• ETRI Journal
• /
• 제44권4호
• /
• pp.672-685
• /
• 2022

This article proposes a subject-independent application of brain-computer interfacing (BCI). A 32-channel Electroencephalography (EEG) device is used to measure imagined speech (SI) of four words (sos, stop, medicine, washroom) and one phrase (come-here) across 13 subjects. A deep long short-term memory (LSTM) network has been adopted to recognize the above signals in seven EEG frequency bands individually in nine major regions of the brain. The results show a maximum accuracy of 73.56% and a network prediction time (NPT) of 0.14 s which are superior to other state-of-the-art techniques in the literature. Our analysis reveals that the alpha band can recognize SI better than other EEG frequencies. To reinforce our findings, the above work has been compared by models based on the gated recurrent unit (GRU), convolutional neural network (CNN), and six conventional classifiers. The results show that the LSTM model has 46.86% more average accuracy in the alpha band and 74.54% less average NPT than CNN. The maximum accuracy of GRU was 8.34% less than the LSTM network. Deep networks performed better than traditional classifiers.

• Nuclear Engineering and Technology
• /
• 제50권4호
• /
• pp.582-588
• /
• 2018

Accident diagnosis is one of the complex tasks for nuclear power plant (NPP) operators. In abnormal or emergency situations, the diagnostic activity of the NPP states is burdensome though necessary. Numerous computer-based methods and operator support systems have been suggested to address this problem. Among them, the recurrent neural network (RNN) has performed well at analyzing time series data. This study proposes an algorithm for accident diagnosis using long short-term memory (LSTM), which is a kind of RNN, which improves the limitation for time reflection. The algorithm consists of preprocessing, the LSTM network, and postprocessing. In the LSTM-based algorithm, preprocessed input variables are calculated to output the accident diagnosis results. The outputs are also postprocessed using softmax to determine the ranking of accident diagnosis results with probabilities. This algorithm was trained using a compact nuclear simulator for several accidents: a loss of coolant accident, a steam generator tube rupture, and a main steam line break. The trained algorithm was also tested to demonstrate the feasibility of diagnosing NPP accidents.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1999년도 추계학술대회 논문집 학회본부 A
• /
• pp.239-241
• /
• 1999

This paper proposes a novel wavelet transform and Kohonen neural network based technique for short-time load forecasting of power systems. Firstly. Kohonen Self-organizing map(KSOM) is applied to classify the loads and then the Daubechies D2, D4 and D10 wavelet transforms are adopted in order to forecast the short-term loads. The wavelet coefficients associated with certain frequency and time localisation are adjusted using the conventional multiple regression method and then reconstructed in order to forecast the final loads through a four-scale synthesis technique. The outcome of the study clearly indicates that the proposed composite model of Kohonen neural network and wavelet transform approach can be used as an attractive and effective means for short-term load forecasting.

• 융합보안논문지
• /
• 제17권5호
• /
• pp.57-61
• /
• 2017

국가 연구망(National Research Network)은 각국의 첨단과학기술을 지원하기 위해 존재한다. 국가 연구망은 각 분야의 과학기술에 대한 요구사항을 충족하여야 하며, 지속적으로 변해가는 주변 환경에 대해서 끊임없이 발전해 나가야 한다. 최근 데이터 중심의 과학기술 연구형태에 따른 수요가 증가하고 있으며, 이러한 수요에 대응하기 위해 선진 국가 연구망들은 중장기적인 전략을 수립하고 있다. 본 논문에서는 선진 연구망에 대한 현황과 국가 연구망 동향 및 요구사항을 연구망환경의 변화, 과학/기술의 변화, 산업과 삶의 변화의 관점에서 분석하였다. 또한 분석한 요구사항에 대응하기 위해 미래형 네트워크 백본 구축, 사용자 중심의 협업 연구 환경 제공, 글로벌 협업 네트워크 서비스 제공, 첨단과학기술 연구데이터 정보 보호 서비스 제공 등에 관해 중장기 발전 방향 및 방안을 제시한다.

• 보건의료산업학회지
• /
• 제2권1호
• /
• pp.125-136
• /
• 2008

Government came to enforce long-term care insurance system in preparation for the rapid aging society. Whether this system is successful or not depends on the professionalism of long-term care-givers who are professional population in charge of care service. Currently in the early stage of enforcement, such problems as a race cutting fee resulted from numerical increase of educational facilities, insolvent operation, degradation of education level resulted from unprofessional instructor, are pointed out. As a mean of manpower-training on long-term care insurance system, this study is to research public-private-university network model of the Academy of Continuing Education attached to Ko Sin University which is the case of Yong Do Gu Busan city. Networking between the vision and development strategy of Yong Do Gu on continuing education city, education system on community manpower-training supported by Ko Sin University, and service field of welfare for the elderly can not only contribute to the professionalism of long-term care-givers but also play an ideal role in manpower-utilization and arrangement of community. Through this networking, high quality of education level and circumstance, using the existing infra, manpower-training and utilization for continuing education of Yong Do Gu can be accomplished. Additionally, the connection with facilities related with welfare for the elderly can contribute to professionalism and accountability of manpower-networking.

• 한국물환경학회지
• /
• 제29권3호
• /
• pp.343-353
• /
• 2013

Flow duration curve (FDC) can be developed by linking the daily flow data of stream flow monitoring network to 8-day interval flow data of the unit watersheds for the management of Total Maximum Daily Loads. This study investigated the applicable method for the development of long term FDC with the selection of the stream flow reference sites, and suggested the development of the FDC in 4 river basins. Out of 142 unit watersheds in 4 river basins, 107 unit watersheds were shown to estimate daily flow data for the unit watersheds from 2006 to 2010. Short term FDC could be developed in 64 unit watersheds (45%) and long term FDC in 43 unit watersheds (30%), while other 35 unit watersheds (25%) were revealed to have difficulties in the development of FDC itself. Limits in the development of the long term FDC includes no stream monitoring sites in certain unit watersheds, short duration of stream flow data set and missing data by abnormal water level measurements on the stream flow monitoring sites. To improve these limits, it is necessary to install new monitoring sites in the required areas, to keep up continuous monitoring and make normal water level observations on the stream flow monitoring sites, and to build up a special management system to enhance data reliability. The development of long term FDC for the unit watersheds can be established appropriately with the normal and durable measurement on the selected reference sites in the stream flow monitoring network.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제31권12호
• /
• pp.1660-1671
• /
• 2004

본 논문은 두 문서간 유사도 측정 방법을 제안한다. 제안한 유사도 측정 모델의 주안점은 문서간 관련성의 정도를 두 문서간 일치하는 단어(term)및 단어쌍(tenn-phrase)에 기반하여 이들이 해당 문서에서 차지하는 가중치를 통해 측정하는 것이다. 유사도 측정 과정에 영향을 미치는 특징을 설계함에 있어 기존의 연구들이 하나의 특징만을 고려하였던 것에 비하여 본 논문은 여러 가지 특징들을 고려한다 즉, 단어뿐만 아니라 단어쌍과 관련된 특징을 결합하여 신경망을 통해 유사도를 측정한다. 제안된 방법의 우수성을 입증하기 위해 두 가지 측면에서 실험하였다. 첫 번째는 두 문서의 동일성 여부를 검증하는 문제이며, 두 번째는 다수의 문서를 대상으로 유사한 문서를 찾는 검색 문제이다. 이 두 가지 실험 모두에서 제안 방법이 기존의 Cosine 유사도 계산 방법 및 구색인 방법에 비해 우수한 성능을 보였다.

• 해양환경안전학회지
• /
• 제28권4호
• /
• pp.469-476
• /
• 2022

본 연구는 조위 관측자료를 이용하여 부산항에서의 장기 조위 자료를 생성하는 Long Short-Term Memory (LSTM)으로 구현된 순환신경망 모델을 개발하였다. 국립해양조사원의 부산 신항과 통영에서 관측된 조위 자료를 모델 입력 자료로 사용하여 부산항의 조위를 예측하였다. 모델에 대하여 2019년 1월 한 달의 학습을 수행하였으며, 이후 2019년 2월에서 2020년 1월까지 1년에 대하여 정확도를 계산하였다. 구축된 모델은 부산 신항과 통영의 조위 시계열을 함께 입력한 경우에 상관계수 0.997 및 평균 제곱근 오차 2.69 m로 가장 성능이 높았다. 본 연구 결과를 바탕으로 딥러닝 순환신경망 모델을 이용하여 임의 항만의 장기 조위 자료 예측이 가능함을 알 수 있었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권2호
• /
• pp.522-541
• /
• 2016

Network virtualization provides an effective way to overcome the Internet ossification problem. As one of the main challenges in network virtualization, virtual network embedding refers to mapping multiple virtual networks onto a shared substrate network. However, existing heuristic embedding algorithms evaluate the embedding potential of the nodes simply by the product of different resource attributes, which would result in an unbalanced embedding. Furthermore, ignoring the hops of substrate paths that the virtual links would be mapped onto may restrict the ability of the substrate network to accept additional virtual network requests, and lead to low utilization rate of resource. In this paper, we introduce and extend five node attributes that quantify the embedding potential of the nodes from both the local and global views, and adopt the technique for order preference by similarity ideal solution (TOPSIS) to rank the nodes, aiming at balancing different node attributes to increase the utilization rate of resource. Moreover, we propose a novel two-stage virtual network embedding algorithm, which maps the virtual nodes onto the substrate nodes according to the node ranks, and adopts a shortest path-based algorithm to map the virtual links. Simulation results show that the new algorithm significantly increases the long-term average revenue, the long-term revenue to cost ratio and the acceptance ratio.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.2155-2158
• /
• 2003

A term of home automation that was in fashion only a few years ago has not been used any more. Nowadays, We have been used a term of home network or digital home than home automation much. While internet infra is diffused at home, data network corp., communication corp., electric appliance corp. and home automation control system corp. which we did not mind each other particularly constructed consortium, and they have designs on home network market. Also, cellular phone's growth tried home networking by using not only wired internet but also broadband wireless communication. Regardless, many solutions are coming out it is few to be applied to real life because the standard is not determined with the protocol each other. Therefore, we developed home network system using USB(Universal Serial Bus) that has the possibility most in home networking standard. The mobile USB home control system is excellent at expansibility and portability. Also we can complete low cost and stable system using an embedded system.