• 대기
• /
• 제30권2호
• /
• pp.169-181
• /
• 2020

We conducted a study on the impact of observation station density; this was done in order to enable the accurate estimation of spatial meteorological variables. The purpose of this study is to help operate an efficient observation network by examining distributions of temperature, relative humidity, and wind speed in a test area of a three-dimensional meteorological observation project in the Yeongdong region in 2019. For our analysis, we grouped the observation stations as follows: 41 stations (for Step 4), 34 stations (for Step 3), 17 stations (for Step 2), and 10 stations (for Step 1). Grid values were interpolated using the kriging method. We compared the spatial accuracy of the estimated meteorological grid by using station density. The effect of increased observation network density varied and was dependent on meteorological variables and weather conditions. The temperature is sufficient for the current weather observation network (featuring an average distance about 9.30 km between stations), and the relative humidity is sufficient when the average distance between stations is about 5.04 km. However, it is recommended that all observation networks, with an average distance of approximately 4.59 km between stations, be utilized for monitoring wind speed. In addition, this also enables the operation of an effective observation network through the classification of outliers.

• 융합신호처리학회논문지
• /
• 제23권4호
• /
• pp.234-240
• /
• 2022

본 논문은 직종별 근무 환경에 따른 상대적 운동량을 고려한 맞춤형 AI 운동 추천 서비스 방법을 제안한다. 가속도 및 자이로 센서를 활용하여 수집된 데이터를 18가지 일상생활의 신체활동으로 분류한 WISDM 데이터베이스를 기반으로 전신, 하체, 상체의 3가지 활동으로 분류한 후 인식된 활동 지표를 통해 적절한 운동을 추천한다. 본 논문에서 신체활동 분류를 위해서 사용하는 1차원 합성곱 신경망(1D CNN; 1 Dimensional Convolutional Neural Network) 모델은 커널 크기가 다른 다수의 1D 컨볼루션(Convolution) 계층을 병렬적으로 연결한 컨볼루션 블록을 사용한다. 컨볼루션 블록은 하나의 입력 데이터에 다층 1D 컨볼루션을 적용함으로써 심층 신경망 모델로 추출할 수 있는 입력 패턴의 세부 지역 특징을 보다 얇은 계층으로도 효과적으로 추출 할 수 있다. 제안한 신경망 모델의 성능 평가를 위해서 기존 순환 신경망(RNN; Recurrent Neural Network) 모델과 비교 실험한 결과 98.4%의 현저한 정확도를 보였다.

• 한국컴퓨터정보학회논문지
• /
• 제24권8호
• /
• pp.87-93
• /
• 2019

These days, as interest in the recommendation system with deep learning is increasing, a number of related studies to develop a performance for collaborative filtering through autoencoder, a state-of-the-art deep learning neural network architecture has advanced considerably. The purpose of this study is to propose autoencoder which is used by the recommendation system to predict ratings, and we added more hidden layers to the original architecture of autoencoder so that we implemented deep autoencoder with 3 to 5 hidden layers for much deeper architecture. In this paper, therefore we make a comparison between the performance of them. In this research, we use 2-dimensional arrays and 3-dimensional tensor as the input dataset. As a result, we found a correlation between matrix entry of the 3-dimensional dataset such as item-time and user-time and also figured out that deep autoencoder with extra hidden layers generalized even better performance than autoencoder.

• Journal of Welding and Joining
• /
• 제21권4호
• /
• pp.31-38
• /
• 2003

Line heating is an effective and economical process for forming flat metal plates into three-dimensional shapes for plating of ships. Because the nature of the line heating process is a transient thermal process, followed by a thermo elastic plastic stress field, predicting deformed shapes of plate is very difficult and complex problem. In this paper, neural network model o3r solving the inverse problem of metal forming is proposed. The backpropagation neural network systems for determining line-heating positions from object shape of plate are reported in this paper. Two cases of the network are constructed-the first case has 18 lines which have different positions and directions and the second case has 10 parallel heating lines. The input data are vertical displacements of plate and the output data are selected heating lines. The train sets of neural network are obtained by using an analytical solution that predicts plate deformations in line heating process. This method shows the feasibility that the neural network can be used to determine the heating-line positions in line heating process.

• Bulletin of the Korean Chemical Society
• /
• 제34권8호
• /
• pp.2375-2380
• /
• 2013

Two metal-organic frameworks based on the connectivity co-effect between rigid benzenedicarboxylic acid and bridging ligand have been synthesized $[Zn_2(3-NO_2-bdc)_2(4,4'-bpy)_2H_2O]_n$ (1), $[Co(3-NO_2-bdc)(4,4'-bpy)H_2O]_n$ (2) (where $3-NO_2-bdcH_2$ = 3-nitro-1,2-benzenedicarboxylic acid, 4,4'-bpy = 4,4'-bipyridine). The two novel complexes were characterized by IR spectrum, elemental analysis, fluorescent properties, thermogravimetric analysis, single-crystal X-ray diffraction and powder X-ray diffraction (PXRD). X-ray structure analysis reveals that 1 and 2 are two-dimensional (2D) network structures. Complex 1 and complex 2 belong to triclinic crystal with P-1 space group. The luminescence measurements reveal that two complexes exhibit good fluorescent emissions in the solid state at room temperature. Also, thermal decomposition process and powder X-ray diffraction of complexes were investigated.

• ETRI Journal
• /
• 제46권2호
• /
• pp.290-306
• /
• 2024

To treat the novel COronaVIrus Disease (COVID), comparatively fewer medicines have been approved. Due to the global pandemic status of COVID, several medicines are being developed to treat patients. The modern COVID medicines development process has various challenges, including predicting and detecting hazardous COVID medicine responses. Moreover, correctly predicting harmful COVID medicine reactions is essential for health safety. Significant developments in computational models in medicine development can make it possible to identify adverse COVID medicine reactions. Since the beginning of the COVID pandemic, there has been significant demand for developing COVID medicines. Therefore, this paper presents the transferlearning methodology and a multilabel convolutional neural network for COVID (MLCNN-COV) medicines development model to identify negative responses of COVID medicines. For analysis, a framework is proposed with five multilabel transfer-learning models, namely, MobileNetv2, ResNet50, VGG19, DenseNet201, and Inceptionv3, and an MLCNN-COV model is designed with an image augmentation (IA) technique and validated through experiments on the image of three-dimensional chemical conformer of 17 number of COVID medicines. The RGB color channel is utilized to represent the feature of the image, and image features are extracted by employing the Convolution2D and MaxPooling2D layer. The findings of the current MLCNN-COV are promising, and it can identify individual adverse reactions of medicines, with the accuracy ranging from 88.24% to 100%, which outperformed the transfer-learning model's performance. It shows that three-dimensional conformers adequately identify negative COVID medicine responses.

• 한국정보과학회논문지:정보통신
• /
• 제29권6호
• /
• pp.696-705
• /
• 2002

본 논문은 멀티캐스트 ATM 스위치를 위한 새로운 복사망에 관하여 언급한다. Lee의 복사망 제안 이후 현재까지 많은 연구가 진행되어 왔으나 스위치 내에서의 오버플로우와 셀 충돌 문제는 여전히 문제점으로 제기되어 왔다. 본 논문에서는 치러한 문제점을 개선하기 위해 공유 버퍼를 갖는 3차원 멀티캐스트 스위치 구조를 제안함으로서 높은 부하에서도 다중경로와 다중출력을 통해 오버플로우와 충돌을 줄일 수 있는 방법에 관하여 언급한다. 이를 위해 큰 팬아웃의 셀을 처리하기 위한 셀 분할 알고리즘(Cell splitting algorithm)과 Lee의 Broadcast Banyan Network을 확장시킨 복사망을 재안하였다. 셀 복사는 셀의 두 비트 멀티캐스트 패턴과 부울 방식의 내부분할 알고리즘(Boolean interval splitting algorithm)을 사용하여 네트워크가 갖는 자가-라우팅의 특성에 따라 이루어지도록 하였다. 제안된 복사망에서는 Lee의 복사망에서 문제가 되었던 오버플로우 문제, 큰 팬아웃의 셀 처리 문제, 셀 충돌 문제 등을 개선시킴으로서 기존의 멀티캐스트 ATM 스위치 성능을 향상시키고자 하였다. 시뮬래이션에 의한 성능 평가 결과 기존의 방법과 비교하여 산출량, 셀 손실률, 셀 지연에 대해 좋은 성능을 보이고 있다.

• Journal of Information Processing Systems
• /
• 제17권3호
• /
• pp.505-517
• /
• 2021

This research presented an implementation of autostereoscopic virtual three-dimensional (3D) button device as non-contact style. The proposed device has several characteristics about visible feature, non-contact use and artificial intelligence (AI) engine. The device was designed to be contactless to prevent virus contamination and consists of 3D buttons in a virtual stereoscopic view. To specify the button pressed virtually by fingertip pointing, a simple deep learning network having two stages without convolution filters was designed. As confirmed in the experiment, if the input data composition is clearly designed, the deep learning network does not need to be configured so complexly. As the results of testing and evaluation by the certification institute, the proposed button device shows high reliability and stability.

• 한국수소및신에너지학회논문집
• /
• 제20권5호
• /
• pp.361-370
• /
• 2009

Ni-PTFE composite plated on graphite (C/Ni-PTFE) and PTFE (PTFE/Ni-PTFE) particles were prepared uniformly by electroless composite plating. The conductivity of C/Ni-PTFE particles was 280 S/m higher than 95 S/m of PTFE/Ni-PTFE particles at same composite plating condition (Ni:35~36 wt%, PTFE:8 wt%). The C/Ni-PTFE particles were formed into the C/Ni-PTFE plate using heat treatment at $350^{\circ}C$ under 10~$1000\;kg/cm^2$. The C/Ni-PTFE plate showed 1) high conductivity of $5.7\;{\times}\;10^4\;S/m$ due to the existence of graphite as conducting aid and the formation of 3-dimensional Ni network 2) good gas diffusion caused by various pore volumes (0.01~$100\;{\mu}m$) in the plate. The plate could be useful for an electrode in an alkaline fuel cell (AFC). The current density of C/Ni-PTFE electrode indicated $84\;mA/cm^2$ at 0.3V and it was 3.0 times higher than that of PTFE/Ni-PTFE electrode.

• 한국멀티미디어학회논문지
• /
• 제10권11호
• /
• pp.1460-1471
• /
• 2007

휴대용 임베디드 기기에서의 삼차원 엔진은 크게 바이트 코드를 실시간으로 해석하며 실행하는 자바 기반의 JSR184와 C언어 기반의 OpenGL/ES가 있다. 이들 두 표준에서 자바 객체를 지원하는 JSR184는 OpenGL/ES에 비하여 상대적으로 많은 프로세서의 자원을 사용하여 제한된 연산능력을 보유하고 있는 임베디드 기기에 적용할 경우 제약이 따를 수 밖에 없다. 반면에 기존 개인용 컴퓨팅 환경에서 사용되는 삼차원 컨텐츠는 자바의 장점을 이용하여 제작되었기 때문에 유럽에서 많은 사용자 층을 확보하고 있고, 또한 그 컨텐츠의 품질이 우수하여 상용 통신망인 GSM 망에서 많이 서비스 되고 있다. 따라서 GSM 망에서 사용되는 휴대용 임베디드 기기에 기존의 자바 기반 삼차원 컨텐츠를 별도의 변환 과정 없이 지원할 수 있는 JSR184의 지원이 필요하지만, 현재 개발되어 사용되는 자바 기반 삼차원 엔진은 휴대용 기기가 보유한 연산능력에 비하여 상대적으로 많은 연산량을 필요로 하기 때문에 상용제품에 적용하기에 많은 어려움이 따른다. 본 논문에서는 휴대용 임베디드 기기가 가지고 있는 충분하지 않은 연산능력을 바탕으로 자바 객체의 장점을 수용하면서 삼차원 컨텐츠의 처리속도를 향상 시킬 수 있는 바인딩 기법을 제안하였다. 제안된 바인딩 기법은 자바를 이용한 삼차원 컨텐츠를 지원하기 위하여, JSR184의 표준 인터페이스를 상위 계층에서 지원하고, OpenGL/ES와 JSR184를 서로 연결하기 위하여 이기종 코드 변환 언어인 KNI(Kilo Native Interface)를 중간 계층에서 사용하였고, 하위 계층에서 OpenGL/ES의 표준을 구현하였다. 제안하는 바인딩 기법은 모의실험을 통하여 기능을 검증하였고, ARM을 장착한 FPGA를 사용하여 그 성능을 평가하였다.