• Biotechnology and Bioprocess Engineering:BBE
• /
• v.7 no.5
• /
• pp.275-280
• /
• 2002

In order to control glucose concentration during fed-batch culture for antibiotic production, we applied so called “software sensor” which estimates unmeasured variable of interest from measured process variables using software. All data for analysis were collected from industrial scale cultures in a pharmaceutical company. First, we constructed an estimation model for glucose feed rate to keep glucose concentration at target value. In actual fed-batch culture, glucose concentration was kept at relatively high and measured once a day, and the glucose feed rate until the next measurement time was determined by an expert worker based on the actual consumption rate. Fuzzy neural network (FNN) was applied to construct the estimation model. From the simulation results using this model, the average error for glucose concentration was 0.88 g/L. The FNN model was also applied for a special culture to keep glucose concentration at low level. Selecting the optimal input variables, it was possible to simulate the culture with a low glucose concentration from the data sets of relatively high glucose concentration. Next, a simulation model to estimate time course of glucose concentration during one day was constructed using the on-line measurable process variables, since glucose concentration was only measured off-line once a day. Here, the recursive fuzzy neural network (RFNN) was applied for the simulation model. As the result of the simulation, average error of RFNN model was 0.91 g/L and this model was found to be useful to supervise the fed-batch culture.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.97.1-97
• /
• 2001

One of the main problems to constitute control method on biologically oriented wastewater purification processes, e.g. activated sludge process, sequential batch reactor, rotating bio-disk contactor, etc. is that it is hard to control the target component directly. For instance, while biochemical oxygen demand, suspended solids, and chemical oxygen demand are the key components to check the process performance, one may not control them directly since they are the results of microbial activities related to numerous physiochemical factors. Therefore controllers for bioprocess should be designed to make favorable condition for microorganisms´ living, e.g. dissolved oxygen level favorable, mixed liquor suspended solids concentration suitable ...

• Bulletin of the Korean Space Science Society
• /
• 2008.10a
• /
• pp.34.2-34.2
• /
• 2008

The batch least square filter is widely used for ground estimations. However, in orbit determination (OD) under inaccurate initial conditions and few measurement data the performance by the batch least square filter can lead an unstable results. To complement weak part of the batch filter, the batch unscented transformation without any linearization process is developed by ACL (Astrodynamics and Control Laboratory) in YONSEI University. In this paper, the batch unscented transformation is introduced and applied to satellite orbit determination using Satellite Laser Ranging (SLR) data. Only range of the satellite measured from ground tracking stations is used for measurement data. The results of simulation test are compared with those of the weighted batch least square filter for various initial states errors (position and velocity). Simulation results show that the batch unscented transformation is comparable or slightly superior to batch least square filter in the orbit determination.

• Journal of Institute of Control, Robotics and Systems
• /
• v.13 no.2
• /
• 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.

• Environmental Engineering Research
• /
• v.19 no.4
• /
• pp.395-399
• /
• 2014

Operation characteristics of the sequencing batch reactor (SBR) process with electro-flotation (EF) as a solid liquid separation method (EF-SBR) were investigated. EF-SBR process showed excellent solid-liquid separation performance which enabled to separate biosolids from liquid phase within 30 min and to extend cyclic reaction time. Although influent organic loading rate was increased stepwise from 5 to 15 g COD/day, food to microorganisms (F/M) ratio could be maintained about 0.3 g COD/g VSS/day in EF-SBR because biomass concentration could be easily controlled at desired level by EF. However, it was impossible to increase biomass concentration at the same level in control SBR (C-SBR) process because solid-liquid separation by gravity settling showed a limitation at higher mixed liquor suspended solids (MLSS) concentration with 60 min of settling time. Total chemical oxygen demand (TCOD) removal efficiency of EF-SBR process was not decreased although influent organic loading rate became 3 times higher than initial value. However, it was seriously deteriorated in C-SBR process after increasing the rate over 10 g COD/day, which was accounted for insufficient organic removal by relatively higher food to microorganisms (F/M) ratio as well as biosolids wash-out by a limitation of gravity sedimentation.

• KSBB Journal
• /
• v.6 no.1
• /
• pp.35-43
• /
• 1991

The production of poly-$\beta$-hydroxybutyrate(PHB) from methanol by batch and fed-batch cultivations of Methylobacterium sp. GL-10 was studied. PHB accumulation was stimulated by the nutrients deficiency including, NH4+, SO42-, and K+. The nitrogen deficiency was the most critical factor for PHB accumulation. In batch cultivation, the maximum cell concentration and PHB content were 1.86g/l and 0.62g/l, respectively, with 1.0%(v/v) of methanol and 0.5g/1 of ammonium sulfate. The mass doubling time of Methylobacterum sp. GL-10 was in the range of 4-5 hrs. The cell growth and PHB accumulation were severely inhibited at the methanol concentration over than 2% (v/v). To overcome methanol Inhibition, constant feeding and intermittent feedillg fed-batch cultivations were adopted, using C/N molar ratio as a control factor. In constant feeding fed-batch process, cell concentration was increased up to 2.67g/1, and PHB yield was enhanced from 0.33 of batch culture to 0.53. The relatively low cell concentration was caused by methanol accumulated in culture broth at late growth phase. To prevent methanol accumulation and to maximize PHB production, DO-state intermittent fed-batch cultivation was attempted. The cell and PHB concentration was reached up to 4.55g/1 and 1.80g/1, respectively. It was possible to maintain methanol concentration low and also to feed nutrient of desired C/N molar ratio.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.11 no.3
• /
• pp.195-202
• /
• 1998

The LPCVD system of batch type for the massproduction of semiconductor fabrication has a problem of phosphorous concentration uniformity in the boat. In this paper we study an improvement of the uniformity for phosphorous concentration and sheet resistance. These property was improved by using the nitrogen process and modified long nozzle for gas injection tube in the doped polysilicon deposition system. The phosphorous concentration and its uniformity for polysilicon film are measured by XRF(X-ray Fluorescence) for the conventional process condition and nitrogen process. In conventional process condition, the phosphorous concentration, it uniformity and sheet resistance for polysilicon film are in the range of 3.8~5.4$\times$10\ulcorner atoms/㎤, 17.3% and 59~$\Omega$/ , respectively. For the case of nitrogen process the corresponding measurements exhibited between 4.3~5.3$\times$10\ulcorner atoms/㎤, 10.6% and 58~81$\Omega$/ . We find that in the nitrogen process the uniformity of phosphorous concentration improved compared with conventional process condition, however, the sheet resistance in the up zone of the boat increased about 12 $\Omega$/ . In modified long nozzle, the phosphorous concentration, its uniformity and sheet resistance for polysilicon films are in the range of 4.5~5.1$\times$10\ulcorner atoms/㎤, 5.3% and 60~65$\Omega$/ respectively. Annealing after $N_2$process gives the increment of grain size and the decrement of roughness. Modification of nozzle gives the increment of injection amount of PH$_3$. Both of these suggestion result in the stable phosphorous concentration and sheet resistance. The results obtained in this study are also applicable to process control of batch type system for memory device fabrication.

• Journal of Microbiology and Biotechnology
• /
• v.12 no.4
• /
• pp.687-691
• /
• 2002

To enhance the production of flavonoids [baicalin, wogonin-7-Ο-glucuronic acid (GA)], which are secondary metabolites of Scutellaria baicalensis Georgi(G.) plant cells, a multilayer perceptron control system was applied to regulate the substrate feeding in a fed-batch cultivation. The optimal profile for the substrate feeding rate in a fed-batch culture of S. baicalensis G. was determined by simulating a kinetic model using a genetic algorithm. Process variable profiles were then prepared for the construction of a multilayer perceptron controller that included massive parallelism, trainability, and fault tolerance. An error back-propagation algorithm was applied to train the multiplayer perceptron. The experimental results showed that neurocontrol incorporated with a genetic algorithm improved the flavonoid production compared with a simple fuzzy logic control system. Furthermore, the specific production yield and flavonoid productivity also increased.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10b
• /
• pp.1028-1031
• /
• 1996

Dynamic scheduling is very important in constructing CIM and improving productivity of chemical processing systems. Computation at the scheduling level requires mostly a long time to generate an optimal schedule, so it is difficult to immediately respond to actual process events in real-time. To solve these problems, we developed dynamic scheduling algorithms such as DSMM(Dynamic Shift Modification Method), PUOM(Parallel Unit Operation Method) and UVVM(Unit Validity Verification Method). Their main functions are to minimize the effects of unexpected disturbances such as process time variations and unit failure, to predict a makespan of the updated dynamic schedule and to modify schedule desirably in real-time responding to process time variations. As a result, the algorithms generate a new pertinent schedule in real-time which is close to the original schedule but provides an efficient way of responding to the variation of process environment. Examples in a shampoo production batch process illustrate the efficiency of the algorithms.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10b
• /
• pp.539-544
• /
• 1992

An iterative lear-rung control scheme is newly designed in tile frequency domain. Purposing for batch process control, a generic form of feedback-assisted first-order learning is considered first, and the inverse model-based learning algorithm is derived through convergence analysis in the frequency domain. To enhance the robustness of the proposed scheme, a filtered version is also presented. Performance of the proposed scheme is evaluated through numerical simulations.