• Journal of Korea Water Resources Association
• /
• v.55 no.5
• /
• pp.321-331
• /
• 2022

In this study, we tried to present a method for calculating the amount of regression using a watershed modeling method that can simulate the hydrological mechanism of water balance analysis and agricultural water based on watershed unit. Using the soil water assessment tool (SWAT), a watershed water balance analysis was conducted considering the simulation of paddy fields for the Manbongcheon Standard Basin (97.34 km²), which is a representative agricultural area of the Yeongsan river basin. Before evaluating return flow, the SWAT was calibrated and validated using the daily streamflow observation data at Naju streamflow gauge station (NJ). The coefficient of determination (R²), Nash-Sutcliffe Efficiency (NSE), Root-Mean-Square Error (RMSE) of NJ were 0.73, 0.70, 0.64 mm/day. Based on the calibration results for three years (2015-2017), the quick return flow and the return rate compared to the water supply amount for the irrigation period (April 1 to September 30) were calculated, and the average return flow rate was 53.4%. The proposed method of this study may be used as foundation data to optimal agricultural water supply plan for rational watershed management.

• Journal of the Korean Geotechnical Society
• /
• v.35 no.11
• /
• pp.51-62
• /
• 2019

In this study, a simple method to estimate the soil water content variation in a wide area was proposed using hyperspectral near-infrared images. The reflectance data of a sand, granite soils, and a kaolinite were measured by reflecting the soil samples with different wavelengths in the visible and near-infrared (VNIR) regions using hyperspectral cameras. The measured reflectances and parameters were used to build a water content prediction model using the Partial Least Square Regression (PLSR) analysis. In the water content prediction model, the Area of Reflectance (Near-infrared, NIR) parameter was the most suitable parameter to determine the water content. The parameter was applicable regardless of the soil type, as the coefficient of determination (R²) exceeded 0.9 for each soil sample. Additionally, the mean absolute percentage error (MAPE) was less than 15% when compared with the actual water content of the soil. Therefore, the predictability of water content variation for soils with water content lower than 50% was confirmed. Accordingly through this study, the predictability of water content variation in several soil types using the hyperspectral near-infrared images was confirmed. For further development, a model that incorporates soil classification would be required to improve the accuracy of the model and to predict higher range of water contents.

• Tunnel and Underground Space
• /
• v.30 no.3
• /
• pp.214-225
• /
• 2020

Uniaxial compressive strength (UCS) of rock is one of the important factors to determine the advance speed during shield TBM tunnel excavation. UCS can be obtained through the Geotechnical Data Report (GDR), and it is difficult to measure UCS for all tunneling alignment. Therefore, the purpose of this study is to predict UCS by utilizing TBM machine driving data and machine learning technique. Several machine learning techniques were compared to predict UCS, and it was confirmed the stacking model has the most successful prediction performance. TBM machine data and UCS used in the analysis were obtained from the excavation of rock strata with slurry shield TBMs. The data were divided into 8:2 for training and test and pre-processed including feature selection, scaling, and outlier removal. After completing the hyper-parameter tuning, the stacking model was evaluated with the root-mean-square error (RMSE) and the determination coefficient (R²), and it was found to be 5.556 and 0.943, respectively. Based on the results, the sacking models are considered useful in predicting rock strength with TBM excavation data.

• Journal of Korea Water Resources Association
• /
• v.53 no.1
• /
• pp.57-70
• /
• 2020

This study is to establish the situation of inter-basin transfer from Seomjin river basin to Yeongsan river basin using SWAT (Soil and Water Assessment Tool). Firstly, the SWAT modeling was conducted for each river basin. After, the inter-basin transfer was established using SWAT reservoir operating parameters WURESN (Water Use Reservoir Withdrawn) and inlet function from Juam dam of Seomjin river basin to Gwangju stream of Yeongsan river basin respectively. Each river basin was calibrated and validated using 13 years (2005~2017) data of Seomjin- Juam dam reservoir storage (JAD), release, transfer and Yeongsan-Mareuk (MR) stream gauge station. The results of root mean square error RMSE, Nash-Sutcliffe efficiency NSE, and determination coefficient R² of JAD were 2.22 mm/day, 0.62 and 0.86 respectively. The RMSE, NSE, and R² of MR were 1.38 mm/day, 0.69 and 0.84 respectively. To evaluate the downstream effects by the transferred water, the water levels of 2 multi-function weirs (SCW, JSW) in Yeongsan river basin and the Gokseong (GS) and Gurye (GR) stream gauge stations in Seomjin river basin were also calibrated. The RMSE, NSE, and R² of SCW, JSW, GS and GR were 1.49~2.49 mm/day, 0.45~0.76, 0.81~0.90 respectively.

• Korean Journal of Remote Sensing
• /
• v.37 no.3
• /
• pp.515-530
• /
• 2021

This study is to estimate soil moisture (SM) using Sentinel-1A/B C-band SAR (synthetic aperture radar) images and Multiple Linear Regression Model(MLRM) in the Yongdam-Dam watershed of South Korea. Both the Sentinel-1A and -1B images (6 days interval and 10 m resolution) were collected for 5 years from 2015 to 2019. The geometric, radiometric, and noise corrections were performed using the SNAP (SentiNel Application Platform) software and converted to backscattering coefficient of VV and VH polarization. The in-situ SM data measured at 6 locations using TDR were used to validate the estimated SM results. The 5 days antecedent precipitation data were also collected to overcome the estimation difficulty for the vegetated area not reaching the ground. The MLRM modeling was performed using yearly data and seasonal data set, and correlation analysis was performed according to the number of the independent variable. The estimated SM was verified with observed SM using the coefficient of determination (R²) and the root mean square error (RMSE). As a result of SM modeling using only BSC in the grass area, R² was 0.13 and RMSE was 4.83%. When 5 days of antecedent precipitation data was used, R² was 0.37 and RMSE was 4.11%. With the use of dry days and seasonal regression equation to reflect the decrease pattern and seasonal variability of SM, the correlation increased significantly with R² of 0.69 and RMSE of 2.88%.

• Analytical Science and Technology
• /
• v.32 no.2
• /
• pp.56-64
• /
• 2019

Recently, a number of adulterated products, which are advertised as hair-growth enhancer have been emerged among those who suffer hair loss disease. For continuous control of illegal products, in this study, a rapid and sensitive method for simultaneous screening of 12 compounds that enhance hair-growth was established to protect public health by ultrahigh-performance liquid chromatography coupled to quadrupole-orbitrap mass spectrometry (UHPLC-Q-Orbitrap-MS). Fragmentation pathways of them were proposed based on $MS^2$ spectral data obtained using the established method. In this analysis, the LODs and LOQs ranged from 0.05 to 50 ng/mL and from 0.17 to 167 ng/mL, respectively. The square of the linear correlation coefficient ($R^2$) was determined as more than 0.995. The intra- and inter-assay accuracies were respective 88-112 % and 88-115 %. Their precision values were measured within 5 % (intra-day) and 10 % (inter-day). Mean recoveries of target compounds in adulterated products ranged from 84 to 115%. The relative standard deviation of stability was less than 12 % at $4^{\circ}C$ for 48 h. The method was employed to screen 14 dietary supplements advertised to be effective for the treatment of hair loss. Some of the products (~21 %) were proven to contain synthetic drugs that promote hair growth such as triaminodil, minoxidil, and finasteride.

• Geomechanics and Engineering
• /
• v.31 no.6
• /
• pp.545-556
• /
• 2022

Evaluation and optimization of tunnel wall convergence (TWC) plays a vital role in preventing potential problems during tunnel construction and utilization stage. When convergence occurs at a high rate, it can lead to significant problems such as reducing the advance rate and safety, which in turn increases operating costs. In order to design an effective solution, it is important to accurately predict the degree of TWC; this can reduce the level of concern and have a positive effect on the design. With the development of soft computing methods, the use of deep learning algorithms and neural networks in tunnel construction has expanded in recent years. The current study aims to employ the long-short-term memory (LSTM) deep neural network predictor model to predict the TWC, based on 550 data points of observed parameters developed by collecting required data from different tunnelling projects. Among the data collected during the pre-construction and construction phases of the project, 80% is randomly used to train the model and the rest is used to test the model. Several loss functions including root mean square error (RMSE) and coefficient of determination (R²) were used to assess the performance and precision of the applied method. The results of the proposed models indicate an acceptable and reliable accuracy. In fact, the results show that the predicted values are in good agreement with the observed actual data. The proposed model can be considered for use in similar ground and tunneling conditions. It is important to note that this work has the potential to reduce the tunneling uncertainties significantly and make deep learning a valuable tool for planning tunnels.

• Structural Engineering and Mechanics
• /
• v.85 no.5
• /
• pp.635-644
• /
• 2023

The design of honeycomb sandwich structures is often challenging because these structures can be tailored from a variety of possible cores and face sheets configurations, therefore, the design of sandwich structures is characterized as a time-consuming and complex task. A data-driven computational approach that integrates the analytical method and Artificial Neural Network (ANN) is developed by the authors to rapidly predict the design of sandwich structures for a targeted maximum structural deflection. The elaborated ANN reverse design approach is applied to obtain the thickness of the sandwich core, the thickness of the laminated face sheets, and safety factors for composite sandwich structure. The required data for building ANN model were obtained using the governing equations of sandwich components in conjunction with the Monte Carlo Method. Then, the functional relationship between the input and output features was created using the neural network Backpropagation (BP) algorithm. The input variables were the dimensions of the sandwich structure, the applied load, the core density, and the maximum deflection, which was the reverse input given by the designer. The outstanding performance of reverse ANN model revealed through a low value of mean square error (MSE) together with the coefficient of determination (R2) close to the unity. Furthermore, the output of the model was in good agreement with the analytical solution with a maximum error 4.7%. The combination of reverse concept and ANN may provide a potentially novel approach in designing of sandwich structures. The main added value of this study is the elaboration of a reverse ANN model, which provides a low computational technique as well as savestime in the design or redesign of sandwich structures compared to analytical and finite element approaches.

• Transactions of the Korean hydrogen and new energy society
• /
• v.31 no.5
• /
• pp.436-443
• /
• 2020

The performance prediction model of a solid oxide fuel cell stack has been developed using deep neural network technique, one of the machine learning methods. The machine learning has been received much interest in various fields, including energy system mo- deling. Using machine learning technique can save time and cost requried in developing an energy system model being compared to the conventional method, that is a combination of a mathematical modeling and an experimental validation. Results reveal that the mean average percent error, root mean square error, and coefficient of determination (R2) range 1.7515, 0.1342, 0.8597, repectively, in maximum. To improve the predictability of the model, the pre-processing is effective and interpolative machine learning and application is more accurate than the extrapolative cases.

• Journal of Forest and Environmental Science
• /
• v.36 no.2
• /
• pp.122-132
• /
• 2020

The study examined revenue generation trend and perception of challenges facing it by forestry personnel in Akwa Ibom State, Nigeria. Data for the study was generated through primary and secondary sources. Primary sources involved the use of questionnaire which was administered to all Forest Officers and Uniformed Field Staff in all the 31 Forest Division and Headquarter in the state. Secondary sources involved collation of generated revenue from all the divisions for the study period. Data obtained were analyzed using descriptive and inferential statistics including Least square regression. The results indicated an increasing trend in forest revenue for the state statistically defined by the function y=45631x-900000000＋e (p>0.05) with a coefficient of determination of 0.7492 or 74.92%. There was also a positive correlation (r=0.866) between generated revenue and year for the 20 years under review. The mean revenue was ₦4776247.00 with the highest generated revenue (₦9823550.00) in 2014. However, majority (55.13%) of the respondents perceived revenue generation in the state to be decreasing and attributed the decline majorly to lack of mobility (16.84%) and insufficient man power (15.79%). Attitude and level of offence in the study area was perceived to be fairly cooperative (62.81%) and high (43.80%), while recruitment of more personnel (11.05%) and provision of mobility (10.03%) was considered an effective means of improving revenue generation in the state. Also, educating the people and regular patrol by forest personnel was considered as the best ways of curtailing forest offences in the area. The study recommended increased allocation of funds to the sector in addition to tackling the challenges faced by the personnel.