• 환경위생공학
• /
• 제4권2호
• /
• pp.91-99
• /
• 1989

An anaerobic acidogenic fermentation experiment was carried out in order to investigate the distribution of volatile acid products and gas generations with varing temperatures and pH values. The experiment was carried out using $1\%$ glucose as substrate and a pair of 3.5 liter vessle as bench scale batch reactors. The reactors were operated for 7 days at 25, 30 and $35^{\circ}C$ and at pH values of 4.0, 4.5, 5.0, 5.5 and 6.0 at each temperature conditions. Major products at all experiment pH's at $35^{\circ}C$ were acetic acids and butyric acids which together composed around $90^{\circ}F$ of total product acids. At higher pH values at $35^{\circ}C$, propionic acid reached around $10\%$. At all experiment conditions, 52 to $55\%$ of generated gases comprised of hydrogen gas and 45 to $48\%$ of carbon dioxide. With temperature increase from 25 to $35^{\circ}C$, the production rate of acetic acid increased 2.9 fold, butyric acid 22 fold, hydrogen gas 2.0 fold and carbon dioxide gas 2.3 fold. Optimum reaction conditions for highest production of acetic acid and hydrogen gas was determined to be pH 5.5 at $35^{\circ}C$.

• 한국근적외분광분석학회:학술대회논문집
• /
• 한국근적외분광분석학회 2001년도 NIR-2001
• /
• pp.1126-1126
• /
• 2001

The paper presented here is the initial part of a larger study, in which it was determined which quality parameters in cheese powder could already be predicted by NIR at an early stage in the process and which could only be predicted at the final stages of the process. This initial study was performed in order to establish the levels and nature of variation within and between batches such that the subsequent data collection could be tackled optimally. The perspectives evolved into more than was originally planned and revealed some interesting uses of NIR-technology. Cheese powder production starts as a batch process, where waste cheese from other dairies is melted down in a vat. The process then turns into a continual process as the vat is emptied and the melted cheese is then filtered, homogenized, pasteurized and finally spray dried. Between each batch the powder is to a greater or lesser degree a mixture of 2 batches. This paper is divided into 2 aspects, one regarding the optimization of sampling time and the other is a study of process dynamics. Optimizing sampling time This initial study included 9 powder samples from 9 different batches produced during one day. The raw materials for the batches were chosen with the aim of creating a relatively high level of variation in the data. The total of 81 samples were taken out at regular intervals and spectra were collected on a NIR-systems 6500 instrument. The subsequent reduction of the data by PCA to score values shows the power of NIR as a tool to determine not only when samples are representative of a certain batch, but also which batches are stable enough to include in a further study. Studying process dynamics To take this experiment a step further 1 of the 81 samples were sent to the laboratory for further analyses. The samples were chosen on the criteria that they covered the spectral variation in the dataset. These samples were analysed for 4 chemical components and 5 physical attributes, which are essential for describing the quality of the product. The latent structure of the 7 samples, using the chemical and physical variables, is totally comparable to the latent structure of the NIR spectra. This outcome makes it possible to describe the dynamics of one day's production both chemically and physically with relatively little resources. Additionally it raises the question as to whether reference values are needed, as the latent structure of the NIR-spectra appears to be sufficient in providing information on the quality of the product. To be able to use NIR in this way would require defining quality limits in the principal component space as opposed to each of the reference values. The potential of NIR applied in an explorative fashion with batch processes opens a whole new gateway for the use of this technology. This study explains yet again after so many years in the field “why I'm crazy about NIR!”.

• 대한원격탐사학회지
• /
• 제36권2_1호
• /
• pp.199-215
• /
• 2020

잘피는 연안해역에 서식하는 해양수생관속식물로 해양생태계의 중요한 역할을 하고 있어, 주기적인 잘피 서식지의 모니터링이 이루어지고 있다. 최근 효율적인 잘피 서식지의 모니터링을 위해 고해상도의 영상 획득이 가능한 드론의 활용도가 높아지고 있다. 그리고 의미론적 분할에 있어 합성곱 신경망 기반의 딥러닝이 뛰어난 성능을 보임에 따라, 원격탐사 분야에 이를 적용한 연구가 활발하게 이루어지고 있다. 그러나 다양한 딥러닝 모델, 영상, 그리고 하이퍼파라미터에 의해 의미론적 분할의 정확도가 다르게 나타나고, 영상의 정규화와 타일과 배치 크기에서도 정형화되어 있지 않은 상태이다. 이에 따라 본 연구에서는 우수한 성능을 보여주는 딥러닝 모델을 이용하여 드론의 광학 영상에서 잘피 서식지를 분할하였다. 그리고 학습 자료의 정규화 및 타일의 크기를 중점으로 결과를 비교 및 분석하였다. 먼저 정규화와 타일, 배치 크기에 따른 결과 비교를 위해 흑백 영상을 만들고 흑백 영상을 Z-score 정규화 및 Min-Max 정규화 방법으로 변환한 영상을 사용하였다. 그리고 타일 크기를 특정 간격으로 증가시키면서 배치 크기는 메모리 크기를 최대한 사용할 수 있도록 하였다. 그 결과, Z-score 정규화가 적용된 영상이 다른 영상보다 IoU가 0.26 ~ 0.4 정도 높게 나타났다. 또한, 타일과 배치 크기에 따라 최대 0.09까지 차이가 나타나는 것을 확인하였다. 딥러닝을 이용한 의미론적 분할에 있어 정규화, 타일의 배치 크기의 변화에 따른 결과가 다르게 나타났다. 그러므로 실험을 통해 이들 요소에 대한 적합한 결정 과정이 있어야 함을 알 수 있었다.

• 한국컴퓨터정보학회논문지
• /
• 제29권9호
• /
• pp.1-7
• /
• 2024

본 논문은 MNIST 데이터셋을 활용한 손글씨 숫자 인식에서 합성곱 신경망(CNN)과 배치정규화(BN)를 결합한 모델을 제안한다. LeCun et al.의 LeNet-5 모델의 성과를 뛰어넘는 것을 목표로 6계층 신경망 구조를 설계하였다. 제안된 모델은 28×28 픽셀 이미지를 입력으로 받아 합성곱, 맥스 풀링, 완전연결계층을 거쳐 처리하며, 특히 배치정규화계층을 도입하여 학습 안정성과 성능을 향상시켰다. 실험에서는 60,000개의 훈련 이미지와 10,000개의 테스트 이미지를 사용하였으며, Momentum 최적화 알고리즘을 적용하였다. 모델 구성에서는 30개의 필터, 필터 사이즈 5×5, 패딩 0, 스트라이드 1을 사용하였고, ReLU 활성화 함수를 채택하였다. 훈련 과정에서는 미니배치 사이즈 100, 총 20 에포크, 학습률 0.1로 설정하였다. 결과적으로 제안된 모델은 99.22%의 테스트 정확도를 달성하여 LeNet-5의 99.05%를 상회하였으며, F1-score 0.9919를 기록하여 모델의 성능을 입증하였다. 또한, 본 논문에서 제안한 6계층 모델은 LeCun et al.의 LeNet-5(7계층 모델)와 Ji, Chun and Kim(10계층 모델)이 제안한 모델보다 더 단순한 구조로 모델의 효율성을 강조하였다. 본 연구의 결과는 AI 비전 검사기 등 실제 산업 응용에서 활용 가능성을 보여주며, 특히 스마트팩토리에서 부품의 불량 상태를 판별하는 데 효과적으로 적용될 수 있을 것으로 기대된다.

• 방사성폐기물학회지
• /
• 제5권2호
• /
• pp.113-122
• /
• 2007

본 연구에서는 타당한 지하조건을 모사하기 위한 실험장치를 글로버박스(Glove-box) 내에 설치하고 천연지하수 및 자연균열을 가진 화강암 시추코어를 이용하여 핵종이동 실험을 수행하였다. 암반코어의 균열을 통한 지하수 유동을 해석하기 위하여 비수착성 음이온 핵종인 Br로 지하수 유동실험을 수행하였다. 암반 균열을 통한 우라늄 이동 실험결과에서 유출된 우라늄의 파과곡선이 비수착성 핵종인 Br와 유사한 거동을 보여주었는데, 이는 주어진 지하수 조건에서 우라늄이 주로 탄산염과 결합된 음이온 복합체로 이동하기 때문인 것으로 추정된다. 아울러 균열충전광물에 대한 우라늄의 회분식 수착실험을 수행한 결과, 균열충전광물에 대한 우라늄의 분배계수 $K_d$는 약 2.7 mL/g로 낮게 나타났다. 이러한 우라늄 수착실험 결과는 빠른 유출을 보인 우라늄 이동실험 결과와 일치한다. 균열암반을 통한 우라늄 이동의 지 연 특성을 보다 자세히 분석하기 위하여 회분식 수착실험으로 부터 구한 $K_d$값을 이용해 지연인자 $R_d$값 $({\sim}16.2)$를 구하고 이동실험 결과로부터 구한 $R_d$값 $({\sim}14.3)$과 비교한 결과, 서로 매우 유사한 지연인자 값을 가진다는 것을 알 수 있었다. 이는 화강암 코어의 균열을 통한 우라늄의 이동 지연이 주로 균열충전광물에 의해 이루어지고 있음을 의미하는 것이라고 하겠다.

• 한국환경과학회지
• /
• 제25권9호
• /
• pp.1233-1239
• /
• 2016

This study aimed to evaluate changes in the TN and TP removal efficiencies, depending on whether or not a settling process is applied, in a sequencing batch reactor (SBR) process with a membrane bioreactor (MBR). Nutrient removal was considered in terms of developing an advanced water treatment system for ships in accordance with water quality standards set forth by 227(64). For these purposes, the TN and TP concentrations in the inflow and outflow water were measured to calculate the TN and TP removal efficiencies, depending on whether or not a settling process was used. Water discharged from a bathroom, which was constructed for the experiment, was used as the raw water. The experiment that included a settling process was conducted twice, and the operating conditions were: aeration for 90 min, settling for 30 min, agitation for 15 min, and settling for 15 min for one experiment; and aeration for 150 min, settling for 45 min, agitation for 15 min, and settling for 15 min in the other. Operating conditions for the experiment that did not include a settling process were: aeration for 180 min and agitation for 60 min. The concentration of the mixed liquor suspended solids (MLSS) in the reactor was 3,500 mg/L, while the aeration rate was 121 L/min and the water production rate was 1.5 L/min. For the two experiments where a settling process was applied, the average TN removal efficiencies were 44.39% and 41.05%, and the average TP removal efficiencies were 47.85% and 46.04%. For the experiment in which a settling process was not applied, the average TN removal efficiency was 65.51%, and the average TP removal efficiency was 52.51%. Although the final nutrient levels did not satisfy the water quality standards of MEPC 227(64), the TN and TP removal efficiencies were higher when a settling process was not applied.

• 한국환경과학회지
• /
• 제2권4호
• /
• pp.331-336
• /
• 1993

A mathematical model for organic removal efficiency was investigated in a fluidized bed biofilm reactor by changing the feed flow rate, the residence time and the recycle flow rate. In batch experiment, organic removal could be assumed as first order and an intrinsic first order rate constant(k1) was found $6.4{\times}^{-6}cm^3/mg{\cdot}sec$ at influent COD range of 3040 - 6620 mg/L. In continuous experiment, at the condition of the influent COD, 3040 mg/L, the superficial upflow velocity, 0.47 cm/sec, the biofilm thickness 336 ${\mu}m$ and the biofilm dry density 0.091 g/mL, the calculated COD removal efficiency from the mathematical model gave 60% which was very close to the observed value of 66 %. As the feed flow rate was increased, the COD removal efficiency was sharply decreased and at constant feed flow rate, the COD removal efficiency was decreased also as the residence time being decreased.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2004년도 추계학술대회 논문집
• /
• pp.364-368
• /
• 2004

Micro Cellular Plastics create a sensation at polymer industrial for lowering product cost & overcoming a lowering of mechanical intensity. This research based on the experiment of sound absorption & transmission characteristics inquire into acoustic property of Micro Cellular Plastics. This experiment clarify the change of cell foaming rate for foaming time and the change of sound absorption & transmission for foaming rate.

• 한국환경보건학회지
• /
• 제19권4호
• /
• pp.44-50
• /
• 1993

This paper has concentrated on estimating the improvement of sludge dewaterability for the application of ozone in sewage treatment plant sludge. The experiment for the study was conducted by batch reactor, contacting ozone (5.0 g O$_3$/hr/l) to waste sludge and measured the Specific Resistance to Filtrate (SRF) varying pH, pressure and reaction time of ozone. And then checked the dissolved solids concentration of flitrate. The results of experiment were as follows: 1. When the total solids concentration of excess sludge was 9, 000 mg/l, the optimum injection rate of ozone was 5.0 g O$_3$/hr/l, and then pressure was 50 cm Hg for the measuring SRF. 2. In the range of pH 3~5, the effect of ozone injection was excellent, but it was unsatisfactory in the range of pH 9~11. Therefore, the ozone injection by acidifying pH level was effective in improving the dewaterability of sludge. 3. It was estimated that the dissolved solids concentration of flitrate was increased in proportion to the injection rate of ozone.

• 상하수도학회지
• /
• 제18권4호
• /
• pp.470-479
• /
• 2004

The experimental conditions and relationships between parameters such as organic matter, aeration volume, aeration time, and precipitation time for the effective treatment of domestic wastewater were investigated. With the batch systems, the adsorption amount of unit microbe was measured with the change of MLSS concentration, precipitation time, and aeration amount. Theoretical adsorption amount of microbes was then numerically formulated by use of a SPSS multiple analysis as follows: $$Y=-0.0106(X_1)+0.07310(X_2)+42.705(X_3)+62.700$$ In this study, the amount of organisms to be removed in the range of MLSS concentration 2,000~4,500 mg/l were examined. In order to investigate the optimal condition of nitrification, the upper water in the biosorption stage was used as the initial experiment water. The results showed that the C/N ratio was 1.5 and the reaction time for the optimal nitrification was 1.5 hr. When the adsorption efficiency for microbe biosorption was 66%, the optimum denitrification efficiency was 83.3%. When the optimum parameters obtained from the batch experiment were applied to the lab-scale operation, the total retention time from the flow-in to flow-out was 10 hours and the removal efficiency was 93.8% for $COD_{cr}$ and 80.9% for TN. For the full-scale operation, the total retention time was 9.0 hours and the removal efficiency was 94.4% for BOD, 89.6% for $COD_{cr}$, 88.0% for TN, and 86.2% for TP.