• Journal of Korea Water Resources Association
• /
• v.36 no.2
• /
• pp.161-172
• /
• 2003

This study has evaluated the stream gauge network with the main emphasis on if the current stream gauge network can catch the runoff characteristics of the basin. As the evaluation of the stream gauge network in this study does not consider a special purpose of a stream gauge, nor the effect from a hydraulic structure, it becomes an optimization of current stream gauge network under the condition that each stream gauge measures the natural runoff volume. This study has been applied to the Nam-Han River Basin for the optimization of total 31 stream gauge stations using the entropy concept. Summarizing the results are as follows. (1) The unit hydrograph representing the basin response from rainfall can be transferred into a probability density function for the application of the entropy concept to optimize the stream gauge network. (2) Accurate derivation of unit hydrographs representing stream gauge sites was found the most important part for the evaluation of stream gauge network, which was assured in this research by comparing the measured and derived unit hydrographs. (3) The Nam-Han River Basin was found to need at least 28 stream gauge stations, which was derived by considering both the shape of the unit hydrograph and the runoff volume. If considering only the shape of the unit hydrograph, the number of stream gauges required decreases to 23.

• Korean Chemical Engineering Research
• /
• v.56 no.2
• /
• pp.229-239
• /
• 2018

The purpose of this study was to effectively produce the biosugar from cell wall of lipid extracted microalgae (LEA) by using microwave-assisted pretreatment without enzymatic hydrolysis process. Response surface methodology (RSM) was applied to optimization of microwave-assisted pretreatment conditions for the production of biosugar based on enzyme-free process from LEA. Microwave power (198~702 W), extraction time (39~241 sec), and sulfuric acid (0~1.0 mol) were used as independent variables for central composite design (CCD) in order to predict optimum pretreatment conditions. It was noted that the pretreatment variables that affect the production of glucose (C6) and xylose (C5) significantly have been identified as the microwave power and extraction time. Additionally, the increase in microwave power and time had led to an increase in biosugar production. The superimposed contour plot for maximizing dependent variables showed the maximum C6 (hexose) and C5 (pentose) yields of 92.7 and 74.5% were estimated by the predicted model under pretreatment condition of 700 w, 185.7 sec, and 0.48 mol, and the yields of C6 and C5 were confirmed as 94.2 and 71.8% by experimental validation, respectively. This study showed that microwave-assisted pretreatment under low temperature below $100^{\circ}C$ with short pretreatment time was verified to be an effective enzyme free pretreatment process for the production of biosugar from LEA compared to conventional pretreatment methods.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.5
• /
• pp.479-487
• /
• 2021

In this paper, we propose an Sound Event Detection (SED) model using self-training based on a noisy student model. The proposed SED model consists of two stages. In the first stage, a mean-teacher model based on an Residual Convolutional Recurrent Neural Network (RCRNN) is constructed to provide target labels regarding weakly labeled or unlabeled data. In the second stage, a self-training-based noisy student model is constructed by applying different noise types. That is, feature noises, such as time-frequency shift, mixup, SpecAugment, and dropout-based model noise are used here. In addition, a semi-supervised loss function is applied to train the noisy student model, which acts as label noise injection. The performance of the proposed SED model is evaluated on the validation set of the Detection and Classification of Acoustic Scenes and Events (DCASE) 2020 Challenge Task 4. The experiments show that the single model and ensemble model of the proposed SED based on the noisy student model improve F1-score by 4.6 % and 3.4 % compared to the top-ranked model in DCASE 2020 challenge Task 4, respectively.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.4
• /
• pp.611-620
• /
• 2021

The maintenance industry is mainly progressing based on condition-based maintenance after corrective maintenance and preventive maintenance. In condition-based maintenance, maintenance is performed at the optimum time based on the condition of equipment. In order to find the optimal maintenance point, it is important to accurately understand the condition of the equipment, especially the remaining useful life. Thus, using simulation data (C-MAPSS), a prediction model is proposed to predict the remaining useful life of a turbofan engine. For the modeling process, a C-MAPSS dataset was preprocessed, transformed, and predicted. Data pre-processing was performed through piecewise RUL, moving average filters, and standardization. The remaining useful life was predicted using principal component analysis and the k-NN method. In order to derive the optimal performance, the number of principal components and the number of neighbor data for the k-NN method were determined through 5-fold cross validation. The validity of the prediction results was analyzed through a scoring function while considering the usefulness of prior prediction and the incompatibility of post prediction. In addition, the usefulness of the RUL prediction model was proven through comparison with the prediction performance of other neural network-based algorithms.

• Journal of the Korean Society of Radiology
• /
• v.14 no.7
• /
• pp.991-1001
• /
• 2020

Supervised deep learning technologies for improving the image quality of computed tomography (CT) need a lot of training data. When input images have different characteristics with training images, the technologies cause structural distortion in output images. In this study, an imaging model based on the deep reinforcement learning (DRL) was developed for overcoming the drawbacks of the supervised deep learning technologies and reducing noise in CT images. The DRL model was consisted of shared, value and policy networks, and the networks included convolutional layers, rectified linear unit (ReLU), dilation factors and gate rotation unit (GRU) in order to extract noise features from CT images and improve the performance of the DRL model. Also, the quality of the CT images obtained by using the DRL model was compared to that obtained by using the supervised deep learning model. The results showed that the image accuracy for the DRL model was higher than that for the supervised deep learning model, and the image noise for the DRL model was smaller than that for the supervised deep learning model. Also, the DRL model reduced the noise of the CT images, which had different characteristics with training images. Therefore, the DRL model is able to reduce image noise as well as maintain the structural information of CT images.

• Applied Chemistry for Engineering
• /
• v.22 no.4
• /
• pp.358-366
• /
• 2011

A homologous series of linear liquid crystal dimers, the ${\alpha},{\omega}$-bis(4-cyano-azobenzene-4'-oxy)alkanes (CATWETn, where n, the number of methylene units in the spacer, is 2~10) were synthesized, and their thermotropic liquid crystalline phase behavior were investigated. The CATWETn with n of 3 and 6 exhibited monotropic nematic phases, whereas other derivatives showed enantiotropic nematic phases. The nematic-isotropic transition temperatures of the dimers and their entropy variation at the phase transition showed a large odd-even effect as a function of n. This phase transition behavior was rationalized in terms of the change in the average shape of the spacer on varying the parity of the spacer. The thermal stability and degree of order in the nematic phase and the magnitude of the odd-even effect of CATWETn were similar to those for the methoxy-, nitro-, and pentyl-substituted dimers, while they were significantly different from those for the monomesogenic compounds, 1-{4-(4'-cyanophenylazo)phenoxy}alkylbromides and the side-chain liquid-crystalline polymers, the poly[1-{4-(4'-cyanophenylazo)phenoxyalkyloxy}ethylene]s. The results were discussed in terms of 'virtual trimer model' by Imrie.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.33 no.3
• /
• pp.101-109
• /
• 2021

Wave observations using a marine X-band radar are conducted by analyzing the backscattered radar signal from sea surfaces. Wave parameters are extracted using Modulation Transfer Function obtained from 3D wave number and frequency spectra which are calculated by 3D FFT of time series of sea surface images (42 images per minute). The accuracy of estimation of the significant wave height is, therefore, critically dependent on the quality of radar images. Wave observations during Typhoon Maysak and Haishen in the summer of 2020 show large errors in the estimation of the significant wave heights. It is because of the deteriorated radar images due to raindrops falling on the sea surface. This paper presents the algorithm developed to increase the accuracy of wave heights estimation from radar images by adopting convolution neural network(CNN) which automatically classify radar images into rain and non-rain cases. Then, an algorithm for deriving the Hs is proposed by creating different ANN models and selectively applying them according to the rain or non-rain cases. The developed algorithm applied to heavy rain cases during typhoons and showed critically improved results.

• KOREAN JOURNAL OF PACKAGING SCIENCE & TECHNOLOGY
• /
• v.28 no.2
• /
• pp.97-104
• /
• 2022

This study investigated the characteristics for the optimal concentration of addition of the mixing solution through the corn husk pulverization and surface modification of biomass byproducts adding mixed solution between ESO and silane. And surveyed the specific surface area, water absorption, particle size and physical properties of bio- degradable plastic sheet. The specific surface area was 1.105 m²/g, particle size was the highest at 19 ㎛. The impact strength, tensile strength, elongation and hardness of plastic sheet showed the highest at the 1% concentration among the mixing solutions. The flexural strength and modulus was high according to the increasing the mixing solution. The results above showed that it was the best the adding 1% of mixed solution after silane treatment of corn husks for its manufacture as a bio-based plastic sheet.

• Journal of IKEEE
• /
• v.27 no.3
• /
• pp.340-344
• /
• 2023

In this paper, we propose a study on the development of deep learning structure to secure visibility of outdoor LED display board according to weather change. The proposed technique secures the visibility of the outdoor LED display board by automatically adjusting the LED luminance according to the weather change using deep learning using an imaging device. In order to automatically adjust the LED luminance according to weather changes, a deep learning model that can classify the weather is created by learning it using a convolutional network after first going through a preprocessing process for the flattened background part image data. The applied deep learning network reduces the difference between the input value and the output value using the Residual learning function, inducing learning while taking the characteristics of the initial input value. Next, by using a controller that recognizes the weather and adjusts the luminance of the outdoor LED display board according to the weather change, the luminance is changed so that the luminance increases when the surrounding environment becomes bright, so that it can be seen clearly. In addition, when the surrounding environment becomes dark, the visibility is reduced due to scattering of light, so the brightness of the electronic display board is lowered so that it can be seen clearly. By applying the method proposed in this paper, the result of the certified measurement test of the luminance measurement according to the weather change of the LED sign board confirmed that the visibility of the outdoor LED sign board was secured according to the weather change.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.219-219
• /
• 2021

서해안 중부 아산만 안쪽에 위치하는 평택·당진항에서 장래 개발 예정인 면적 6.9km²의 내항2공구 수역은 내항2공구 외곽호안 - 내항가호안 - 내항2공구 중앙 분리호안으로 둘러싸여 있으며, 투수성 제체인 내항가호안 사석 공극을 통하여 해수가 유통되어 조석 현상이 나타나고 있다. 2020년 8~9월의 2개월간 내항2공구 외곽호안 내·외측에서 조석 관측 결과, 2공구 수역의 최대 조차는 1.97m로서 외측 해역 최대 조차 9.79m의 20.1%이고 내·외측의 순간 수위차는 최대 5.82m에 달한다. 내항가호안은 내항2공구 개발이 거의 완료되는 시기까지 유지될 예정이므로 2공구 개발에 따른 내측 조차와 내·외측 수위차의 변화를 정확하게 예측하는 것은 내항가호안 제체 안전에 매우 중요하다. 이 연구의 목적은 장래 개발단계별 변화 예측에 앞서, 관측이 이루어진 2개월간의 실시간 내측 조석과 내·외측 수위차 시계열을 Delft3D-Flow를 이용하여 기 구축된 아산만 수치모델에서 재현하는 것이다. 내항가호안 제체 통과 유량은 내·외측 수위차에 비례하는 것으로 가정하고, 수위차 - 유량 관계식을 도출하였다. 수위차는 평택 조위관측소와 내항2공구 수역의 1분 간격 관측 조위로부터 산출하였고, 제체 통과 유량은 내측 조위(z, 평택항 DL 기준, m) - 수용적(V, 10⁶m³) 관계식으로 계산하였다. 내측 조위 - 수용적 관계식은 수심측량 성과로부터 V = 0.28z² + 3.73z + 2.96 (r²=1.00)으로 얻어졌다. 다양한 함수식의 적합성을 검토한 결과, 다음과 같은 수위차(𝚫z, m) - 제체 통과 유량(Q, m³/s) 관계식을 도출하였다. [내항가호안 내측으로 유입시] $Q_{IN}=\{\begin{array}{lll}{\exp}\{0.54\;{\ln}({\Delta}z)+6.00\}&&\text{; }{\Delta}z{\leq}1.8\\219.82{\Delta}z+158.56&&\text{; }{\Delta}z>1.8\end{array}\;\;(r^2=0.86)$ [내항가호안 외측으로 유출시] Q_OUT = -exp{0.44 ln(-𝚫z) + 5.70} (r²=0.59) 매 𝚫t 마다 제체 통과 유량을 계산하는 알고리즘을 Delft3D 소스 코드에 추가하고, 8개 분조 합성조석(M₂, S₂, K₁, O₁, N₂, K₂, P₁, Q₁)을 외력조건으로 설정하여 2개월간 조석 수치모델링을 수행하였다. 내항2공구 수역의 매 시별 조위 관측치와 모델치를 비교한 결과, 오차는 -0.37~0.37m의 범위이고, 오차 평균은 0.02m, 절대오차 평균은 0.08m로 상당히 정확하게 실시간 조위 변동을 모의하였다. 보정·검정된 이 모델을 이용하여 향후 내항2공구 개발에 따른 내측 조석과 내·외측 수위차 변화에 대한 예측모의를 진행할 예정이다.