• Food Science and Biotechnology
• /
• v.18 no.4
• /
• pp.939-947
• /
• 2009

In order to produce vinegar using onions, 12 acetic acid bacteria were screened from the juice of fallen peaches, and a strain showing the highest acetic acid productivity among them was selected and identified as Acetobacter tropicalis No. 22. The culture broth containing 2.5%(w/v) of initial sugar concentration showed maximum acetic production after 10 days of cultivation, and the acetic acid was produced at the highest rate and reached the maximum acidity after 2 to 6 days of cultivation when the residual sugar and the ethanol concentration were in the range of 1.6 to 2%(w/v) and 0.6 to 1.8%(v/v), respectively. Also optimum conditions for acetic acid production by response surface method using the fractional factorial design with 3 variables and 5 levels were involved with initial ethanol content of 4.67%(v/v), initial acidity of 0.03%, and initial glucose concentration of 2.35%(w/v) and predicted level of acetic acid production at these conditions was 3.77%.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.29 no.7 s.238
• /
• pp.921-929
• /
• 2005

This paper presents an accelerated life test for burnout of tungsten filament of incandescent lamp. From failure analyses of field samples, it is shown that their root causes are local heating or hot spots in the filament caused by tungsten evaporation and wire sag. Finite element analysis is performed to evaluate the effect of vibration and impact for burnout, but any points of stress concentration or structural weakness are not found in the sample. To estimate the burnout life of lamp, an accelerated life test is planned by using quality function deployment and fractional factorial design, where voltage, vibration, and temperature are selected as accelerating variables. We assumed that Weibull lifetime distribution and a generalized linear model of life-stress relationship hold through goodness of fit test and test for common shape parameter of the distribution. Using accelerated life testing software, we estimated the common shape parameter of Weibull distribution, life-stress relationship, and accelerating factor.

• Journal of Microbiology and Biotechnology
• /
• v.15 no.1
• /
• pp.66-71
• /
• 2005

Doxorubicin is a clinically important anticancer polyketide compound that is typically produced by Streptomyces peucetius var. caesius. To improve doxorubicin productivity by S. peucetius, a doxorubicin pathway-specific regulatory gene, dnrI, was cloned into a high-copy-number plasmid containing a catechol promoter system. The S. peucetius containing the recombinant plasmid exhibited approximately 9.5-fold higher doxorubicin productivity compared with the wild-type S. peucetius. The doxorubicin productivity by this recombinant S. peucetius strain was further improved through the optimization of culture media composition. Based on the Fractional Factorial Design (FFD), cornstarch, $K_2HPO_4$, and $MgSO_4$ were identified to be the key factors influencing doxorubicin productivity. The Response Surface Method (RSM) results based on 20 independent culture conditions with varying amounts of key factors predicted the highest theoretical doxorubicin productivity of 11.1 mg/l with corn starch of 46.33 g/l, $K_2HPO_4$ of 4.63 g/l, and $MgSO_4$ of 9.26 g/l. The doxorubicin productivity of the recombinant S. peucetius strain with the RSM-based optimized culture condition was experimentally verified to be 11.46 mg/l, which was approximately 30.8-fold higher productivity compared with the wild-type S. peucetius without culture media optimization.

• Korean journal of food and cookery science
• /
• v.23 no.1 s.97
• /
• pp.1-8
• /
• 2007

Extraction conditions were optimized using response surface methodology for preparing high-quality, hot water extracts from cultivated Chrysanthemum indicum L. petals. A fractional factorial design was applied to investigate the effects of added sample ratio ($X_1$), extraction temperature ($X_2$) and extraction time ($X_3$) variables on extract properties, such as, soluble solid contents ($Y_1$), sugar content ($Y_2$), ?E ($Y_3$), turbidity ($Y_4$), total phenol ($Y_5$) and DPPH ($Y_6$). Second-order models were employed to generate a 3-dimensional response surface for dependent variables and their coefficients of determination ($R^2$) ranged from 0.8408${\sim}$0.9914. The range of optimum conditions at $80^{\circ}C$ extraction for maximize characteristics of hot water extracts was 2.3${\sim}$2.7 g and 9.2${\sim}$11.2 hr.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.18 no.6
• /
• pp.652-657
• /
• 2009

MPW(Magnetic Pulse Welding) is a technologies for welding of metals by means of repulsive force on account of the interaction between electro-magnetic field of coil and current induced in outer pipe. These MPW is one of the most useful welding process of welding ability of the dissimilar metal in which cylindrical materials, such as pipe, tube. As the quality of a weld joint is strongly influenced by process parameters during the welding process and the success of the welding to evaluated according to the leakage pressure. Generally, the process parameters is magnetic pressure, the gap between outer pipe and inner pipe, and the ratio of thickness to diameter of pipe(D/T) in MPW. Therefore, the goal of this study was to explain the effect of parameters on the weld joint leakage pressure. For these purposes, FFD(Fractional Factorial Design) were used for the experiment. The measured data were analyzed by regression analysis and verification experiments with random condition were conducted to confirm the suggested experimental model.

• 제어로봇시스템학회:학술대회논문집
• /
• 1998.10a
• /
• pp.195-200
• /
• 1998

Since the neural network was introduced, significant progress has been made on data handling and learning algorithms. Currently, the most popular learning algorithm in neural network training is feed forward error back-propagation (FFEBP) algorithm. Aside from the success of the FFEBP algorithm, polynomial neural networks (PNN) learning has been proposed as a new learning method. The PNN learning is a self-organizing process designed to determine an appropriate set of Ivakhnenko polynomials that allow the activation of many neurons to achieve a desired state of activation that mimics a given set of sampled patterns. These neurons are interconnected in such a way that the knowledge is stored in Ivakhnenko coefficients. In this paper, the PNN model has been developed using the plasma enhanced chemical vapor deposition (PECVD) experimental data. To characterize the PECVD process using PNN, SiO$_2$films deposited under varying conditions were analyzed using fractional factorial experimental design with three center points. Parameters varied in these experiments included substrate temperature, pressure, RF power, silane flow rate and nitrous oxide flow rate. Approximately five microns of SiO$_2$were deposited on (100) silicon wafers in a Plasma-Therm 700 series PECVD system at 13.56 MHz.

• Journal of the Korean Society for Precision Engineering
• /
• v.25 no.6
• /
• pp.159-167
• /
• 2008

Cooling lubricants are used in machining operations in order to reduce friction at the tool-chip and tool-workpiece interfaces, cool both chip and tool, and remove chip. Furthermore, they influence a strong effect on the shearing mechanisms and, consequently, on the machined surface quality and tool wear. However, several researchers state that the costs related to cutting fluids is frequently higher than those related to cutting tools. Moreover, the cooling lubricants cause an increase in both worker's health and social problems related to their use and correct disposal. Therefore, many researchers have focused on the environmentally conscious machining technologies. One of the technologies is known as MQL(Minimum Quantity Lubrication) machining. In this paper, an experimental model to obtain the optimal cutting conditions in MQL turning was suggested, and the effects of cutting conditions on surface roughness and cutting force were analyzed. For these purposes, FFD (Fractional Factorial Design) and RSM (Response Surface Methods) were used for the experiment. Cutting force and surface roughness with different cutting conditions were measured through the external cylindrical turning of Al 6061 based on the experiment plan. The measured data were analyzed by regression analysis and verification experiments with random conditions were conducted to confirm the suggested experimental model.

• International Journal of Quality Innovation
• /
• v.3 no.2
• /
• pp.113-131
• /
• 2002

The relationships among multi-dimensional data (such as medical examination data) with ambiguity and variation are difficult to explore. The traditional approach to building a data classification system requires the formulation of rules by which the input data can be analyzed. The formulation of such rules is very difficult with large sets of input data. This paper first describes two classification approaches using back-propagation (BP) neural network and Mahalanobis distance (MD) classifier, and then proposes two classification approaches for multi-dimensional feature selection. The first one proposed is a feature selection procedure from the trained back-propagation (BP) neural network. The basic idea of this procedure is to compare the multiplication weights between input and hidden layer and hidden and output layer. In order to simplify the structure, only the multiplication weights of large absolute values are used. The second approach is Mahalanobis-Taguchi system (MTS) originally suggested by Dr. Taguchi. The MTS performs Taguchi's fractional factorial design based on the Mahalanobis distance as a performance metric. We combine the automatic thresholding with MD: it can deal with a reduced model, which is the focus of this paper In this work, two case studies will be used as examples to compare and discuss the complete and reduced models employing BP neural network and MD classifier. The implementation results show that proposed approaches are effective and powerful for the classification.

• Korean Chemical Engineering Research
• /
• v.54 no.1
• /
• pp.44-51
• /
• 2016

The powder core, conventionally fabricated from iron particles coated with insulator, showed large eddy current loss under high frequency, because of small specific resistance. To overcome the eddy current loss, the increase in the specific resistance of powder cores was needed. In this study, copper oxide coating onto electrically conductive iron particles was performed using a planetary ball mill to increase the specific resistance. Coating factors were optimized by the Response surface methodology. The independent variables were the CuO mass fraction, mill revolution number, coating time, ball size, ball mass and sample mass. The response variable was the specific resistance. The optimization of six factors by the fractional factorial design indicated that CuO mass fraction, mill revolution number, and coating time were the key factors. The levels of these three factors were selected by the three-factors full factorial design and steepest ascent method. The steepest ascent method was used to approach the optimum range for maximum specific resistance. The Box-Behnken design was finally used to analyze the response surfaces of the screened factors for further optimization. The results of the Box-Behnken design showed that the CuO mass fraction and mill revolution number were the main factors affecting the efficiency of coating process. As the CuO mass fraction increased, the specific resistance increased. In contrast, the specific resistance increased with decreasing mill revolution number. The process optimization results revealed a high agreement between the experimental and the predicted data ($Adj-R^2=0.944$). The optimized CuO mass fraction, mill revolution number, and coating time were 0.4, 200 rpm, and 15 min, respectively. The measured value of the specific resistance of the coated pellet under the optimized conditions of the maximum specific resistance was $530k{\Omega}{\cdot}cm$.

• Journal of Soil and Groundwater Environment
• /
• v.12 no.5
• /
• pp.73-85
• /
• 2007

In this study, the experiment of solid-phase microextraction (SPME) technique using GC/FID was conducted as a possible alternative to liquid-liquid extraction for the analysis of EGBE, DGBE, DBM and TGME in water, and also, an optimization condition of trace analysis for disel oxygenates including EGBE by the design of experiment (DOE) was described. Experiments used a fractional factorial design method followed by central composite design allowing optimization of a number of factors as well as statistical analysis of the results. The response surface analysis showed that the extraction efficiency could be represented by a second-order polynomial equation in which the salts concentration, extraction temperature, extraction time and sonication time are the major influences. Using DOE method, a new datadependent method was developed to improve the quantity of confidently analyzed disel oxygenates in water samples.