave characteristics were investigated numerically. The result showed that the flow fields might be effectively controlled by the suitable combination between non-equilibrium condensation and the position of porous wall.

• 한국정밀공학회지
• /
• 제20권5호
• /
• pp.124-131
• /
• 2003

As the new model development lead time becoming shorter and the market requirements becoming more strict, automobile manufacturer begins to utilize 3-dimensional CAD system based techniques such as DMU (Digital Mock-up), Rf (Rapid Prototyping), VE (Virtual Engineering) to meet the market trends. But, no satisfactory utilities have been developed yet, to represent emotional properties such as embossment on the surface of interior parts, touch originated from material characteristics in virtual environment, so it is inevitable to manufacture prototype parts to verify actual feeling which passengers feel in automobile. This paper suggests a methodology to enhance the human sensitivity via embodying embossment on the surface of prototype car interior trim without deterioration of dimensional accuracy using RIM (Reaction Injection Molding) and thermoforming method.

• 대한의용생체공학회:의공학회지
• /
• 제13권1호
• /
• pp.1-8
• /
• 1992

In the field of prosthesis arm control, the pattern classification of the EMG signal is a required basis process and also the estimation of force from collected EMG data is another necessary duty. But unfortunately, what we've got is not real force but an EMG signal which contains the information of force. This is the reason why we estimate the force from the EMG data. In this paper, when we handle the EMG signal to estimate the force, spatial prewhitening process is applied from which the spatial correlation between the channels are removed. And after the orthogonal transformation which is used in the force estimation process, the transformed signal Is inputed into the probabilistic model for pattern classification. To verify the different results of the multiple channels, SNR(signal to noise ratio) function is introduced.

• 말소리와 음성과학
• /
• 제7권4호
• /
• pp.49-55
• /
• 2015

It is argued in this paper that, in determining semantic similarity, Korean words should be recategorized with a focus on the semantic relation to ontology in light of cross-linguistic morphological variations. It is proposed, in particular, that Korean semantic similarity should be measured on three tracks, human judgements track, relatedness track, and cross-part-of-speech relations track. As demonstrated in Yang et al. (2015), GloVe, the unsupervised learning machine on semantic similarity, is applicable to Korean with its performance being compared with human judgement results. Based on this compatability, it was further thought that the model's performance might most likely vary with different kinds of specific relations in different languages. An attempt was made to analyze them in terms of two major Korean-specific categories involved in their lexical and cross-POS-relations. It is concluded that languages must be analyzed by varying methods so that semantic components across languages may allow varying semantic distance in the vector space models.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2000년도 추계학술대회 논문집
• /
• pp.358-361
• /
• 2000

To reduce the leadtime for a new model according to the strict market requirements, automobile manufacturer begins to utilize 3-dimensional CAD based techniques such as DMU(Digital Mock-Up). RP(Rapid Prototyping), VE(Virtual Engineering). But, not so many satisfactory utilities have been introduced to deal with the emotional properties such as embossment on the surface of interior parts and touch from material characteristics in virtual environment. It is required to manufacture prototype parts to verify actual feeling of the passengers in real automobile. This paper suggests a methodology to enhance emotional property via embedding embossment on the surface of prototype car interior trim without deterioration of dimensional accuracy using RIM(Reaction Injection Molding) and vacuum forming method.

• 한국CDE학회논문집
• /
• 제15권5호
• /
• pp.343-353
• /
• 2010

Time to market and cost-efficient production are some of the challenge that manufacturing industries face. Modern methods of engineering can't help such organizations attain competitive advantage. To help these situations, MEMPHIS (Middleware for Exchanging Machinery and Product Data in Highly Immersive Systems) was introduced as an approach that enables VE (Virtual Engineering) and links engineering applications with VR (Virtual Reality) solutions. Thus an environment is provided to implement virtual design reviews and enable the application of virtual prototyping methods. However MEMPHIS could just handle Product data for virtual design review and simulation. In this paper, we newly define and develop the extended MEMPHIS that enables virtual manufacturing with Process, Resource and Plant data as well as Product data.

• 한국의학물리학회:학술대회논문집
• /
• 한국의학물리학회 2005년도 제30회 춘계학술대회
• /
• pp.79-82
• /
• 2005

In this study, we have tested the feasibility of the convex non-linear objective model and the line search optimization method for the fluence map optimization (FMO). We've created the convex nonlinear objective function with simple bound constraints and attained the optimal solution using well-known gradient algorithms with an Armijo line search that requires sufficient objective function decrease. The algorithms were applied to 10 head-and-neck cases. The numbers of beamlets were between 900 and 2,100 with a 3 mm isotropic dose grid. Nonlinear optimization methods could efficiently solve the IMRT FMO problem in well under a minute with high quality for clinical cases.

• 한국CDE학회논문집
• /
• 제6권4호
• /
• pp.236-243
• /
• 2001

A key factor far realizing the internet-based virtual manufacturing system(VMS) and virtual enter-prise(VE) is how to precisely and effectively represent the machine elements and mechanics. In this paper, we present methods to represent the numerically controlled machine tools in the internet environment. The method is composed of: 1) geometrical modeling of the machine tools, 2) kinematic modeling for the movements of the machine tools, and 3) representing the developed model in the internet infrastructure. Based on the models. a web-based virtual machine tools (WVMT) is developed, and it can be accessed at hrrp://wvmt.postech.ac.kr. The WVMT can be used for various purposes: 1) web-based virtual manufacturing system, 2) web-based CAM system, and 3) CNC educational tools for the vocational school through internet.