• 한국방재학회:학술대회논문집
• /
• 한국방재학회 2008년도 정기총회 및 학술발표대회
• /
• pp.695-698
• /
• 2008

기후변화와 지구온난화현상은 지구 전체에 걸쳐 분명하게 나타나고 있으며 그에 따라 발생할 수 있는 수문 변화에 대한 연구가 다양하게 이루어지고 있다. 본 연구에서는 기후변화에 따른 수문 변동 분석을 위하여 SWAT 모형을 이용하였으며 금강 상류유역에 적용하였다. 모형의 보정은 1982-1995년의 월평균 하천유량을 이용하였고 1996-2005년의 자료를 이용하여 검증하였다. 기후변화에 따른 수문 변동을 정량적으로 분석하기 위하여 1988-2002년을 기준시나리오 기간으로 설정하였으며 이산화탄소 농도, 기온, 강수의 변화에 따른 총 6개의 시나리오를 구성하였다. 시나리오 $1\sim6$은 수문 변화의 민감도를 나타내는 시나리오로 배증 이산화탄소를 반영하는 시나리오는 평균 11%의 하천유량 증가를 예측하였고, -42, -17, 17, 42%의 강수량 변화에 따라서는 -55, -24, 26, 65%의 하천유량 증감이 예측되었다.

• Smart Structures and Systems
• /
• 제24권6호
• /
• pp.745-757
• /
• 2019

Road surface deteriorations such as potholes have caused motorists heavy monetary damages every year. However, effective road condition monitoring has been a continuing challenge to road owners. Depth cameras have a small field of view and can be easily affected by vehicle bouncing. Traditional image processing methods based on algorithms such as segmentation cannot adapt to varying environmental and camera scenarios. In recent years, novel object detection methods based on deep learning algorithms have produced good results in detecting typical objects, such as faces, vehicles, structures and more, even in scenarios with changing object distances, camera angles, lighting conditions, etc. Therefore, in this study, a Deep Learning Pothole Detector (DLPD) based on the deep region-based convolutional neural network is proposed for autonomous detection of potholes from images. About 900 images with potholes and road surface conditions are collected and divided into training and testing data. Parameters of the network in the DLPD are calibrated based on sensitivity tests. Then, the calibrated DLPD is trained by the training data and applied to the 215 testing images to evaluate its performance. It is demonstrated that potholes can be automatically detected with high average precision over 93%. Potholes can be differentiated from manholes by training and applying a manhole-pothole classifier which is constructed using the convolutional neural network layers in DLPD. Repeated detection of the same potholes can be prevented through feature matching of the newly detected pothole with previously detected potholes within a small region.

• 인터넷정보학회논문지
• /
• 제24권6호
• /
• pp.1-11
• /
• 2023

Federated learning (FL) is a ground breaking machine learning paradigm that allow smultiple participants to collaboratively train models in a cloud environment, all while maintaining the privacy of their raw data. This approach is in valuable in applications involving sensitive or geographically distributed data. However, one of the challenges in FL is dealing with heterogeneous and non-independent and identically distributed (non-IID) data across participants, which can result in suboptimal model performance compared to traditionalmachine learning methods. To tackle this, we introduce FedGCD, a novel FL algorithm that employs Graph Neural Network (GNN)-based community detection to enhance model convergence in federated settings. In our experiments, FedGCD consistently outperformed existing FL algorithms in various scenarios: for instance, in a non-IID environment, it achieved an accuracy of 0.9113, a precision of 0.8798,and an F1-Score of 0.8972. In a semi-IID setting, it demonstrated the highest accuracy at 0.9315 and an impressive F1-Score of 0.9312. We also introduce a new metric, nonIIDness, to quantitatively measure the degree of data heterogeneity. Our results indicate that FedGCD not only addresses the challenges of data heterogeneity and non-IIDness but also sets new benchmarks for FL algorithms. The community detection approach adopted in FedGCD has broader implications, suggesting that it could be adapted for other distributed machine learning scenarios, thereby improving model performance and convergence across a range of applications.

• International Journal of Computer Science & Network Security
• /
• 제24권6호
• /
• pp.23-32
• /
• 2024

Mobile Ad hoc Network is a network of multiple wireless nodes which communicate and exchange information together without any fixed and centralized infrastructure. The core objective for the development of MANET is to provide movability, portability and extensibility. Due to infrastructure less network topology of the network changes frequently this causes many challenges for designing routing algorithms. Many routing protocols for MANET have been suggested for last few years and research is still going on. In this paper we review three main routing protocols namely Proactive, Reactive and Hybrid, performance comparison of Proactive such as DSDV, Reactive as AODV, DSR, TORA and Hybrid as ZRP in different network scenarios including dynamic network size, changing number of nodes, changing movability of nodes, in high movability and denser network and low movability and low traffic. This paper analyzes these scenarios on the performance evaluation metrics e.g. Throughput, Packet Delivery Ratio (PDR), Normalized Routing Load(NRL) and End To-End delay(ETE).This paper also reviews various network layer security attacks challenge by routing protocols, detection mechanism proposes to detect these attacks and compare performance of these attacks on evaluation metrics such as Routing Overhead, Transmission Delay and packet drop rates.

• Journal of Preventive Medicine and Public Health
• /
• 제50권2호
• /
• pp.83-90
• /
• 2017

Objectives: The aim of this study was to model the avoidable burden of the risk factors of road traffic crashes in Iran and to prioritize interventions to reduce that burden. Methods: The prevalence and the effect size of the risk factors were obtained from data documented by the traffic police of Iran in 2013. The effect size was estimated using an ordinal regression model. The potential impact fraction index was applied to calculate the avoidable burden in order to prioritize interventions. This index was calculated for theoretical, plausible, and feasible minimum risk level scenarios. The joint effects of the risk factors were then estimated for all the scenarios. Results: The highest avoidable burdens in the theoretical, plausible, and feasible minimum risk level scenarios for the non-use of child restraints on urban roads were 52.25, 28.63, and 46.67, respectively. In contrast, the value of this index for speeding was 76.24, 37.00, and 62.23, respectively, for rural roads. Conclusions: On the basis of the different scenarios considered in this research, we suggest focusing on future interventions to decrease the prevalence of speeding, the non-use of child restraints, the use of cell phones while driving, and helmet disuse, and the laws related to these items should be considered seriously.

• 인터넷정보학회논문지
• /
• 제24권6호
• /
• pp.107-117
• /
• 2023

사이버 공간에서 이루어지는 전투 행위가 군의 주요 임무체계 및 무기체계에 어떠한 영향을 미치는지를 평가할 수 있어야 한다. 사이버공격에 의한 임무영향을 사이버 M&S로 분석하기 위해서는 대상이 되는 임무체계와 사이버전 요소를 모델로 구축하고, 시뮬레이션을 위한 시나리오를 저작하여야 한다. 사이버전에 의한 임무영향 분석 관련 연구는 미국을 중심으로 많은 연구가 수행되었으며, 기존의 연구에서는 물리전장과 사이버전장에 대해 별개로 시나리오를 저작하였다. 임무영향 분석의 정확도를 높이기 위해서는 물리전장 모델과 사이버전장 모델을 결합한 시뮬레이션 환경을 구축하고, 임무 시나리오와 사이버공격/방어 시나리오를 통합해서 저작할 수 있어야 한다. 또한 물리전장과 사이버전장은 업무영역이 상이함을 고려하여 시나리오를 효율적으로 저작할 수 있는 방법이 필요하다. 본 논문에서는 임무체계 정보를 이용하여 시나리오 저작에 필요한 자료를 사전에 작성하고, 선작업된 자료를 이용하여 통합시나리오를 저작하는 방법을 제안한다. 제안한 방법은 시나리오 저작도구의 설계에 반영하여 개발하고 있으며, 제안 방법을 입증하기 위해 대화력전 분야의 통합시나리오 저작을 수행하였다. 향후, 제안한 방법을 반영한 시나리오 저작도구를 활용하면 임무영향 분석을 위한 통합시나리오를 짧은 시간에 쉽게 저작할 수 있게 될 것이다.

• 한국농공학회논문집
• /
• 제60권2호
• /
• pp.121-132
• /
• 2018

Along with climate change, it is reported that the extreme climate events such as severe drought could cause difficulties of agricultural water supply. To minimize such damages, it is necessary to secure the agricultural water resources by using or saving the amount of irrigation water efficiently. The objectives of this study were to develop paddy water management scenarios and to evaluate their effectiveness on water saving. Three water management scenarios (a) deep irrigation with ponding depth of 20~80 mm (control, CT), (b) no/intermittent irrigation until paddy cracks (water management A, WM-A), and (c) intermittent irrigation with ponding depth under 20 mm (water management B, WM-B) were developed. Water saving effects were analyzed using monitored data from experimental paddy fields, and agricultural water supply was analyzed on a reservoir-scale using MASA model. The observed irrigation amounts were reduced by 21 % and 17 % for WM-A and WM-B compared to CT, respectively, and mainly occurred by the increase of effective rainfall. The simulation results showed that water management scenarios could reduce irrigation by 21~51 % and total inflow by 10~24 % compared to CT. The long-term simulated water level change of agricultural reservoir resulted in the decrease of dead level occurrence for WM-A and WM-B. The study results showed that WT-A and WT-B have more benefit than CT in the aspect of agricultural reservoir water supply.

• 한국수자원학회논문집
• /
• 제56권11호
• /
• pp.737-749
• /
• 2023

본 연구에서는 18개 기후모형으로부터 도출된 SSP 기반의 미래 기후변화 시나리오와 유역모델링(SWAT-K)을 연계하여 제주도 지역의 미래 기후변화에 따른 수문학적 영향을 분석하였다. 기후모형에 따른 편차가 있으나 대체로 미래 기온의 증가에 따라 증발산량이 증가하고, 강수량의 증가로 인해 유출량과 함양량 또한 크게 증가할 것으로 전망되었다. 특히 SSP5-8.5 시나리오에서 이러한 증가가 더욱 뚜렷하게 나타났으며, 미래 후반기로 갈수록 GCM 모형 간의 차이가 크게 나타났다. 연평균 값에 대한 과거기간(1981~2010년) 대비 증감률로는 미래 후반기(2071~2100년)에 SSP5-8.5 시나리오에서 강수량 +21.4%, 증발산량 +19.2%, 유출량 +40.9%, 함양량 +16.6%의 전망을 나타내었다. 월별로 변화율을 보면 SSP5-8.5 시나리오에서 강수량은 9월에 24.5%, 증발산량은 4월에 34.1%, 유출량은 10월에 58.1%, 함양량은 9월에 33.8%까지 증가할 것으로 전망되었다. 또한 극한 기후 시나리오에 따른 전망을 위해, 미래에 최다 강수량을 예측한 CanESM5 모형과 최소 강수량을 예측한 ACCESS-ESM1-5 모형의 미래 기후자료를 사용하여 연평균 수문학적 변화를 비교하였다. 그 결과 강수량이 최대로 나타난 CanESM5 모형에서는 유출률과 함양률이 상대적으로 높게 나타난 반면, 강수량이 최저로 전망된 ACCESS-ESM1-5의 경우에는 증발산 비율이 높게 나타났다. 본 연구에서 적용한 기후변화 시나리오 기준으로 제주도 전체의 가용수자원은 증가한다고 전망할 수 있으나, 기후모형에 따라 계절별 지역별로 상이한 결과를 도출할 수 있기 때문에 가능한 다양한 시나리오를 활용한 종합적 분석과 대응방안이 필요하다고 생각된다.

• 로봇학회논문지
• /
• 제10권1호
• /
• pp.24-32
• /
• 2015

This paper introduces a graphical user interface design that is aimed to apply to the surveillance and security robot, which is the pilot program for the army unmanned light combat vehicle. It is essential to consider the activities of robot users under the changing security environment in order to design the efficient graphical user interface between user and robot to accomplish the designated mission. The proposed design approach firstly identifies the user activities to accomplish the mission in the standardized scenarios of military surveillance and security operation and then develops the hierarchy of the interface elements that are required to execute the tasks in the surveillance and security scenarios. The developed graphical user interface includes input control component, navigation component, information display component, and accordion and verified by the potential users from the various skilled levels with the military background. The assessment said that the newly developed user interface includes all the critical elements to execute the mission and is simpler and more intuitive compared to the legacy interface design that was more focused on the technical and functional information and informative to the system developing engineers rather than field users.

• 한국해양공학회지
• /
• 제27권1호
• /
• pp.24-30
• /
• 2013

The Korea Navy has been making constant efforts to apply M&S (modeling and simulation) to naval ship development, and the generalization of M&S for ship development is a trend. M&S for ship design is used for the V&V (verification and validation) of its design and operation, including design verification and ergonomic design that considers the crew using the Naval Ship Product Model. In addition, many parts of this M&S are repeatedly accomplished regardless of the kinds of ships. This study aims to standardize M&S, which repeatedly applies similar verifications for operation scenarios. A congestion assessment simulation for the major spaces of ships was the subject of the standardization based on the leading research results of various researchers, and a simulation automation solution was suggested. An information model using XML was proposed through the simulation automation concept, and a prototype system based on it was implemented. The usability was shown through a case study that verified the operability performance of the crew messroom.