• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.5
• /
• pp.785-796
• /
• 2023

In this study, a methodology for real-time classification and prediction of defects that may appear in PVDF(Polyvinylidene fluoride) sensors, which are widely used for structural integrity monitoring, is proposed. The types of sensor defects appearing according to the sensor attachment environment were classified, and an impact test using an impact hammer was performed to obtain an output signal according to the defect type. In order to cleary identify the difference between the output signal according to the defect types, the time domain statistical features were extracted and a data set was constructed. Among the machine learning based classification algorithms, the learning of the acquired data set and the result were analyzed to select the most suitable algorithm for detecting sensor defect types, and among them, it was confirmed that the highest optimization was performed to show SVM(Support Vector Machine). As a result, sensor defect types were classified with an accuracy of 92.5%, which was up to 13.95% higher than other classification algorithms. It is believed that the sensor defect prediction technique proposed in this study can be used as a base technology to secure the reliability of not only PVDF sensors but also various sensors for real time structural health monitoring.

• Smart Media Journal
• /
• v.13 no.4
• /
• pp.33-48
• /
• 2024

With the advent of 5G, characterized by Enhanced Mobile Broadband (eMBB), Ultra-Reliable Low Latency Communications (URLLC), and Massive Machine Type Communications (mMTC), efficient network management and service provision are becoming increasingly critical. This paper proposes a novel approach to address key challenges of 5G networks, namely ultra-high speed, ultra-low latency, and ultra-reliability, while dynamically optimizing network slicing and resource allocation using machine learning (ML) and deep learning (DL) techniques. The proposed methodology utilizes prediction models for network traffic and resource allocation, and employs Federated Learning (FL) techniques to simultaneously optimize network bandwidth, latency, and enhance privacy and security. Specifically, this paper extensively covers the implementation methods of various algorithms and models such as Random Forest and LSTM, thereby presenting methodologies for the automation and intelligence of 5G network operations. Finally, the performance enhancement effects achievable by applying ML and DL to 5G networks are validated through performance evaluation and analysis, and solutions for network slicing and resource management optimization are proposed for various industrial applications.

• Food Science and Preservation
• /
• v.17 no.2
• /
• pp.281-289
• /
• 2010

We determined the optimum ethanolic conditions for extraction of $\gamma$-oryzanol and other functional components from rice bran, using response surface methodology (RSM). A central composite design was used to investigate the effects of the independent variables of solvent ratio ($X_1$), extraction temperature ($X_2$), and extraction time ($X_3$), on dependent variables including yield ($Y_1$), total phenolic content ($Y_2$), electron-donating activity ($Y_3$), ferulic acid level ($Y_4$), and $\gamma$-oryzanol concentration ($Y_5$). Solvent ratio and extraction temperature were the most important factors in extraction. The maximum yield was at 22.56 mL/g ($X_1$), 78.19C ($X_2$), and 522.15 min ($X_3$), at the saddle point. Total phenolic levels were little affected by solvent ratio or extraction temperature. The maximum concentration of extracted total phenolics was 90.78mg GAE/100 g at 21.26 mL/g, $94.65^{\circ}C$, and 567.97 min. A maximum electron-donating ability of 54.72% was obtained with the parameters 20.20 mL/g,$81.89^{\circ}C$, and 701.87 min, at the highest point. The maximum level of ferulic acid components was 210.47 mg/100g at 5.22 mL/g, $79.66^{\circ}C$, and 575.24 min. In addition, the maximum $\gamma$-oryzanol concentration was 660.39 mg/100g at 5.10 mL/g, $81.83^{\circ}C$, and 587.39 min. The optimum extraction conditions were a solvent ratio of 10.45 mL/g, $80^{\circ}C$ extraction temperature, and 535 min extraction time. Predicted extraction levels under optimized conditions were in line with experimental values.

• Journal of Life Science
• /
• v.23 no.4
• /
• pp.542-553
• /
• 2013

The aim of this work was to establish the optimal conditions for the production of cellobiase by a marine bacterium, Cellulophaga lytica LBH-14, using response-surface methodology (RSM). The optimal conditions of rice bran, ammonium chloride, and the initial pH of the medium for cell growth were 100.0 g/l, 5.00 g/l, and 7.0, respectively, whereas those for the production of cellobiase were 91.1 g/l, 9.02 g/l, and 6.6, respectively. The optimal concentrations of $K_2HPO_4$, NaCl, $MgSO_4{\cdot}_{7H2}O$, and $(NH_4)_2SO_4$ for cell growth were 6.25, 0.62, 0.28, and 0.42 g/l, respectively, whereas those for the production of cellobiase were 4.46, 0.36, 0.27, and 0.73 g/l, respectively. The optimal temperatures for cell growth and for the production of cellobiase by C. lytica LBH-14 were 35 and $25^{\circ}C$, respectively. The maximal production of cellobiase in a 100 L bioreactor under optimized conditions in this study was 92.3 U/ml, which was 5.4 times higher than that before optimization. In this study, rice bran and ammonium chloride were developed as carbon and nitrogen sources for the production of cellobiase by C. lytica LBH-14. The time for the production of cellobiase by the marine bacterium with submerged fermentations was reduced from 7 to 3 days, which resulted in enhanced productivity of cellobiase and a decrease in its production cost. This study found that the optimal conditions for the production of cellobiase were different from those of CMCase by C. lytica LBH-14.

• Food Science and Preservation
• /
• v.17 no.5
• /
• pp.659-666
• /
• 2010

A central composite design was used to investigate the effects of the three independent variables of extraction temperature ($X_1$), ethanol concentration ($X_2$), and extraction time ($X_3$), on dependent variables including yield ($Y_1$), total phenol levels ($Y_2$), electron-donating ability ($Y_3$), brownness ($Y_4$), and reducing sugar content ($Y_5$) of Vitis Coignetiae. Yield was affected by extraction temperature and time. The maximum yield was obtained at $91.62^{\circ}C(X_1)$, and, 25.37% (w/v) ethanol ($X_2$), after 317.70 min of extraction ($X_3$), evident as a saddle when displayed graphically. Total phenol levels were essentially unaffected by extraction temperature or ethanol concentration, but were highly influenced by extraction time. The maximum total phenol levels was 4,763.46 GAE mg/100 g obtained at $88.20^{\circ}C(X_1)$, and 47.79% (w/v) ethanol ($X_2$), after 349.32 min ($X_3$) of extraction. Electron-donating ability (EDA) was affected by extraction temperature and time. Maximum EDA was 55.90% at $86.72^{\circ}C(X_1)$, 50.61% (w/v) ethanol ($X_2$), and 265.96 min ($X_3$) of extration time, again shown by a graphical saddle. Brownness was affected by extraction time. The maximum extent of brown coloration was obtained at $82.66^{\circ}C(X_1)$, 99.27% (w/v) ethanol ($X_2$), and 252.63 min of extraction time ($X_3$), once again shown by graphical saddle. The maximum reducing sugar content was obtained at $96.24^{\circ}C(X_1)$, 22.59% (w/v) ethanol ($X_2$), and 216.06 min extraction time($X_3$).

• Food Engineering Progress
• /
• v.14 no.2
• /
• pp.92-98
• /
• 2010

In this study, various active functional components in Chinese Quince were extracted by solvent extraction method. A central composit design for optimization was applied to investigate the effects of independent variables such as solvent to sample ratio ($X_{1}$), extraction temperature ($X_{2}$), and extraction time ($X_{3}$) on the soluble solid contents ($Y_{1}$), total phenols ($Y_{2}$), electron donating ability ($Y_{3}$), browning color ($Y_{4}$) and reducing sugar contents ($Y_{5}$). It was found that extraction temperature and extraction time were the main effective factors in this extraction process. The maximum soluble solid contents of 35.77% was obtained at 26.38 mL/g ($X_{1}$), 72.82$^{\circ}C$ ($X_{2}$) and 74.86 min ($X_{3}$) in saddle point. Total phenols were rarely affected by solvent ratio and extraction time, but it was affected by extraction temperature. The maximum total phenols of 20.70% was obtained at 22.61 mL/g ($X_{1}$), 84.49$^{\circ}C$ ($X_{2}$), 77.25 min ($X_{3}$) in saddle point. The electron donating ability was affected by extraction time. The maximum electron donating ability of 94.12% was obtained at 10.65 mL/g ($X_{1}$), 67.78$^{\circ}C$ ($X_{2}$), 96.75 min ($X_{3}$) in saddle point. The maximum browning color of 0.32% was obtained at 23.77 mL/g ($X_{1}$), 87.27$^{\circ}C$ ($X_{2}$), 96.68 min ($X_{3}$) in saddle point. The maximum value of reducing sugar content of 10.55% was obtained at 26.83 mL/g ($X_{1}$), 82.167$^{\circ}C$ ($X_{2}$), 81.94 min ($X_{3}$). Reducing sugar content was affected by extraction time.

• Food Engineering Progress
• /
• v.15 no.4
• /
• pp.388-397
• /
• 2011

The purpose of this study was to determine the optimum ethanol extraction conditions for maximum extraction of functional components such as ferulic acid, oryzanol, and toopherol from black rice bran using Response Surface Methodology (RSM). A central composite design was applied to investigate the effects of the independent variables of solvent ratio ($X_{1}$), extraction temperature ($X_{2}$) and extraction time ($X_{3}$) on the dependent variables such as total phenol components ($Y_{1}$), total flavonoids compounds ($Y_{2}$), electron donating ability ($Y_{3}$), $\gamma$-oryzanol ($Y_{4}$), ferulic acid ($Y_{5}$) and $\alpha$-toopherol components ($Y_{6}$). ANOVA results showed that coefficients of determination (R-square) of estimated models for dependent variables ranged from 0.8939 to 0.9470. It was found that solvent ratio and extraction temperature were the main effective factors in this extraction proess. Particularly, the extraction efficiency of ferulic acid, $\gamma$-oryzanol and $\alpha$-toopherol components were significantly affected by extraction temperature. As a result, optimum extraction conditions were 20.35 mL/g of solvent ratio, 79.4$^{\circ}C$ of extraction temperature and 2.88 hr of extraction time. Predicted values at the optimized conditions were acceptable when compared with experimental values.

• Korean Journal of Food Science and Technology
• /
• v.32 no.4
• /
• pp.843-850
• /
• 2000

We have produced the microcapsule composed of ${\alpha}-tocopherol$ as a core material (Cm) and the gelatinized polysaccharide as a wall material (Wm). Firstly, we have developed a simple, sensitive, and quantitative analysis method of the microencapsulation product using 5% cupric acetate pyridine solution. We could then optimize all the conditions for the microencapsulation process such as the ratio of [Cm] to [Wm], the temperature of dispersion fluid, and the emulsifier concentration using response surface methodology (RSM). As for the microencapsulation of ${\alpha}-tocopherol$, the regression model equation for the yield of microencapsulation (YM, %) to the change of an independent variable could be predicted as follows : YM=99.77-1.76([Cm]:[Wm])-1.72$([Cm]\;:\;[Wm])^2$. From the ridge of maximum response, the optimum conditions for the microencapsulation of ${\alpha}-tocopherol$ were able to be determined as the ratio of [Cm] to [Wm] of 4.6:5.4(w/w), the emulsifier concentration of 0.49%, and dispersion fluid temperature of $25.5^{\circ}C$, respectively. Finally, the microcapsules produced under the optimal conditions were applied for the analysis of storage stability. The optimal conditions for the storage were found to be the values of pH 9.0 and $25{\sim}35^{\circ}C$. And the storage stability of the microcapsules containing ${\alpha}-tocopherol$ were higher than 99% for a week at pH 9.0 and $25^{\circ}C$.

• Microbiology and Biotechnology Letters
• /
• v.26 no.5
• /
• pp.435-441
• /
• 1998

Optimal conditions for the production of natural color, betacyanin were investigated by varying light intensity, C/N ratio, concentrations of phosphate and kinds of elicitors. Batch cultivation was employed to characterize cell growth and betacyanin production of 32 days. The maximum specific growth rate, ${\mu}$$\sub$max/, was 0.3 (1/day) for batch cultivation. The maximum specific production rate, q$\^$max/$\sub$p/, was enhanced 0.11 (mg/g-cell/day) at 3 klux. A light intensity of 3 klux was shown to the best for both cell growth and betacyanin production. The maximum specific production rate was 0.125 (mg/g-cell/day) at 0.242 (1/day), the maximum specific growth rate. The dependence of specific growth rate on the light lintensity is fit to the photoinhibition model. The correlation between ${\mu}$ and q$\sub$p/ showed that the product formation parameters, ${\alpha}$ and ${\beta}$$\sub$p/ were 0.3756 (mg/cell) and 0.001 (mg/g-cell/day), respectively. The betacyanin production was partially cell growth related process, which is different from the production of a typical product in plant cell cultures. In C/N ratio experiment, high carbon concentration, 42.1 (w/w) improved cell growth rate while lower concentration, 31.6 (w/w) increased the betacyanin production rate. The ${\mu}$$\sub$max/ and q$\^$max/$\sub$p/ were 0.26 (1/day) and 0.075 (mg/g-cell/day), respectively. Beta vulgaris L. cells under 1.25 mM phosphate concentration produced 10.15 mg/L betacyanin with 13.46 (g-dry wt./L) of maximum cell density. The production of betacyanin was elongated by adding 0.1 ${\mu}$M of kinetin. This also increased the cell growth. Optimum culture conditions of light intensity, C/N, phosphate concentration were obtained as 5.5 klux, 27 (w/w), 1.25 mM, respectively by the response surface methodology. The maximum cell density, X$\sub$max/, and maximum production, P$\sub$max/, in optimized conditions were 16 (g-dry wt./L), 12.5 (mg/L) which were higher than 8 (g-dry wt./L), 4.48 (mg/L) in normal conditions. The ${\mu}$$\sub$max/ and q$\^$max/$\sub$p/ were 0.376 (1/day) and 0.134 (mg/g-cell/day) at the optimal condition. The overall results may be useful in scaling up hairy root cell culture system for commercial production of betacyanin.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.39 no.8
• /
• pp.1206-1212
• /
• 2010

This study was performed to establish optimum extraction condition of Tricholoma matsutake. A central composite design was applied to investigate the effects of independent variables, extraction temperature ($X_1$), extraction time ($X_2$) and water per sample ($X_3$) on dependent variables such as soluble solids contents ($Y_1$), total phenolics contents ($Y_2$), reducing sugar contents ($Y_3$), electron donating ability ($Y_4$) and nitrite scavenging ability ($Y_5$). The optimum extraction conditions were predicted and monitored by response surface methodology using SAS program based regression analysis. Soluble solids content, electron donating ability and nitrite scavenging ability were highly affected by water per sample. However, the contents of total phenolics and reducing sugar were affected by water per sample and extraction temperature as well. The optimum extraction conditions for soluble solids were 34.84 mL/g (water/sample) at $78.85^{\circ}C$, for 3.33 hr. In contrast, the optimum extraction conditions of electron donating ability were temperature of $91.00^{\circ}C$, time of 1.62 hr and water per sample of 39.42 mL/g. Taken together, the optimum ranges for hot water extraction of Tricholoma matsutake were $70{\sim}90^{\circ}C$, 2~4 hr and 30~50 mL/g.