• Korean Chemical Engineering Research
• /
• v.58 no.4
• /
• pp.565-572
• /
• 2020

This study developed a software platform using machine learning of artificial intelligence to optimize the distillation column system. The distillation column is representative and core process in the petrochemical industry. Process stabilization is difficult due to various operating conditions and continuous process characteristics, and differences in process efficiency occur depending on operator skill. The process control based on the theoretical simulation was used to overcome this problem, but it has a limitation which it can't apply to complex processes and real-time systems. This study aims to develop an empirical simulation model based on machine learning and to suggest an optimal process operation method. The development of empirical simulations involves collecting big data from the actual process, feature extraction through data mining, and representative algorithm for the chemical process. Finally, the platform for the distillation column was developed with verification through a developed model and field tests. Through the developed platform, it is possible to predict the operating parameters and provided optimal operating conditions to achieve efficient process control. This study is the basic study applying the artificial intelligence machine learning technique for the chemical process. After application on a wide variety of processes and it can be utilized to the cornerstone of the smart factory of the industry 4.0.

• Journal of the Korean Society for Railway
• /
• v.15 no.1
• /
• pp.9-16
• /
• 2012

The dynamic stability of railway vehicle has been one of the important issues in railway safety. The dynamic simulator has been used in the study about the dynamic stability of railway vehicle and wheel/rail interface optimization. Especially, a small scale simulator has been widely used in the fundamental study in the laboratory instead of full scale roller rig which is not cost effective and inconvenient to achieve diverse design parameters. But the technique for the design of the small scale simulator about the dynamic characteristics of the wheel-rail system and the bogie system has not been well developed in Korea. Therefore, the research using the small-scaled derailment simulator and the 1/5 scaled bogie has been conducted. In this paper, we did running stability test of 1/5 scaled bogie using small-scaled derailment simulator. Also, for the operation of the small scaled simulator, it is required to investigate the performance and characteristics of the simulator system. This could be achieved by a comparative study between an analysis and an experiment. This paper presented the analytical model which could be used for verifying the test results and understanding of the physical behavior of the dynamic system comprising the small- scaled derailment simulator and the 1/5 scaled bogie.

• Journal of the Korean Vacuum Society
• /
• v.21 no.1
• /
• pp.29-35
• /
• 2012

The Hydrogenated silicon nitride (SiNx:H) using plasma enhanced chemical vapor deposition is widely used in photovoltaic industry as an antireflection coating and passivation layer. In the high temperature firing process, the $SiN_x:H$ film should not change the properties for its use as high quality surface layer in crystalline silicon solar cells. Initially PECVD-$SiN_x:H$ film trends were investigated by varying the deposition parameters (temperature, electrode gap, RF power, gas flow rate etc.) to optimize the process parameter conditions. Then by varying gas ratios ($NH_3/SiH_4$), the hydrogenated silicon nitride films were analyzed for its optical, electrical, chemical and surface passivation properties. The $SiN_x:H$ films of refractive indices 1.90~2.20 were obtained. The film deposited with the gas ratio of 3.6 (Refractive index=1.98) showed the best properties in after firing process condition. The single crystalline silicon solar cells fabricated according to optimized gas ratio (R=3.6) condition on large area substrate of size $156{\times}156mm$ (Pseudo square) was found to have the conversion efficiency as high as 17.2%. Optimized hydrogenated silicon nitride surface layer and high efficiency crystalline silicon solar cells fabrication sequence has also been explained in this study.

• Journal of Korea Water Resources Association
• /
• v.41 no.4
• /
• pp.413-422
• /
• 2008

It is essential to locally adjust the Hargreaves parameter for estimating reference evapotranspiration with short data as a substitute of Penman-Monteith equation. In this study, evaluation of daily-based reference evapotranspiration is computed with Hargreaves equation. in Gyeonggi bay area including Ganghwa, Incheon, Suwon, Seosan, and Cheonan station for the time period of 1997-2004. Hargreaves coefficient is adjusted to give the best fit with Penman-Monteith evapotranspiration, being regarded as a reference. Then, the preferred parameters are validated for the same stations for the time period of 2005-2006. The optimization-based correction in calibration for 1997-2004 shows improved performance of the Hargreaves equation, giving 0.68-0.77 to 0.92-0.98 in Nash-Sutcliffe coefficient of efficiency (NSC) and 14.63-23.30 to 5.23-11.75 in RMSE. The validation for 2005-2006 shows improved performance of the Hargreaves equation, giving 0.43-0.85 to 0.93-0.97 in NSC and 14.43-26.81 to 6.48-9.09 in RMSE.

• Korean Journal of Soil Science and Fertilizer
• /
• v.40 no.4
• /
• pp.249-254
• /
• 2007

Inoculation dose of agriculturally important microbes is an important criterion that decides the establishment and hence their effects on plant growth. Effects of the inoculation dose of Azospirillum brasilense strain CW903 on the growth and nutrient absorption of three different crops, tomato, rice and red pepper were assessed under green house condition. Three different concentrations of A. brasilense strain CW903 ($10^5$, $10^6$ and $10^8cfu\;mL^{-1}$) were applied through seed treatment and through the soil near the root zone (1 mL per plant) at 20 and 30 days after sowing. Positive effects on the growth of tomato, rice and red pepper were found at $10^6$ and $10^8cfu\;mL^{-1}$ inoculation doses of A. brasilense strain CW903. The inoculation dose of $10^8cfu\;mL^{-1}$ of A. brasilense strain CW903 recorded the best effects on growth parameters like shoot and root length and the absorption of important nutrients.

• Korean Journal of Optics and Photonics
• /
• v.20 no.3
• /
• pp.175-181
• /
• 2009

Holographic lithography is one of the potential technologies for next generation lithography which can print large areas (6") as well as very fine patterns ($0.35{\mu}m$). Usually, photolithography has been developed with two target purposes. One was for LCD applications which require large areas (over 6") and micro pattern (over $1.5{\mu}m$) exposure. The other was for semiconductor applications which require small areas (1.5") and nano pattern (under $0.2{\mu}m$) exposure. However, holographic lithography can print fine patterns from $0.35{\mu}m$ to $1.5{\mu}m$ keeping the exposure area inside 6". This is one of the great advantages in order to realize high speed fine pattern photolithography. How? It is because holographic lithography is taking holographic optics instead of projection optics. A hologram mask is the key component of holographic optics, which can perform the same function as projection optics. In this paper, Surface-Relief TIR Hologram Mask technology is introduced, and enables more robust hologram masks than those previously reported that were formed in photopolymer recording materials. We describe the important parameters in the fabrication process and their optimization, and we evaluate the patterns printed from the surface-relief TIR hologram masks.

• Korean Journal of Materials Research
• /
• v.2 no.3
• /
• pp.180-190
• /
• 1992

Some solutions to several major problems in ZMR such as agglomeration of polysilicon, slips and local substrate melting are described. Experiments are performed with varying polysilicon thickness and capping oxide thickness. The aggmeration can be eliminated when nitrogen is introduced at the capping oxide layer-to-polysilicon interface and polysilicon-to-buried oxide layer interface by annealing the SOI samples at $1100^{\circ}$ in $NH_3$ ambient for three hours. The slips and local substrate melting are removed when the back surface of silicon substrate is sandblasted to produce the back surface roughness of about $20{\mu}m$. The subboundary spacing increases with increasing polysilicon thickness and the uniformity of recrystallized SOI film thickness improves with increasing capping oxide thickness, improving the quality of recrystallized SOI film. When the polysilicon thickness is about $1.0{\mu}m$ and the capping oxide thickness is $2.5{\mu}m$, the thickness variation of the recrystallized SOI film is about ${\pm}200{\AA}$ and the subboundary spacing is about $70-120{\mu}m$.

• Journal of Digital Convergence
• /
• v.18 no.11
• /
• pp.259-266
• /
• 2020

Many developments have been steadily carried out by researchers with applying knowledge-based expert system or machine learning algorithms to the financial field. In particular, it is now common to perform knowledge based system trading in using stock prices. Recently, deep learning technologies have been applied to real fields of stock trading marketplace as GPU performance and large scaled data have been supported enough. Especially, LSTM has been tried to apply to stock price prediction because of its compatibility for time series data. In this paper, we implement stock price prediction using LSTM. In modeling of LSTM, we propose a fitness combination of model parameters and activation functions for best performance. Specifically, we propose suitable selection methods of initializers of weights and bias, regularizers to avoid over-fitting, activation functions and optimization methods. We also compare model performances according to the different selections of the above important modeling considering factors on the real-world stock price data of global major companies. Finally, our experimental work brings a fitness method of applying LSTM model to stock price prediction.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.11 no.1 s.53
• /
• pp.11-19
• /
• 2007

This paper proposes an optimal design method of distribution and capacities of linear viscous dampers for vibration control of two adjacent buildings. The previous researches have dealt with suboptimal design problem under the assumption that linear viscous dampers are distributed uniformly or proportionally to the sensitivity of the modal damping ratio according to floors, whereas this study deals with global optimization problem in which the damping capacities of each floor are independently selected as design parameters. For this purpose, genetic algorithm to effectively search multiple design variables in large searching domains is adopted and objective function leading to the global optimal solutions is established through the comparison of several optimal design values obtained from different objective functions with control performance and damping capacity. The effectiveness of the proposed method is investigated by comparing the control performance and total damping capacity designed by the proposed method with those of the previous method. In addition, the time history analyses are performed by using three historical earthquakes with different frequency contents, and the simulation results demonstrate that the proposed method is an effective seismic design method for the vibration control of the adjacent structures.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.40 no.7
• /
• pp.975-980
• /
• 2011

1-Palmitoyl-2-oleoyl-3-oleoyl glycerol (POO) and 1-palmitoyl-2-oleoyl-3-palmitoyl glycerol (POP) were enriched from palm stearin using an acetone fractionation process. Response surface methodology was employed to optimize the purity of POO ($Y_1$, %) and POP ($Y_2$, %) along with POO+POP content ($Y_3$, g) based on independent variables such as fractionation temperature ($X_1$, 25, 30, and $35^{\circ}C$) and the ratio of palm stearin to acetone ($X_2$, 1:3, 1:6 and 1:9, w/v). Fractionation conditions were optimized to maximize $Y_1$, $Y_2$, and $Y_3$, in which fractionation temperature was $29.3^{\circ}C$ with a 1:5.7 acetone ratio. With such parameters, 60.9% of POP and 23.8% of POO purity were expected with a 75% yield (3.0 g) of POO and POP.