• Journal of the Korean Society of Hazard Mitigation
• /
• v.11 no.2
• /
• pp.45-52
• /
• 2011

To improve fire-resistance of a high strength concrete against spalling under elevated temperature, fibers can be mixed to provide flow paths of evaporated water to the surface of concrete when heated. In this study, the experiment of a column under fire and mechanical loads is conducted and the material model for predicting temperature of reinforcement steel bar and mechanical behavior of fiber-mixed high strength concrete is suggested. The material model in previous studies is modified by incorporating physical behavior of internal concrete and thermal characteristics of concrete at the elevated temperature. Thermo-mechanical analysis of the fiber-mixed high strength concrete column is conducted using the calibrated material model. The performance of the proposed material model is confirmed by comparing thermo-mechanical analysis results with the experiment of a column under fire and mechanical loads.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.14 no.1
• /
• pp.257-264
• /
• 2019

This paper presents a single image haze removal method via a pixel-based joint BDCP (bright and dark channel prior) and a hierarchical bilateral filter in order to reduce computational complexity and memory requirement while improving the dehazing performance. Pixel-based joint BDCP reduces the computational complexity compared to the patch-based DCP, while making it possible to estimate the atmospheric light in pixel unit and the transmission more accurately. Moreover the bilateral filter, which can smooth an image effectively while preserving edges, refines the transmission to reduce the halo effects, and its hierarchical structure applied to edges only prevents the increase of complexity from the iterative application. Experimental results on various hazy images show that the proposed method exhibits excellent haze removal performance with low computational complexity compared to the conventional methods, and thus it can be applied in various fields.

• Journal of Internet of Things and Convergence
• /
• v.8 no.5
• /
• pp.17-25
• /
• 2022

Recently the massive amount of data has been generated because of the number of edge devices increases. And especially, the number of raw unstructured HTML documents has been increased. Therefore, MRC(Machine Reading Comprehension) in which a natural language processing model finds the important information within an HTML document is becoming more important. In this paper, we propose HTDE(HTML Tag Depth Embedding Method), which allows the BERT to train the depth of the HTML document structure. HTDE makes a tag stack from the HTML document for each input token in the BERT and then extracts the depth information. After that, we add a HTML embedding layer that takes the depth of the token as input to the step of input embedding of BERT. Since tokenization using HTDE identifies the HTML document structures through the relationship of surrounding tokens, HTDE improves the accuracy of BERT for HTML documents. Finally, we demonstrated that the proposed idea showing the higher accuracy compared than the accuracy using the conventional embedding of BERT.

• Journal of the Korean Geotechnical Society
• /
• v.38 no.5
• /
• pp.19-33
• /
• 2022

Newmark's sliding block analysis is the most commonly used method for predicting earthquake-induced permanent displacement of embankment slopes. Additionally, it yields the amount of slip circle sliding using the limit equilibrium theory. Thus, permanent displacement does not occur until the seismic load exceeds the yield acceleration, which induces sliding of the slip circle. The evolution of Newmark's sliding block analysis has been made by introducing the numerical seismic response analysis results since it was introduced. This study compares seismic performance evaluation results for the example enclosure dam section with the analysis methods. As a result, earthquake-induced permanent displacement using Newmark's sliding block analysis did not occur for the enclosure dam, indicating a high safety factor. However, nonlinear response history analysis gave reasonable results.

• Journal of Internet of Things and Convergence
• /
• v.8 no.3
• /
• pp.69-74
• /
• 2022

Big data related analysis techniques, the prescriptive analytics methodology improves the performance of passive learning models by ensuring that active learning secures high-quality learning data. Prescriptive analytics is a performance maximizing process by enhancing the machine learning models and optimizing systems through active learning to secure high-quality learning data. It is the best subscription value analysis that constructs the expensive category data efficiently. To expand the value of data by collecting research field, research propensity, and research activity information, customized researcher through prescriptive analysis such as predicting the situation at the time of execution after data pre-processing, deriving viable alternatives, and examining the validity of alternatives according to changes in the situation Provides research information service.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.8 no.3
• /
• pp.57-67
• /
• 2018

The ensemble is a unified approach used for getting better performance by using multiple algorithms in machine learning. In this paper, we introduce boosting and bagging, which have been widely used in ensemble techniques, and design a method using support vector regression, radial basis function network, Gaussian process, and multilayer perceptron. In addition, our experiment was performed by adding a recurrent neural network and MOHID numerical model. The drifter data used for our experimental verification consist of 683 observations in seven regions. The performance of our ensemble technique is verified by comparison with four algorithms each. As verification, mean absolute error was adapted. The presented methods are based on ensemble models using bagging, boosting, and machine learning. The error rate was calculated by assigning the equal weight value and different weight value to each unit model in ensemble. The ensemble model using machine learning showed 61.7% improvement compared to the average of four machine learning technique.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.419-419
• /
• 2022

최근 전 지구적 기후변화의 발생으로 수문현상의 규모와 빈도가 예측하기 어려운 수준으로 변화되고 있다. 이에 따라 정밀한 데이터를 활용한 수공구조물 운영 및 관리의 중요성이 대두되고 있다. 이 중에서도 다목적댐은 이·치수 측면에서 모두 활용되기 때문에 정밀한 댐 운영을 위한 댐 유입량 자료의 수집 및 관리가 필요하지만 현실적 한계로 인해 간접적으로 측정되고 있다. 현재 국내 다목적댐 저수지의 유입량은 댐시설 유지관리 기준(MW, 1994)에서 제시한 저수지 수위 변동량과 댐 방류량의 추정치로부터 계산하는 간접측정방법을 통해 산정되고 있다. 그러나 이와 같은 방법은 태풍이나 집중호우 등 대규모 홍수 발생 시 저수지 수위의 불균일성으로 인한 오차가 나타나며, 음유입량 및 톱니바퀴 형태의 자료가 발생하는 등 정확도 측면에서 한계가 있다. 따라서 본 연구에서는 한국건설기술연구원에서 2008년 개발한 물리적 기반의 분포형 유출해석 모형인 GRM(Grid based Rainfall-Runoff Model)을 활용하여 상류 유량관측소(옥동교 관측소, 영춘 관측소) 관측유량과 충주댐 지점 모의유량간의 경험공식을 도출하였으며, 이를 통해 상류 유량 관측소의 유량자료를 활용한 댐 유입량 직접산정이 가능하도록 하였다. 또한 다중 관측소 활용 시 댐 유입량 모의 성능이 개선되는지 여부를 확인하기 위해 3가지 경우(옥동교 관측소 단일, 영춘 관측소 단일, 옥동교·영춘 관측소 다중)로 구분하고 각 공식의 성능을 비교 평가하였다. 분석 결과 상류 관측소 관측유량과 댐 본체 지점의 모의유량이 비교적 높은 상관관계(0.79~0.96)를 보였으며, 단일 관측소를 활용한 공식 대비 다중 관측소를 활용한 공식이 더 높은 결정계수를 보였다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.6
• /
• pp.33-39
• /
• 2023

Recently, reinforcement learning has been used to improve the communication performance of flying ad-hoc networks (FANETs) and to design mobility models. Mobility model is a key factor for predicting and controlling the movement of unmmaned aerial vehicle (UAVs). In this paper, we designed and analyzed the performance of Q-learning with fourier basis function approximation and Deep-Q Network (DQN) models for optimal path finding in a three-dimensional virtual environment where UAVs operate. The experimental results show that the DQN model is more suitable for optimal path finding than the Q-learning model in a three-dimensional virtual environment.

• Journal of Korea Society of Industrial Information Systems
• /
• v.29 no.2
• /
• pp.15-25
• /
• 2024

For effective analysis of animal ecosystems, technology that can automatically identify the current status of animal habitats is crucial. Specifically, animal sound classification, which identifies species based on their sounds, is gaining great attention where video-based discrimination is impractical. Traditional studies have relied on a single deep learning model to classify animal sounds. However, sounds collected in outdoor settings often include substantial background noise, complicating the task for a single model. In addition, data imbalance among species may lead to biased model training. To address these challenges, in this paper, we propose an animal sound classification scheme that combines predictions from multiple models using Focal Loss, which adjusts penalties based on class data volume. Experiments on public datasets have demonstrated that our scheme can improve recall by up to 22.6% compared to an average of single models.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.12 no.4
• /
• pp.87-97
• /
• 2008

The purpose of this study is to analyse and compare experimentally flexural behavior of RC beams strengthened with CFRP plates by different methods, and finally develop the nonlinear analysis model with the aim of predicting the improving effects of structural capacity and the structural behaviors of RC beams. From this study, the characteristics of bond and flexural behavior of the prestressed CFRP plates were analyzed and examined. In deed, the beams were tested with experimental parameters of strengthening methods and prestressing level, and the developed analysis model was evaluated with the testing results. From this study, it is concluded that the developed analysis model have a good reliability and can be applied to the strengthening design of beams using CFRP plates.