• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2021년도 학술발표회
• /
• pp.136-136
• /
• 2021

In this study, a deep convolutional neural network (DCNN) model is proposed for short-term precipitation forecasting using weather radar-based images. The DCNN model is a combination of convolutional neural networks, autoencoder neural networks, and U-net architecture. The weather radar-based image data used here are retrieved from competition for rainfall forecasting in Korea (AI Contest for Rainfall Prediction of Hydroelectric Dam Using Public Data), organized by Dacon under the sponsorship of the Korean Water Resources Association in October 2020. This data is collected from rainy events during the rainy season (April - October) from 2010 to 2017. These images have undergone a preprocessing step to convert from weather radar data to grayscale image data before they are exploited for the competition. Accordingly, each of these gray images covers a spatial dimension of 120×120 pixels and has a corresponding temporal resolution of 10 minutes. Here, each pixel corresponds to a grid of size 4km×4km. The DCNN model is designed in this study to provide 10-minute predictive images in advance. Then, precipitation information can be obtained from these forecast images through empirical conversion formulas. Model performance is assessed by comparing the Score index, which is defined based on the ratio of MAE (mean absolute error) to CSI (critical success index) values. The competition results have demonstrated the impressive performance of the DCNN model, where the Score value is 0.530 compared to the best value from the competition of 0.500, ranking 16^th out of 463 participating teams. This study's findings exhibit the potential of applying the DCNN model to short-term rainfall prediction using weather radar-based images. As a result, this model can be applied to other areas with different spatiotemporal resolutions.

• 정보보호학회논문지
• /
• 제32권3호
• /
• pp.501-511
• /
• 2022

2021년에는 코로나의 여파로 랜섬웨어를 활용한 공격이 유행했으며 그 수는 매년 급증하고 있다. 그 중 파워쉘은 랜섬웨어에 주요 기술로 사용되고 있어 파워쉘 기반 악성코드 탐지 기법의 필요성은 증가하고 있으나 기존의 탐지기법은 난독화가 적용된 스크립트를 탐지하지 못하거나 역난독화에 시간이 오래 소요되는 한계가 존재한다. 이에 본 논문에서는 간단하고 빠른 역난독화 처리과정, Word2Vec과 CNN(Convolutional Neural Network)으로 구성되어 스크립트의 의미를 학습하고 특징을 추출해 악성 여부를 판단할 수 있는 딥러닝 기반의 분류 모델을 제안한다. 2021 사이버보안 AI/빅데이터 활용 경진대회의 AI 기반 파워쉘 악성 스크립트 탐지 트랙에서 제공된 1400개의 악성코드와 8600개의 정상 스크립트를 이용하여 제안한 모델을 테스트한 결과 기존보다 5.04배 빠른 역난독화 실행시간, 100%의 역난독화 성공률, 0.01의 FPR(False Positve Rate), 0.965의 TPR(True Positive Rate)로 악성코드를 빠르고 효과적으로 탐지함을 보인다.

• 정보보호학회논문지
• /
• 제32권4호
• /
• pp.691-708
• /
• 2022

머신러닝과 딥러닝의 기술이 보편화되면서 산업제어시스템의 이상(비정상) 탐지 연구에도 적용이 되기 시작하였다. 국내에서는 산업제어시스템의 이상 탐지를 위한 인공지능 연구를 활성화시키기 위하여 HAI 데이터셋을 개발하여 공개하였고, 산업제어시스템 보안위협 탐지 AI 경진대회를 시행하고 있다. 이상 탐지 연구들은 대개 기존의 딥러닝 학습 알고리즘을 변형하거나 다른 알고리즘과 함께 적용하는 앙상블 학습 모델의 방법을 통해 향상된 성능의 학습 모델을 만드는 연구가 대부분 이었다. 본 연구에서는 학습 모델과 데이터 전처리(pre-processing)의 개선을 통한 방법이 아니라, 비정상 데이터를 탐지하여 라벨링 한 결과를 보정하는 후처리(post-processing) 방법으로 이상 탐지의 성능을 개선시키는 연구를 진행하였고, 그 결과 기존 모델의 이상 탐지 성능 대비 약 10%이상의 향상된 결과를 확인하였다.

• 한국분말야금학회:학술대회논문집
• /
• 한국분말야금학회 2001년도 추계학술강연 및 발표대회
• /
• pp.7-7
• /
• 2001

The increasing interest in light weight materials coupled to the need for cost -effective processing have combined to create a significant opportunity for aluminum P/M. particularly in the automotive industry in order to reduce fuel emissions and improve fuel economy at affordable prices. Additional potential markets for Al PIM parts include hand tools. Where moving parts against gravity represents a challenge; and office machinery, where reciprocating forces are important. Aluminum PIM adds light weight, high compressibility. low sintering temperatures. easy machinability and good corrosion resistance to all advantages of conventional iron bm;ed P/rv1. Current commercial alloys are pre-mixed of either the AI-Si-Mg or AL-Cu-Mg-Si type and contain 1.5% ethylene bis-stearamide as an internal lubricant. The powder is compacted in closed dies at pressure of 200-500Mpa and sintered in nitrogen at temperatures between $580~630^{\circ}C$ in continuous muffle furnace. For some applications no further processing is required. although most applications require one or more secondary operations such as sizing and finishing. These sccondary operations improve the dimension. properties or appearance of the finished part. Aluminum is often considered difficult to sinter because of the presence of a stable surface oxide film. Removal of the oxide in iron and copper based is usually achieved through the use of reducing atmospheres. such as hydrogen or dissociated ammonia. In aluminum. this occurs in the solid st,lte through the partial reduction of the aluminum by magncsium to form spinel. This exposcs the underlying metal and facilitates sintering. It has recently been shown that < 0.2% Mg is all that is required. It is noteworthy that most aluminum pre-mixes contain at least 0.5% Mg. The sintering of aluminum alloys can be further enhanced by selective microalloying. Just 100ppm pf tin chnnges the liquid phase sintering kinetics of the 2xxx alloys to produce a tensile strength of 375Mpa. an increilse of nearly 20% over the unmodified alloy. The ductility is unnffected. A similar but different effect occurs by the addition of 100 ppm of Pb to 7xxx alloys. The lend changes the wetting characteristics of the sintering liquid which serves to increase the tensile strength to 440 Mpa. a 40% increase over unmodified aIloys. Current research is predominantly aimed at the development of metal matrix composites. which have a high specific modulus. good wear resistance and a tailorable coefficient of thermal expnnsion. By controlling particle clustering and by engineering the ceramic/matrix interface in order to enhance sintering. very attractive properties can be achicved in the ns-sintered state. I\t an ils-sintered density ilpproaching 99%. these new experimental alloys hnve a modulus of 130 Gpa and an ultimate tensile strength of 212 Mpa in the T4 temper. In contest. unreinforcecl aluminum has a modulus of just 70 Gpa.