• Design & Manufacturing
• /
• v.14 no.4
• /
• pp.11-16
• /
• 2020

In this study, forming of carbon composite parts was performed using an injection/compression molding process. An impregnation of matrix is determined by ability of wet and flow rate between the matrix and reinforcement. The flow rate of matrix passing through the reinforcements is a function of permeability of reinforcement, a viscosity of matrix and pressure gradient on molding, and the viscosity of the matrix depends on the mold temperature, molding pressure and shear strain of matrix. Therefore, compression molding experiment was conducted using a heating mold in order to confirm the possibility of matrix impregnation. The impregnation of the matrix through the porosities between the woven yarns was confirmed by the cross-sectional SEM image of compression molded parts. An injection molding process was also performed at a short cycle time, high molding pressure and low mold temperature than those of compression experiment conditions. Deterioration of impregnation on the surface of molded parts were caused by these injection conditions and it could be the reason of decreasing the maximum tensile strength. In order to improve impregnation of matrix on the surface, injection/compression molding and insert-over molding were applied. As a result of applying injection/compression molding and insert-over molding, it was shown that the improvement of impregnation on the surface and the maximum tensile strength was increased about 2.8 times than the virgin matrix.

• The Korean Journal of Applied Statistics
• /
• v.34 no.2
• /
• pp.153-166
• /
• 2021

A main goal of pharmacogenomics studies is to predict individual's drug responsiveness based on high dimensional genetic variables. Due to a large number of variables, feature selection is required in order to reduce the number of variables. The selected features are used to construct a predictive model using machine learning algorithms. In the present study, we applied several hybrid feature selection methods such as combinations of logistic regression, ReliefF, TurF, random forest, and LASSO to a next generation sequencing data set of 400 epilepsy patients. We then applied the selected features to machine learning methods including random forest, gradient boosting, and support vector machine as well as a stacking ensemble method. Our results showed that the stacking model with a hybrid feature selection of random forest and ReliefF performs better than with other combinations of approaches. Based on a 5-fold cross validation partition, the mean test accuracy value of the best model was 0.727 and the mean test AUC value of the best model was 0.761. It also appeared that the stacking models outperform than single machine learning predictive models when using the same selected features.

• Journal of Digital Convergence
• /
• v.20 no.4
• /
• pp.389-396
• /
• 2022

Deep Learning is based on a perceptron, and is currently being used in various fields such as image recognition, voice recognition, object detection, and drug development. Accordingly, a variety of learning algorithms have been proposed, and the number of neurons constituting a neural network varies greatly among researchers. This study analyzed the learning characteristics according to the number of neurons of the currently used SGD, momentum methods, AdaGrad, RMSProp, and Adam methods. To this end, a neural network was constructed with one input layer, three hidden layers, and one output layer. ReLU was applied to the activation function, cross entropy error (CEE) was applied to the loss function, and MNIST was used for the experimental dataset. As a result, it was concluded that the number of neurons 100-300, the algorithm Adam, and the number of learning (iteraction) 200 would be the most efficient in deep learning learning. This study will provide implications for the algorithm to be developed and the reference value of the number of neurons given new learning data in the future.

• Geomechanics and Engineering
• /
• v.27 no.5
• /
• pp.433-445
• /
• 2021

Artificial ground freezing (AGF) is a commonly used geotechnical support technique that can be applied in any soil type and has low environmental impact. Experimental and numerical investigations have been conducted to optimize AGF for application in diverse scenarios. Precise simulation of groundwater flow is crucial to improving the reliability these investigations' results. Previous experimental research has mostly considered horizontal seepage flow, which does not allow accurate calculation of the groundwater flow velocity due to spatial variation of the piezometric head. This study adopted vertical seepage flow-which can maintain a constant cross-sectional area-to eliminate the limitations of using horizontal seepage flow. The closure time is a measure of the time taken for an impermeable layer to begin to form, this being the time for a frozen soil-ice wall to start forming adjacent to the freeze pipes; this is of great importance to applied AGF. This study reports verification of the reliability of our experimental apparatus and measurement system using only water, because temperature data could be measured while freezing was observed visually. Subsequent experimental AFG tests with saturated sandy soil were also performed. From the experimental results, a method of estimating closure time is proposed using the inflection point in the thermal conductivity difference between pore water and pore ice. It is expected that this estimation method will be highly applicable in the field. A further parametric study assessed factors influencing the closure time using a two-dimensional coupled thermo-hydraulic numerical analysis model that can simulate the AGF of saturated sandy soil considering groundwater flow. It shows that the closure time is affected by factors such as hydraulic gradient, unfrozen permeability, particle thermal conductivity, and freezing temperature. Among these factors, changes in the unfrozen permeability and particle thermal conductivity have less effect on the formation of frozen soil-ice walls when the freezing temperature is sufficiently low.

• Journal of the Korean Wood Science and Technology
• /
• v.30 no.2
• /
• pp.108-114
• /
• 2002

Red pine(Pinus densiflora) square timbers with 14.0 cm and 16.5 cm of face size and 24 m long were dried in a vacuum-press kiln and in a radio-frequency/vacuum(RF/V) kiln to compare drying rate, moisture content(MC) distribution and specific energy. RF/V drying rate was higher than vacuum-press drying rate. The effect of size of cross section on the RF/V drying rates were more pronounced than those of vacuum-press drying. The longitudinal- and the transverse MC distribution of dried square timber showed convex profile for the vacuum-press drying and concave profile for the RF/V drying. Moisture gradient of width direction was similar to the thickness direction in vacuum-press dried square timber and was more slight than that of the thickness direction in the RF/V dried large square timber. The specific energy consumption curve increased as MC decreased. Specific energy(kWh/kg of water evaporated) of the vacuum-press process required more than that of the RF/V process.

• Journal of Preventive Medicine and Public Health
• /
• v.56 no.1
• /
• pp.41-49
• /
• 2023

Objectives: Rwanda reported a stunting rate of 33% in 2020, decreasing from 38% in 2015; however, stunting remains an issue. Globally, child deaths from malnutrition stand at 45%. The best options for the early detection and treatment of stunting should be made a community policy priority, and health services remain an issue. Hence, this research aimed to develop a model for predicting stunting in Rwandan children. Methods: The Rwanda Demographic and Health Survey 2019-2020 was used as secondary data. Stratified 10-fold cross-validation was used, and different machine learning classifiers were trained to predict stunting status. The prediction models were compared using different metrics, and the best model was chosen. Results: The best model was developed with the gradient boosting classifier algorithm, with a training accuracy of 80.49% based on the performance indicators of several models. Based on a confusion matrix, the test accuracy, sensitivity, specificity, and F1 were calculated, yielding the model's ability to classify stunting cases correctly at 79.33%, identify stunted children accurately at 72.51%, and categorize non-stunted children correctly at 94.49%, with an area under the curve of 0.89. The model found that the mother's height, television, the child's age, province, mother's education, birth weight, and childbirth size were the most important predictors of stunting status. Conclusions: Therefore, machine-learning techniques may be used in Rwanda to construct an accurate model that can detect the early stages of stunting and offer the best predictive attributes to help prevent and control stunting in under five Rwandan children.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.46 no.1
• /
• pp.23-31
• /
• 2023

Research and interest in sustainable printing are increasing in the packaging printing industry. Currently, predicting the amount of ink required for each work is based on the experience and intuition of field workers. Suppose the amount of ink produced is more than necessary. In this case, the rest of the ink cannot be reused and is discarded, adversely affecting the company's productivity and environment. Nowadays, machine learning models can be used to figure out this problem. This study compares the ink usage prediction machine learning models. A simple linear regression model, Multiple Regression Analysis, cannot reflect the nonlinear relationship between the variables required for packaging printing, so there is a limit to accurately predicting the amount of ink needed. This study has established various prediction models which are based on CART (Classification and Regression Tree), such as Decision Tree, Random Forest, Gradient Boosting Machine, and XGBoost. The accuracy of the models is determined by the K-fold cross-validation. Error metrics such as root mean squared error, mean absolute error, and R-squared are employed to evaluate estimation models' correctness. Among these models, XGBoost model has the highest prediction accuracy and can reduce 2134 (g) of wasted ink for each work. Thus, this study motivates machine learning's potential to help advance productivity and protect the environment.

• Journal of Korean Society of Forest Science
• /
• v.84 no.2
• /
• pp.239-246
• /
• 1995

This research aimed at the contribution to obtaining the scientifical data which were required for planning she environmentally sound and sustainable management, particularly in the field of the logging road construction. Main natural environmental variables including natural vegetation, rainfall, soil runoff were measured in the logging road on-sites and analysed. This project was carried out at the (mt.)Paekunsan Research sorest of Seoul National University, located in Gwangyang, Chollanam-do in southern part of Korea, from 1993 to 1994. 1. The explanatory variables for erosion and sedimentation on logging road surface were accumulated rainfall, erosion distance, cross-sectional gradient, and soil hardness. The erosion and sedimentation on logging road was increasing positively in proportion to the accumulated rainfall, soil distance from starting point of the logging road, and cross-sectional gradient. 2. On cut-slope of logging road, cut-slope shape, part of the slope, plant coverage, soil hardness, sand content, accumulated rainfall, clay content, and silt content were effective factors. Cut-slope erosion and sedimentation on logging roam increased as with the lower plant coverage, the lower accumulated rainfall, the high sand content in the soil. 3. On fill-slope of logging road, there were three significant variables such as total rainfall and number of rainfall-storm. Fill-slope erosion and sedimentation had a positive correlation with the amount of rainfall, the number of rainfall, the soil hardness. 4. The total erosion and sedimentation on logging road were $5.04{\times}10^{-2}m^2/m^2$ in logging road construction year, $7.37{\times}10^{-2}m^2/m^2$ in next year. The erosion and sedimentation on logging road surface were 32.7% of total erosion and sedimentation on Logging road in construction year, and 57.1% in next year, respectively. The erosion and sedimentation on cut-slopes were 30.4% on logging road in construction year, fill-slopes of total erosion and sedimentation and 21.0% in next year, respectively. The erosion and sedimentation on fill-slopes were 36.9% on logging road in construction year, 21.9 in next year. To decrease the erosion and sedimentation at the logging road from the beginning stage of construction, the effective revegetation works should be implemented on the cut-slope and fill slopes, and erosion control measures such as optima. road design must be constructed on read surface.

• Parasites, Hosts and Diseases
• /
• v.25 no.2
• /
• pp.159-167
• /
• 1987

To analyse the antigen specificity of patients sera from 24 confirmed neurocysticercosis and a monoclonal antibody, SDS-PAGE using 10~15% linear gradient gel and EITB were done. Cystic fluid, saline extracts of scolex and of whole worm of C. cellulosae, saline extracts of sparganum, hydatid cyst fluid, saline extracts of Fasciola, Clonorchis and Paragonimus were used as antigen. Of protein bands in cystic fluid of C. cellulosae, patient sera reacted frequently to bands of 152, 94, 64, 48, 24, 15, 10 and 7kDa proteins. To saline extracts of scolex and whole worm of C. cellulosae, patients sera reacted frequently to 94, 64, 52, 39, 34, 15 and 10kDa bands. Two bands in sparganum extract (130 and 64kDa) and two bands in hydatid cyst fluid (52 and 27kDa) were cross-reacting bands with sera from cysticercosis patients. Saline extracts of Fasciola, ClonorchiJ and Paragonimus did 'not exhibit cross-reacting bands. Monoclonal antibody to cystic fluid of C. cellulosae was found to react with low molecular weight proteins of 15, 10 and 7kDa.

• Korean Journal of Remote Sensing
• /
• v.38 no.5_2
• /
• pp.707-723
• /
• 2022

Although satellite-based sea surface temperature (SST) is advantageous for monitoring large areas, spatiotemporal data gaps frequently occur due to various environmental or mechanical causes. Thus, it is crucial to fill in the gaps to maximize its usability. In this study, daily SST composite fields with a resolution of 4 km were produced through a two-step machine learning approach using polar-orbiting and geostationary satellite SST data. The first step was SST reconstruction based on Data Interpolate Convolutional AutoEncoder (DINCAE) using multi-satellite-derived SST data. The second step improved the reconstructed SST targeting in situ measurements based on light gradient boosting machine (LGBM) to finally produce daily SST composite fields. The DINCAE model was validated using random masks for 50 days, whereas the LGBM model was evaluated using leave-one-year-out cross-validation (LOYOCV). The SST reconstruction accuracy was high, resulting in R² of 0.98, and a root-mean-square-error (RMSE) of 0.97℃. The accuracy increase by the second step was also high when compared to in situ measurements, resulting in an RMSE decrease of 0.21-0.29℃ and an MAE decrease of 0.17-0.24℃. The SST composite fields generated using all in situ data in this study were comparable with the existing data assimilated SST composite fields. In addition, the LGBM model in the second step greatly reduced the overfitting, which was reported as a limitation in the previous study that used random forest. The spatial distribution of the corrected SST was similar to those of existing high resolution SST composite fields, revealing that spatial details of oceanic phenomena such as fronts, eddies and SST gradients were well simulated. This research demonstrated the potential to produce high resolution seamless SST composite fields using multi-satellite data and artificial intelligence.