• 한국컴퓨터정보학회논문지
• /
• 제23권12호
• /
• pp.11-19
• /
• 2018

In the Unmanned Aerial Vehicle Line-Of-Sight datalink system, there is a possibility that the communication line is disconnected because line of sight can not be secured by one antenna due to changes in position and posture of the air vehicle. In order to prevent this, both top and bottom of air vehicle are equipped with antennas. At this time, if the signal can be transmitted and received by switching to an antenna advantageous for securing the line of sight, communication disconnection can be minimized. The legacy antenna switching method has disadvantages such that diffraction, fading due to the surface or obstacles, interference and reflection of the air vehicle are not considered, or antenna switching standard is not clear. In this paper, we propose an airborne antenna switching method for improving the performance of UAV LOS datalink system. In the antenna switching method, the performance of each of the upper and lower parts of the mounted antenna according to the position and attitude of the air vehicle is predicted by using the deep learning in an UAV LOS datalink system in which only the antenna except the receiver is duplicated. Simulation using flying test dataset shows that it is possible to switch antennas considering the position and attitude of unmanned aerial vehicle in the datalink system.

• 한국표면공학회지
• /
• 제55권3호
• /
• pp.156-163
• /
• 2022

An ammonium ion with a size and charge similar to that of potassium can bind to valinomycin, which is used as an ion carrier for potassium, and cause a meaningful interference effect on the detection of potassium ions. Currently, there are few ion sensors that correct the interference effect of ammonium ions, and there are few studies that specifically suggest the mechanism of the interference effect. By fabricating a SPCE-based potassium ion-selective electrode, the electromotive force was measured in the concentration range of potassium in the nutrient solution, and the linear range was measured to be 10^-5 to 10^-2 M, and the detection limit was 10^-5.19 M. And the interference phenomenon of the potassium sensor was investigated in the concentration range of ammonium ions present in the nutrient solution. Therefore, a data-based analysis strategy using deep learning was presented as a method to minimize the interference effect.

• Nuclear Engineering and Technology
• /
• 제55권9호
• /
• pp.3277-3290
• /
• 2023

Nuclear power plant (NPP) operations with multiple objectives and devices are still performed manually by operators despite the potential for human error. These operations could be automated to reduce the burden on operators; however, classical approaches may not be suitable for these multi-objective tasks. An alternative approach is deep reinforcement learning (DRL), which has been successful in automating various complex tasks and has been applied in automation of certain operations in NPPs. But despite the recent progress, previous studies using DRL for NPP operations have limitations to handle complex multi-objective operations with multiple devices efficiently. This study proposes a novel DRL-based approach that addresses these limitations by employing a continuous action space and straightforward binary rewards supported by the adoption of a soft actor-critic and hindsight experience replay. The feasibility of the proposed approach was evaluated for controlling the pressure and volume of the reactor coolant while heating the coolant during NPP startup. The results show that the proposed approach can train the agent with a proper strategy for effectively achieving multiple objectives through the control of multiple devices. Moreover, hands-on testing results demonstrate that the trained agent is capable of handling untrained objectives, such as cooldown, with substantial success.

• Structural Engineering and Mechanics
• /
• 제84권3호
• /
• pp.323-335
• /
• 2022

The main idea of the framework is to seamlessly combine a reasonably accurate and fast surrogate model with the importance sampling strategy. Developing a surrogate model for predicting structures' dynamic responses is challenging because it involves high-dimensional inputs and outputs. For this purpose, a novel surrogate model based on cutting-edge deep learning architectures specialized for capturing temporal relationships within time-series data, namely Long-Short term memory layer and Transformer layer, is designed. After being properly trained, the surrogate model could be utilized in place of the finite element method to evaluate structures' responses without requiring any specialized software. On the other hand, the importance sampling is adopted to reduce the number of calculations required when computing the failure probability by drawing more relevant samples near critical areas. Thanks to the portability of the trained surrogate model, one can integrate the latter with the Importance sampling in a straightforward fashion, forming an efficient framework called TTIS, which represents double advantages: less number of calculations is needed, and the computational time of each calculation is significantly reduced. The proposed approach's applicability and efficiency are demonstrated through three examples with increasing complexity, involving a 1D beam, a 2D frame, and a 3D building structure. The results show that compared to the conventional Monte Carlo simulation, the proposed method can provide highly similar reliability results with a reduction of up to four orders of magnitudes in time complexity.

• Journal of Information Technology Applications and Management
• /
• 제29권6호
• /
• pp.81-93
• /
• 2022

As deep learning technology in natural language and visual processing has rapidly developed, collaborative filtering-based recommendation systems using deep learning technology are being actively introduced in the recommendation field. In this study, OCF-GAN, a hybrid collaborative filtering model using GAN, was proposed to solve the one-class and cold-start problems, and its usefulness was verified through performance evaluation. OCF-GAN based on conditional GAN consists of a generator that generates a pattern similar to the actual user preference pattern and a discriminator that tries to distinguish the actual preference pattern from the generated preference pattern. When the training is completed, user preference vectors are generated based on the actual distribution of preferred items. In addition, the cold-start problem was solved by using a hybrid collaborative filtering recommendation method that additionally utilizes user and item profiles. As a result of the performance evaluation, it was found that the performance of the OCF-GAN with additional information was superior in all indicators of the Top 5 and Top 20 recommendations compared to the existing GAN-based recommender. This phenomenon was more clearly revealed in experiments with cold-start users and items.

• 대한원격탐사학회지
• /
• 제36권6_2호
• /
• pp.1591-1604
• /
• 2020

본 연구는 고해상도 위성영상을 딥러닝 알고리즘에 적용하여 토지피복을 분류하고 공간객체별 알고리즘의 성능 검증에 대한 연구이다. 이를 Fully Convolutional Network계열의 알고리즘을 선정하였으며, Kompasat-3 위성영상, 토지피복지도 및 임상도를 활용하여 데이터셋을 구축하였다. 구축된 데이터셋을 알고리즘에 적용하여 각각 최적 하이퍼파라미터를 산출하였다. 하이퍼파라미터 최적화 이후 최종 분류를 시행하였으며, 전체 정확도는 DeeplabV3+가 81.7%로 가장 높게 산정되었다. 그러나 분류 항목별로 정확도를 살펴보면, 도로 및 건물에서 SegNet이 가장 우수한 성능을 나타내었으며, 활엽수, 논의 항목에서 U-Net이 가장 높은 정확도를 보였다. DeeplabV3+의 경우 밭과 시설재배지, 초지 등에서 다른 두 모델보다 우수한 성능을 나타내었다. 결과를 통해 토지피복 분류를 위해 하나의 알고리즘 적용에 대한 한계점을 확인하였으며, 향후 공간객체별로 적합한 알고리즘을 적용한다면, 높은 품질의 토지피복분류 결과를 산출할 수 있을 것으로 기대된다.

• 한국과학교육학회지
• /
• 제17권1호
• /
• pp.75-83
• /
• 1997

The purpose of this study was to determine the strategies that middle school students used in solving problems concerning density and solubility. These were compared in the aspects of problem contexts for 42 students of varying logical reasoning ability, spatial ability, and learning approach. A coding scheme used consists of five categories: reading & organization, production, errors, evaluation, and strategy. Students' protocols were analyzed after intercoder agreement had been established to be .95. The results were as follows: 1. Students had more difficulties in reading and organizing the problems in everyday contexts than in scientific contexts. Students at the concrete-operational stage and / or surface approach were more likely to have difficulties in reading and organizing the problems than those at the formal-operational stage and / or deep approach. 2. Students tended to split up the solubility problems into sub-problems and to solve the density problem in everyday contexts in random manner. These were significantly correlated with the test scores concerning logical reasoning ability, spatial ability, and learning approach at the .1 level of significance. 3. Major errors in solving the density problems were to disregard the given information or generated and to use inappropriate information. Many errors in solving the solubility problems were found to be executive errors. The strategy to use the information given appropriately was positively related to students' logical reasoning ability, spatial ability, and learning approach. 4. More evaluation strategies were found in everyday contexts. Their strategies to grasp the meaning of answers and to check the math were significantly related to students' logical reasoning ability. 5. Students used the random trial-and-error strategy more than the systematic strategy and the systematic trial-and-error strategy, especially in everyday contexts. The strategies used by the students were significantly related to students' logical reasoning ability, spatial ability, and learning approach.

• 영어어문교육
• /
• 제8권2호
• /
• pp.151-176
• /
• 2003

Motivation and learning strategies, some of the important factors affecting language learning, have mostly been studied with reference to their relationship in terms of proficiency. This study investigated motivation and learning strategies and their relationship in order to find the inward predisposition of learners. Data was collected from 200 university students in Taejon and Chungnam province, Korea language learning strategies were measured by the Strategy Inventory for Language Learning(SILL), and motivation by the Attitude / Motivation Test Battery(AMTB), with adaptations for Koreans. The detailed analysis of the data Indicated that Korean university students were more motivated to learn English for a practical goal than a formal one. They had a strong willingness to learn but showed 'the tendency of the new generation' of choosing the easiest and most convenient ways in studying English in terms of motivational intensity and strategy use. Findings imply that there have to be some changes and improvements in the deep-rooted classroom teaching methods. A systematic device is needed to induce students to be autonomous learners, providing them with a variety of activities suitable for their purposes and levels, as in opportunities of contacting native speakers, multi-media language labs, the Internet etc.

• 한국멀티미디어학회논문지
• /
• 제21권8호
• /
• pp.969-981
• /
• 2018

Recent advances in machine learning approaches such as deep neural network and reinforcement learning offer significant performance improvements in generating detailed and varied motions in physically simulated virtual environments. The optimization methods are highly attractive because it allows for less understanding of underlying physics or mechanisms even for high-dimensional subtle control problems. In this paper, we propose an efficient learning method for stochastic policy represented as deep neural networks so that agent can generate various energetic motions adaptively to the changes of tasks and states without losing interactivity and robustness. This strategy could be realized by our novel trajectory search method motivated by the trust region policy optimization method. Our value-based trajectory smoothing technique finds stably learnable trajectories without consulting neural network responses directly. This policy is set as a trust region of the artificial neural network, so that it can learn the desired motion quickly.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2022년도 제66차 하계학술대회논문집 30권2호
• /
• pp.7-10
• /
• 2022

본 연구는 국내 주식의 intraday 가격변화를 딥러닝 모형들로 예측하고 그 예측모형을 이용한 매매전략 딥러닝 모형을 제안한다. 주식의 intraday 가격변화에 따라서, 고빈도 매매, 주문집행문제 (order execution problem), 자동화 매매 등과 같은 intraday 주식 트레이딩의 수익률이 달라지기 때문에, 주식의 intraday 가격변화 예측은 주식 투자에 있어서 중요하다. 해외 시장에 대해서는 인공지능 등을 이용한 intraday 가격변화 예측 연구가 활발히 이루어졌지만, 국내의 경우 관련한 연구가 드물어 그 효용성이 명확히 드러나지 않았었다. 그에 따라서, KOSPI 50의 구성 종목에 대하여 정준의(canonical) 딥러닝 모형들을 적용하여 예측 성능을 비교한다. 또한, 그 예측모형들을 활용하여 간소화된 주문집행문제에서의 매매전략 딥러닝 모형을 제안한다. 그리고, 제안한 매매전략 딥러닝 모형을 KOSPI 50의 구성 종목에 대하여 실험하여, 제안한 방법론이 유효함을 밝힌다. 제시된 모형을 실제 주식 매매에 직접 적용하여 수익성을 개선을 기대할 수 있고, 사람이 직접 거래할지라도 효과적인 보조 지표가 될 수 있기에 본 논문은 실용적 의미를 지닌다.