• Transactions of Materials Processing
• /
• v.29 no.4
• /
• pp.218-228
• /
• 2020

The prediction of final mass and optimized process conditions of injection molded products using Artificial Neural Network (ANN) were demonstrated. The ANN was modeled with 10 input parameters and one output parameter (mass). The input parameters, i.e.; melt temperature, mold temperature, injection speed, packing pressure, packing time, cooling time, back pressure, plastification speed, V/P switchover, and suck back were selected. To generate training data for the ANN model, 77 experiments based on the combination of orthogonal sampling and random sampling were performed. The collected training data were normalized to eliminate scale differences between factors to improve the prediction performance of the ANN model. Grid search and random search method were used to find the optimized hyper-parameter of the ANN model. After the training of ANN model, optimized process conditions that satisfied the target mass of 41.14 g were predicted. The predicted process conditions were verified through actual injection molding experiments. Through the verification, it was found that the average deviation in the optimized conditions was 0.15±0.07 g. This value confirms that our proposed procedure can successfully predict the optimized process conditions for the target mass of injection molded products.

• Fire Science and Engineering
• /
• v.32 no.1
• /
• pp.1-6
• /
• 2018

The flash point is one of the most important parameters used to characterize the ignition and explosion hazards of liquids. Flash points were measured for several binary systems containing toluene, including {methanol+toluene}, {ethanol+toluene}, and {1-propanol+toluene}. Experiments were performed according to the standard test method using a SETA closed cup flash point tester. The measured flash points were compared with the predicted values calculated using the following $G^E$ models: Wilson, NRTL, and UNIQUAC. The average absolute deviation between the predicted and measured lower flash point was less than 1.69 K.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2002.11a
• /
• pp.97-117
• /
• 2002

A chemoinfometrical method for quantitative determination of crystal content of indomethacin (IMC) polymorphs based on fourie-transformed near-infrared (FT-NIR) spectroscopy was established. A direct comparison of the data with the ones collected from using the conventional powder X-ray diffraction method was performed. Pure $\alpha$ and ${\gamma}$ forms of IMC were prepared using published methods. Powder X-ray diffraction profiles and NIR spectra were recorded for six kinds of standard materials with various content of ${\gamma}$ form IMC. The principal component regression (PCR) analyses were performed based on normalized NIR spectra sets of standard samples of known content of IMC ${\gamma}$ form. A calibration equation was determined to minimize the root mean square error of the prediction. The predicted ${\gamma}$ form content values were reproducible and had a relatively small standard deviation. The values of ${\gamma}$ form content predicted by two methods were in close agreement. The results were indicated that NIR spectroscopy provides for an accurate quantitative analysis of crystallinity in polymorphs compared with the results obtained by conventional powder X-ray diffractometry.

• International Journal of Highway Engineering
• /
• v.18 no.1
• /
• pp.91-97
• /
• 2016

PURPOSES : The objective of this paper is to develop a pavement performance model based on the Bayesian algorithm, and compare the measured and predicted performance data. METHODS : In this paper, several pavement types such as SMA (stone mastic asphalt), PSMA (polymer-modified stone mastic asphalt), PMA (polymer-modified asphalt), SBS (styrene-butadiene-styrene) modified asphalt, and DGA (dense-graded asphalt) are modeled in terms of the performance evaluation of pavement structures, using the Bayesian algorithm. RESULTS : From case studies related to the performance model development, the statistical parameters of the mean value and standard deviation can be obtained through the Bayesian algorithm, using the initial performance data of two different pavement cases. Furthermore, an accurate performance model can be developed, based on the comparison between the measured and predicted performance data. CONCLUSIONS : Based on the results of the case studies, it is concluded that the determined coefficients of the nonlinear performance models can be used to accurately predict the long-term performance behaviors of DGA and modified asphalt concrete pavements. In addition, the developed models were evaluated through comparison studies between the initial measurement and prediction data, as well as between the final measurement and prediction data. In the model development, the initial measured data were used.

• Tribology and Lubricants
• /
• v.30 no.6
• /
• pp.356-363
• /
• 2014

Marine diesel engines operate in environments in which damage easily occurs from corrosion. Recently, damage to cylinder liners has increased from corrosion wear caused by increased engine power. This damage can cause serious problems in the economy. Thus, many researchers have treated and studied damaged cylinder liners. However, a method is necessary for real-time monitoring of damage to cylinder liners during operation of the engine, before serious damage can occur. This study carries out reciprocating friction and wear tests on a cast iron specimen under various corrosion atmospheres and verifies the variations of friction coefficient and friction surface. Additionally, the friction coefficient and friction status are predicted by using a neural network that learns the vibration and frequency spectrum data from an acceleration sensor. According to our conclusions, amplitude is distributed highly at high frequencies, and values of standard deviation and kurtosis are high when damage to the friction surface is serious. The accuracy rate of the friction coefficient predicted by the neural network is over 80% of the real measured value without NaCl, and application of the neural network is very effective for diagnosing the friction condition and damage to the cylinder liner.

• Journal of the Korean Society of Visualization
• /
• v.20 no.3
• /
• pp.109-116
• /
• 2022

The optimized design of airfoils is essential to increase the performance and efficiency of wind turbines. The aerodynamic characteristics of airfoils near the stall show large deviation from experiments and numerical simulations. Hence, it is needed to perform repetitive analysis of various shapes near the stall. To overcome this, the artificial intelligence is used and combined with numerical simulations. In this study, three types of airfoils are chosen, which are S809, S822 and SD7062 used in wind turbines. A convolutional neural network model is proposed in the combination of VGG16 and U-Net. Learning data are constructed by extracting pressure fields and aerodynamic characteristics through numerical analysis of 2D shape. Based on these data, the pressure field and lift coefficient of untrained airfoils are predicted. As a result, even in untrained airfoils, the pressure field is accurately predicted with an error of within 0.04%.

• Fire Science and Engineering
• /
• v.29 no.5
• /
• pp.1-6
• /
• 2015

Flash point is one of the most important variables used to characterize fire and explosion hazard of liquids. The lower flash point data were measured for the binary systems {methanol + 1-butanol}, {ethanol + 1-butanol} and {2-propanol + 1-butanol} at 101.3 kPa. Experiments were performed according to the standard test method (ASTM D 3278) using a SETA closed cup flash point tester. The measured flash points were compared with the predicted values calculated using the following activity coefficient models: Wilson, Non-Random Two Liquid (NRTL), and UNIversal QUAsiChemical (UNIQUAC). The measured FP data agreed well with the predicted values of Raoult's law, Wilson, NRTL and UNIQUAC models. The average absolute deviation between the predicted and measured lower FP was less than 1.14 K.

• Computers and Concrete
• /
• v.1 no.3
• /
• pp.325-354
• /
• 2004

Slabs in buildings and bridge decks, which are restrained against lateral displacements at the edges, have ultimate strengths far in excess of those predicted by analytical methods based on yield line theory. The increase in strength has been attributed to membrane action, which is due to the in-plane forces developed at the supports. The benefits of compressive membrane action are usually not taken into account in currently available design methods developed based on plastic flow theories assuming concrete to be a rigid-plastic material. By extending the existing knowledge of compressive membrane action, it is possible to design slabs in building and bridge structures economically with less than normal reinforcement. Recent research on building and bridge structures reflects the importance of membrane action in design. This paper describes the finite element modelling of membrane action in reinforced concrete slabs through optimisation of a simple concrete model. Through a series of parametric studies using the simple concrete model in the finite element simulation of eight fully clamped concrete slabs with significant membrane action, a set of fixed numerical model parameter values is identified and computational conditions established, which would guarantee reliable strength prediction of arbitrary slabs. The reliability of the identified values to simulate membrane action (for prediction purposes) is further verified by the direct simulation of 42 other slabs, which gave an average value of 0.9698 for the ratio of experimental to predicted strengths and a standard deviation of 0.117. A 'deflection factor' is also established for the slabs, relating the predicted peak deflection to experimental values, which, (for the same level of fixity at the supports), can be used for accurate displacement determination. The proposed optimised concrete model and finite element procedure can be used as a tool to simulate membrane action in slabs in building and bridge structures having variable support and loading conditions including fire. Other practical applications of the developed finite element procedure and design process are also discussed.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.21 no.5
• /
• pp.391-404
• /
• 2009

A small hydropower plant(SHP) using cooling water discharged from the power plant was constructed in Samcheonpo. This study presents predicted and measured hydrological data in the construction process of small hydropower plant in order to evaluate characteristics of water level variation of cooling water discharge channel which is a key factor in the design of SHP since the water level rise of channel is related to impact on circulating water system of the existing power plant. Various methods were applied for prediction of water level variation in the design stage from simple empirical formula to sophisticated 3-dimensional CFD method. Measured results reveal that mean value was similar between measured and predicted, but measured results were larger than predicted in deviation. Moreover, simple formula, i.e. standard weir equation and Honma equation, were more useful before installation of SHP, but sophisticated methods during operation of SHP.

• Membrane Journal
• /
• v.6 no.1
• /
• pp.22-31
• /
• 1996

Pervaporation experiments of isopropanol-water mixtures through a polydimethytsiloxane(PDMS) membrane were carried out at 35$^{\circ}$C and the effect of isopropanol concentration on the separation characteristics was investigated. The total permeation rate showed the largest deviation from the ideal permeation rate at the isoprpanol volume fraction from 0.5 to 0.7, which resulted from the interaction effect between permeants. The plasticizing effect of isopropanol enhanced the permeation of water, while the existance of water resulted in the depression of isopropanol permeation. Both the permeation rate and the selectivity were predicted using Flory-Huggins thermodynamics and modified Maxwell-Stefan equation. The concentration-dependent diffusion coefficients were expressed by Vignes equation. The Flory-Huggins interaction parameter between isopropanol and water was calculated using excess Gibbs energy correlation and the interaction parameters between liquid and polymer membrane were determined by equilibrium swelling experiments. The predicted permeation rates were in accord with the experimental ones within maximum error range of 35 %. The predicted permeation selectivities were in good agreement with the experimental values.