• Nuclear Engineering and Technology
• /
• 제54권1호
• /
• pp.269-274
• /
• 2022

An artificial neural network (ANN) that identifies radionuclides from low-count gamma spectra of a NaI scintillator is proposed. The ANN was trained and tested using simulated spectra. 14 target nuclides were considered corresponding to the requisite radionuclide library of a radionuclide identification device mentioned in IEC 62327-2017. The network shows an average identification accuracy of 98.63% on the validation dataset, with the gross counts in each spectrum N_c = 100~10000 and the signal to noise ratio SNR = 0.05-1. Most of the false predictions come from nuclides with low branching ratio and/or similar decay energies. If the N_c>1000 and SNR>0.3, which is defined as the minimum identifiable condition, the averaged identification accuracy is 99.87%. Even when the source and the detector are covered with lead bricks and the response function of the detector thus varies, the ANN which was trained using non-shielding spectra still shows high accuracy as long as the minimum identifiable condition is satisfied. Among all the considered nuclides, only the identification accuracy of ²³⁵U is seriously affected by the shielding. Identification of other nuclides shows high accuracy even the shielding condition is changed, which indicates that the ANN has good generalization performance.

• IMMUNE NETWORK
• /
• 제21권4호
• /
• pp.28.1-28.14
• /
• 2021

Lung-resident memory T cells (T_RM) play an essential role in protecting against pulmonary virus infection. Parenteral administration of DNA vaccine is generally not sufficient to induce lung CD8 T_RM cells. This study investigates whether intramuscularly administered DNA vaccine expressing the nucleoprotein (NP) induces lung T_RM cells and protects against the influenza B virus. The results show that DNA vaccination poorly generates lung T_RM cells and massive secondary effector CD8 T cells entering the lungs after challenge infection do not offer sufficient protection. Nonetheless, intranasal administration of non-replicating adenovirus vector expressing no Ag following priming DNA vaccination deploys NP-specific CD8 T_RM cells in the lungs, which subsequently offers complete protection. This novel 'prime and deploy' strategy could be a promising regimen for a universal influenza vaccine targeting the conserved NP Ag.

• Journal of Power Electronics
• /
• 제18권5호
• /
• pp.1357-1368
• /
• 2018

In this paper, a novel high gain bidirectional modular dc-dc converter (BMC) with unipolar and bipolar structures for dc network interconnections is proposed. When compared with traditional dc grid-connecting converters, the proposed converter can achieve a high voltage gain with a simple modular transformerless structure. A sub-modular structure for the BMC is proposed to eliminate the unbalanced current stress between the different power units (levels) in the BMC. This can realize current sharing and standardized production and assembling. In addition, phase-interval operation is introduced to the sub-modules to realize low voltage and current ripple in both sides of the converter. Furthermore, two types of bipolar topologies of the sub-modular BMC were proposed to extend its application in bipolar dc network connections. In addition, the control system was optimized for grid-connection applications by providing various control strategies. Finally, simulations of a 3-level unipolar sub-modular BMC and a 4-level bipolar sub-modular BMC were conducted, and a 1-kW experimental 3-level unipolar prototype was developed to verify the effectiveness of the proposed converter.

• Journal of Electrical Engineering and Technology
• /
• 제12권5호
• /
• pp.1764-1776
• /
• 2017

To achieve economic stability, distribution Company as an economic institution should be managed by various processes. In this way, knowledge of different processes is the first step. Furthermore, expectations, outputs, requirement data, and sub-processes should be extracted and determined. Accordingly, to assign the performance responsibility of each process, the decision-making points must be introduced and, the deviation or change in set-points should be investigated into processes. Also, the performance of processes could be monitored by introducing of the sub-indictors. In this study, a practical method is presented for monitoring of reliability and power loss indices from viewpoint components' supply chain into the distribution network. At first, the visibility model of the supply chain is illustrated by focus group and the sub-indicators are extracted for each process of this chain. Then, validation and verification of the sub-indicators are accomplished by the Delphi method and, an information dashboard is presented by confirmed the sub-indicators and statistics methods. Finally, the proposed method is investigated by real data in a typical network and the results are analyzed.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2004년도 Proceedings of ISRS 2004
• /
• pp.577-580
• /
• 2004

To overcome with large size noise source distribution network design difficulty in interferometric radiometer system, especially for sub-Y-type array, a new on-board calibration technique using noise re-radiation is proposed in this paper. The suggested calibration technique is using noise re-radiation effect of center antenna after noise source injection from matched load. This approach is especially proper to sub-Y-type array interferometric synthetic aperture radiometer in mm-wave frequency band. Compared with noise injection network of a conventional synthetic aperture radiometer, the system mass, volume, and hardware complexity is reduced and cost-effective. Only one internal noise source, matched load, is used for injection using noise re-radiation technique a small set of sub-Y receiver channels is calibrated. Detailed calibration scenario is discussed and simulation results about noise re­radiation effect are presented.

• Bulletin of the Korean Chemical Society
• /
• 제32권6호
• /
• pp.1859-1864
• /
• 2011

Two lanthanide complexes, $[Ln(NO_3)_2(H_2O)_3(L)_2](NO_3)(H_2O)$ {Ln = Eu (1), Tb (2); L = 2-(4-pyridylium)-ethanesulfonate, $(4-pyH)^+-CH_2CH_2-SO_3^-)$}, were prepared from lanthanide nitrate and 4-pyridineethanesulfonic acid in $H_2O$ under microwave-heating conditions. Complexes 1 and 2 are isostructural, and the lanthanide metal in both complexes is coordinated to nine oxygen atoms. The pyridyl nitrogen in the ligand is protonated to give a zwitter ion that possesses an $NH^+$ (pyridyl) positive end and an $SO_3^-$ negative end. All O-H and N-H hydrogen atoms participate in hydrogen bonds to generate a two-dimensional (complex 1) or a three-dimensional network (complex 2). Complex 1 exhibits an intense red emission, whereas complex 2 exhibits an intense green emission in the solid state at room temperature.

• 전자공학회논문지
• /
• 제53권9호
• /
• pp.26-37
• /
• 2016

현재 전달망의 핵심기술은 링크 또는 노드 장애가 발생했을 경우에 경로 이중화를 통해 50ms이내에 망을 복구하는 OAM 및 보호절체 기술이다. 개별 통신사업자, 지방/중앙 정부, 중요 기업의 전달 망은 장애에 대한 실시간 망 복구를 위해 보호 서브네트워크를 개별적으로 설정, 관리되고 있다. 그래서 개별 보호 서브네트워크의 종단 노드에 대해 노드 이중화를 적용하여 종단 노드 장애에 대해 대비하는 것이 중요하다. 하지만 MPLS-TP, 캐리어 이더넷과 같은 패킷 전달망에서 선형 보호절체가 적용되는 보호 서브네트워크는 이중 노드 상호 연결 방안이 존재하지 않는다. 비록 이더넷 링 보호절체는 이중 노드 상호 연결방안을 포함하고 있지만 이더넷 링 보호절체의 기술적 특성상 연결 노드에서 장애가 발생하면 전이 트래픽이 급격하게 증가될 수 있다. 본 논문에서는 보호 서브네트워크에서 연결 노드 이중화를 위한 선형 보호절체 적용 방안을 제시한다. 그리고 링크와 상호 연결 노드의 장애에 대한 여러 실험을 통해 제안된 선형 보호와 링 보호 프로세스의 다양한 조합이 어떻게 다중 보호 서브네트워크에서 서비스 트래픽의 복구 탄력성에 영향을 미치는 지를 분석한다.

• 마이크로전자및패키징학회지
• /
• 제26권4호
• /
• pp.127-132
• /
• 2019

본 연구에서는 liquid-solid solution 합성법을 통해 고유전 페로브스카이트 구조의 barium strontium titanate(Ba_0.6Sr_0.4TiO₃, BSTO)를 합성하여 trimethylsilyl chloride(TMCS)를 silylation agent로 이용한 표면개질을 진행하였다. Silylation 표면개질을 활용하여 기존 BSTO 나노입자 표면에 있던 -OH 리간드와 TMCS가 갖고 있는 Cl을 반응시켜 나노입자 표면의 리간드를 -Si, -CH₃로 치환하였다. 다양한 TMCS 농도의 변화를 주어 silylation을 진행했고, Fourier-transform infrared spectroscopy 및 X 선 회절 분석, 전계방사 주사전자현미경을 통해 silicon network 및 결정구조, 나노입자의 크기를 확인하였다. 접촉각 변화 관찰을 통해 가장 많이 silylation된 BSTO 나노입자에서 120.9°인 소수성 특성을 확인하였다. 나노입자의 silylation을 통해 D.I water 내 BSTO 나노입자의 소수화 정도를 확인하였다.

• 지식경영연구
• /
• 제19권2호
• /
• pp.47-68
• /
• 2018

Business environment is shifting from industrial economy to knowledge based economy. Enterprises go through numerous trials for successful management in changing environment. Along with trial tests, research area has been growing simultaneously. Unlike initial research which focused on basic concepts such as: form of business model and success points. Current research emphasizes on actualization of business that enterprises plan, which brought academic research with perplex form of knowledge structure. On the other hand, there is limitation in understanding business model systematically due to preceding research primarily centered on analyzing definition and case study. In order to analyze knowledge structure, this study utilized social network analysis based on "relationship". For the analysis, 13,412 keywords were extracted from 36years worth of article or research related to business model stored in SCOPUS database. From the analysis, it was shown core research subject was INNOVATION and the number of co-authors has increased due to the academic diversity. Business model research is divided into five sub-categories (E-commerce, SMEs, sustainability, open-source, and e-book). Through cognitive map analysis on each of research characteristics of sub-category, it has shown that E-commerce, SMEs, sustainability, and open-source are core categories.

• 한국멀티미디어학회논문지
• /
• 제22권6호
• /
• pp.655-664
• /
• 2019

In this paper, we propose a novel way of combining multiple deep convolutional neural network (DCNN) architectures which work well for accurate video face identification by adopting a serial combination of 3D and 2D DCNNs. The proposed method first divides an input video sequence (to be recognized) into a number of sub-video sequences. The resulting sub-video sequences are used as input to the 3D DCNN so as to obtain the class-confidence scores for a given input video sequence by considering both temporal and spatial face feature characteristics of input video sequence. The class-confidence scores obtained from corresponding sub-video sequences is combined by forming our proposed class-confidence matrix. The resulting class-confidence matrix is then used as an input for learning 2D DCNN learning which is serially linked to 3D DCNN. Finally, fine-tuned, serially combined DCNN framework is applied for recognizing the identity present in a given test video sequence. To verify the effectiveness of our proposed method, extensive and comparative experiments have been conducted to evaluate our method on COX face databases with their standard face identification protocols. Experimental results showed that our method can achieve better or comparable identification rate compared to other state-of-the-art video FR methods.