• 제어로봇시스템학회논문지
• /
• 제13권2호
• /
• pp.173-178
• /
• 2007

This paper is to establish an appropriate model for predicting the temperature decreases in the batch transferred from the refining process to the caster in steel-making companies. Mathematical modeling of the temperature decreases between the processes is difficult, since the reaction mechanism by which the temperature changes in a molten steel batch is dynamic, uncertain and complex. Three soft computing techniques are examined using the same data, namely the multiple regression, fuzzy regression, and neural net (NN) models. To compare the accuracy of these three models, a limited number of input variables are selected from those variables significantly affecting the temperature decrease. The results show that the difference in accuracy between the three models is not statistically significant. Nonetheless, the NN model is recommended because of its adaptive ability and robustness. The method presented in this paper allows the temperature decrease to be predicted without requiring any precise metallurgical knowledge.

• 한국생물공학회:학술대회논문집
• /
• 한국생물공학회 2002년도 생물공학의 동향 (X)
• /
• pp.515-518
• /
• 2002

The method of foam fractionation can be applied to enrich proteins from a dilute protein solution if the proteins are hydrophobic and foam. If a protein, such as invertase, is hydrophilic, a dilute solution containing this protein may not foam. In that case, a batch foam fractionation process may not be appropriate for recovering a concentrated solution of that protein. In this paper, various concentrations of invertase were added to a dilute solution containing bromelain (a hydrophobic protein), in order to determine how the presence of a hydrophilic protein can affect the recovery of the desired hydrophobic protein. The effect of invertase on bromelain recovery was studied here at an initial bulk solution pH of 5 and an air superficial velocity of 4.6 cm/s.

• KSBB Journal
• /
• 제6권1호
• /
• pp.15-25
• /
• 1991

The optimal glucose concentration for the high-density culture of recombinant yeasts was obtained using dynamic simulation. An adaptive and predictive algoritilm complimented by the rule base was proposed for the control of the fed-batch fermentation process. The measurement of process variables has relatively long sampling period and relatively long time delay characteristics. As one of the solution on these problems, prediction techniques and rule bases were added to a classical recursive identification and control algorithm. Rule bases were used in the determination of control input considering the difference between the predicted value and the measured value. A mathelnatical model was used in the estimation and interpretation of the changes of state variables and parameters. Better performances were obtained by employing the control algorithm proposed in the present study compared to the conventional adaptive control method.

• 한국지하수토양환경학회:학술대회논문집
• /
• 한국지하수토양환경학회 2006년도 총회 및 춘계학술발표회
• /
• pp.402-406
• /
• 2006

The physical and chemical characteristics of ALC were analyzed and showed 2.2 of specific gravity and 9.05 of pH. The results of leaching tests with standard method for soil and waste indicated heavy metals(Cu, Cd, Pb, $Cr^{6+}$) were under maximum concentration level. The anaerobic digestion sludge was attached in the surface of ALC within 90 hours. As the results of batch test, pH of the ALC and Bio-ALC were decreased from initial pH of ALC to 8.7 and 7.8 respectively Also, the concentration of heavy metals was rapidly eliminated in the solution with the batch test. The result of column experiment indicates that the removal efficiency of ALC was showed 66% of T-P, 60% of T-N, and 67% of CODcr. Also, removal efficiency of Bio-ALC was slightly higher than that of ALC in T-N (64%) and CODcr (74%).

• 한국물환경학회지
• /
• 제31권4호
• /
• pp.399-406
• /
• 2015

A sequencing batch reactor was operated under different pH conditions to see the influence of free ammonia (FA) and free nitrous acid (FNA) to microbial community on ammonium partial nitrification. Long-term influences of FA and FNA were evaluated by polymerase chain reaction-denaturing gradient gel electrophoresis and fluorescence in situ hybridization. Nitrite accumulation was successfully achieved at pH 8.2 and 6.3. The shifts in the microbial community were observed when influent ammonia concentration increased to 1 g $NH_4$-N/L at pH 8.2, and then when pH was dropped to 6.3. Both Nitrosomonas and Nitrosospira were selected during the startup of the reactor, and eventually became dominant members as ammonia-oxidizing bacteria. The results of molecular microbiological analysis strongly suggested that the composition of microbial community was changed according to the method used to control nitrite-oxidizing bacteria.

• Nuclear Engineering and Technology
• /
• 제9권1호
• /
• pp.15-31
• /
• 1977

중수형 원자력발전소의 가동중에 연료를 재장전하는 특성을 고려하여 새로운 핵연료 batch와 주기의 개념을 서정하고, 연속적인 에너지 계산방법으로 개발하여 핵주기비 계산관계식을 유도하였으며, 이러한 관계식들로서 중수형 원자로에 사용될 수 있는 전자계산기 코드 HWRCOST를 개발하였다. 이 코드로서 현재 우리나라에 건설중인 CANDU-PHWR의 전수명에 걸친 핵연료 주기비를 계산하였고 아울러 우라늄 원광비, 성형 가공비, 사용핵연료 보관처리비 및 발전소 가동율의 변화에 대한 핵연료 주기비의 감응도를 분석하였다.

• 공학논문집
• /
• 제1권1호
• /
• pp.31-40
• /
• 1997

ATM망의 모델링에서 발생되어지는 D-BMAP/Geo/1/K queue의 departure 프로세스를 k-state MMBP로 특성화하는 근사화방법을 제시하고, 그 근사화방법과 decomposition algorithm을 이용하여 셀손실을 갖는 tandem 구조의 한정용량 대기 네트워크를 분석하였다.

• 반도체디스플레이기술학회지
• /
• 제19권4호
• /
• pp.65-70
• /
• 2020

CNN's technologies that represent emotional detection include primitive CNN algorithms, deployment normalization, and drop-off. We present the methods and data of the three experiments in this paper. The training database and the test database are set up differently. The first experiment is to extract emotions using Batch Normalization, which complemented the shortcomings of distribution. The second experiment is to extract emotions using Dropout, which is used for rapid computation. The third experiment uses CNN using convolution and maxpooling. All three results show a low detection rate, To supplement these problems, We will develop a deep learning algorithm using feature extraction method specialized in image processing field.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2021년도 춘계학술대회
• /
• pp.547-549
• /
• 2021

인공지능 분석에서 모델을 만들고 이를 검증하는 과정은 이미 생성된 데이터를 가지고 수행하는 Batch Processing이기에 연산 처리시간이 필요한 작업이다. 우리는 주식이나 국방 정보와 같은 실시간으로 발생하는 데이터를 바로 앞에서 발생한 데이터를 가지고 실시간으로 모델을 세우고 검증하여 예측하는 것이 필요하다. 이를 위한 해결책으로, 인공지능 모델링 작업에 필요한 데이터를 시간 처리 순으로 분할하고 데이터를 여러 프로세스에서 분산 처리하는 기법을 적용하여 해결하였다.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2007년도 추계학술대회 논문집
• /
• pp.325-326
• /
• 2007