• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.232-232
• /
• 2021

수재해 저감과 예방을 위해서는 공간적 변동성을 반영한 정확한 면적 강우량의 측정은 필수적이다. 이러한 요구에 부응하기 위해 24 GHz 이중편파 전자파를 기반으로 소규모 공간 범위에 대해 저고도의 지상 강우를 30 m의 거리해상도로 관측할 수 있는 전파강수계가 개발되었다. 전파강수계는 시제품이 개발된 이래로 한국건설기술연구원 연천센터와 국내 여러 현장과 인도네시아 등에서 시험을 실시하였다. 임상훈 등(2020)은 전파강수계의 반사도와 비차등위상차를 이용한 강우 추정식을 개발하여 연천 및 거제 관측 자료에 적용한 바 있다. 본 논문에서는 연천센터에 분산 배치한 우량계 자료를 이용하여 전파강수계의 강우 공간분포 측정 성능을 평가하였다. 공간우량계는 15대 중 음영구역 바깥에서 정상 작동한 7개의 0.5mm급 우량계 자료와 핏게이지에 있는 0.2mm급 우량계 2대가 비교에 사용되었다. 전파강수계 강우강도는 비교 위치에 해당하는 점 주변의 레이 방향 5개(37.5 m에 해당) 및 방위각 방향 5개 게이트 등 총 25개의 복셀에서 산출된 강우 정보를 평균하여 비교하였다. 정확도는 지상우량계를 참값으로 보고 MAE(Mean absolute error)로 평가하였다. 그 결과 평균 4.2%의 오차를 보였으며, 우량계의 오차를 ±5%로 가정할 경우 3.3~7.9%로 나타났다. 전파강수계의 누적 강우량 값은 강우계에 비해 작은데, 이는 지속적인 관측을 통해 강우 산정의 정확도를 개선하는 것이 필요함을 의미한다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.136-136
• /
• 2021

In this study, a deep convolutional neural network (DCNN) model is proposed for short-term precipitation forecasting using weather radar-based images. The DCNN model is a combination of convolutional neural networks, autoencoder neural networks, and U-net architecture. The weather radar-based image data used here are retrieved from competition for rainfall forecasting in Korea (AI Contest for Rainfall Prediction of Hydroelectric Dam Using Public Data), organized by Dacon under the sponsorship of the Korean Water Resources Association in October 2020. This data is collected from rainy events during the rainy season (April - October) from 2010 to 2017. These images have undergone a preprocessing step to convert from weather radar data to grayscale image data before they are exploited for the competition. Accordingly, each of these gray images covers a spatial dimension of 120×120 pixels and has a corresponding temporal resolution of 10 minutes. Here, each pixel corresponds to a grid of size 4km×4km. The DCNN model is designed in this study to provide 10-minute predictive images in advance. Then, precipitation information can be obtained from these forecast images through empirical conversion formulas. Model performance is assessed by comparing the Score index, which is defined based on the ratio of MAE (mean absolute error) to CSI (critical success index) values. The competition results have demonstrated the impressive performance of the DCNN model, where the Score value is 0.530 compared to the best value from the competition of 0.500, ranking 16^th out of 463 participating teams. This study's findings exhibit the potential of applying the DCNN model to short-term rainfall prediction using weather radar-based images. As a result, this model can be applied to other areas with different spatiotemporal resolutions.

• Steel and Composite Structures
• /
• v.49 no.3
• /
• pp.337-348
• /
• 2023

Successive wetting and drying cycles of concrete due to weather changes can endanger the safety of engineering structures over time. Considering wetting and drying cycles in concrete tests can lead to a more correct and reliable design of engineering structures. This study aims to provide a model that can be used to estimate the resistance properties of concrete under different wetting and drying cycles. Complex sample preparation methods, the necessity for highly accurate and sensitive instruments, early sample failure, and brittle samples all contribute to the difficulty of measuring the strength of concrete in the laboratory. To address these problems, in this study, the potential ability of six machine learning techniques, including ANN, SVM, RF, KNN, XGBoost, and NB, to predict the concrete's tensile strength was investigated by applying 240 datasets obtained using the Brazilian test (80% for training and 20% for test). In conducting the test, the effect of additives such as glass and polypropylene, as well as the effect of wetting and drying cycles on the tensile strength of concrete, was investigated. Finally, the statistical analysis results revealed that the XGBoost model was the most robust one with R2 = 0.9155, mean absolute error (MAE) = 0.1080 Mpa, and variance accounted for (VAF) = 91.54% to predict the concrete tensile strength. This work's significance is that it allows civil engineers to accurately estimate the tensile strength of different types of concrete. In this way, the high time and cost required for the laboratory tests can be eliminated.

• Journal of Korean Society for Quality Management
• /
• v.52 no.1
• /
• pp.43-56
• /
• 2024

Purpose: The injection molding process, crucial for plastic shaping, encounters difficulties in sustaining product quality when replacing injection machines. Variations in machine types and outputs between different production lines or factories increase the risk of quality deterioration. In response, the study aims to develop a system that optimally adjusts conditions during the replacement of injection machines linked to molds. Methods: Utilizing a dataset of 12 injection process variables and 52 corresponding sensor variables, a predictive model is crafted using Decision Tree, Random Forest, and XGBoost. Model evaluation is conducted using an 80% training data and a 20% test data split. The dependent variable, classified into five characteristics based on temperature and pressure, guides the prediction model. Bayesian optimization, integrated into the selected model, determines optimal values for process variables during the replacement of injection machines. The iterative convergence of sensor prediction values to the optimum range is visually confirmed, aligning them with the target range. Experimental results validate the proposed approach. Results: Post-experiment analysis indicates the superiority of the XGBoost model across all five characteristics, achieving a combined high performance of 0.81 and a Mean Absolute Error (MAE) of 0.77. The study introduces a method for optimizing initial conditions in the injection process during machine replacement, utilizing Bayesian optimization. This streamlined approach reduces both time and costs, thereby enhancing process efficiency. Conclusion: This research contributes practical insights to the optimization literature, offering valuable guidance for industries seeking streamlined and cost-effective methods for machine replacement in injection molding.

• The Journal of Information Systems
• /
• v.33 no.3
• /
• pp.187-205
• /
• 2024

Purpose A prediction model for occupational injuries can support more proactive, efficient, and effective policy-making. This study aims to develop a model that predicts the severity of occupational injuries, classified into 15 disability grades in South Korea, using machine learning techniques applied to COMWEL data. The primary goal is to improve prediction accuracy, offering an advanced tool for early intervention and evidence-based policy implementation. Design/methodology/approach The data analyzed in this study consists of 290,157 administrative records of occupational injury cases collected between 2018 and 2020 by the Korea Workers' Compensation & Welfare Service, based on the 'Workers' Compensation Insurance Application Form' submitted for occupational injury treatment. Four machine learning models - Decision Tree, DNN, XGBoost, and LightGBM - were developed and their performances compared to identify the optimal model. Additionally, the Permutation Feature Importance (PFI) method was used to assess the relative contribution of each variable to the model's performance, helping to identify key variables. Findings The DNN algorithm achieved the lowest Mean Absolute Error (MAE) of 0.7276. Key variables for predicting disability grades included the severity index, primary disease code, primary disease site, age at the time of the injury, and industry type. These findings highlight the importance of early policy intervention and emphasize the role of both medical and socioeconomic factors in model predictions. The academic and policy implications of these results were also discussed.

• Journal of Korean Society of Environmental Engineers
• /
• v.29 no.6
• /
• pp.706-714
• /
• 2007

This study suggested a method for prediction of residual chlorine and THMs in water distribution system by measurement of residual chlorine, THMs, and other parameters, estimation of chlorine decay coefficients and THM formation coefficients, and simulation of water qualities using pipe network analysis. Bulk decay coefficients of parallel first-order were obtained by bottle tests, and pipe wall decay coefficients of first-order were estimated through evaluation of 5 models, which showed the lowest values of 0.03 for MAE(mean absolute error) and 0.037 MAE in comparison with the observed in field. And bottle tests were conducted to model first-order reaction of THM formation by nonlinear least square regression and the resultant coefficients were compared with the observed in field. As a result, the coefficients of determination$(R^2)$ for the observed and the predicted values were 0.98 in September and 0.82 in November, and the formation of THMs was predicted by modeling.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.21 no.6
• /
• pp.87-96
• /
• 2022

The traffic vulnerable, including elderly pedestrians, have a relatively low walking speed and slow cognitive response time due to reduced physical ability. Although a smart crossing system has been developed and operated to improve problem, it is difficult to operate a signal that reflects the appropriate walking speed for each pedestrian. In this study, a neural network model and a multiple regression model-based traversing speed estimation model were developed using image information collected in an area with a high percentage of traffic vulnerability. to support the provision of optimal walking signals according to real-time traffic weakness. actual traffic data collected from the urban traffic network of Paju-si, Gyeonggi-do were used. The performance of the model was evaluated through seven selected indicators, including correlation coefficient and mean absolute error. The multiple linear regression model had a correlation coefficient of 0.652 and 0.182; the neural network model had a correlation coefficient of 0.823 and 0.105. The neural network model showed higher predictive power.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.32 no.1B
• /
• pp.9-20
• /
• 2012

Recently to overcome limitations of conceptual, hydrological and physics based models for flood stage forecasting, multiple linear regression model as one of data-driven models have been widely adopted for forecasting flood streamflow(stage). The objectives of this study are to compare performance of different multiple linear regression models according to regression coefficient estimation methods and determine most effective multiple linear regression flood stage forecasting models. To do this, the time scale was determined through the autocorrelation analysis of input data and different flood stage forecasting models developed using regression coefficient estimation methods such as LS(least square), WLS(weighted least square), SPW(stepwise) was applied to flood events in Jungrang stream. To evaluate performance of established models, fours statistical indices were used, namely; Root mean square error(RMSE), Nash Sutcliffe efficiency coefficient (NSEC), mean absolute error (MAE), adjusted coefficient of determination($R^{*2}$). The results show that the flood stage forecasting model using SPW(stepwise) parameter estimation can carry out the river flood stage prediction better in comparison with others, and the flood stage forecasting model using LS(least square) parameter estimation is also found to be slightly better than the flood stage forecasting model using WLS(weighted least square) parameter estimation.

• Journal of Wetlands Research
• /
• v.13 no.3
• /
• pp.547-565
• /
• 2011

Different types of schemes have been used in stage prediction involving conceptual and physical models. Nevertheless, none of these schemes can be considered as a single superior model. To overcome disadvantages of existing physics based rainfall-runoff models for stage predicting because of the complexity of the hydrological process, recently the data-derived models has been widely adopted for predicting flood stage. The objective of this study is to evaluate model performance for stage prediction of the Neuro-Fuzzy and regression analysis stage prediction models in these data-derived methods. The proposed models are applied to the Wangsukcheon in Han river watershed. To evaluate the performance of the proposed models, fours statistical indices were used, namely; Root mean square error(RMSE), Nash Sutcliffe efficiency coefficient(NSEC), mean absolute error(MAE), adjusted coefficient of determination($R^{*2}$). The results show that the Neuro-Fuzzy stage prediction model can carry out the river flood stage prediction more accurately than the regression analysis stage prediction model. This study can greatly contribute to the construction of a high accuracy flood information system that secure lead time in medium and small streams.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.35 no.3
• /
• pp.579-588
• /
• 2015

Recent climage changes have resulted in increases in rainfall intensity and flood frequency as well as the risk of flood damage due to typhoons during the summer season. Water-friendly facilities such as ecological parks and sports facilities have been established on floodplains of rivers since the river improvement project was implemented and increases in the flood levels of rivers due to typhoons can lead to direct flood damage to such facilities. To analyze the hydraulic influence of these water-friendly facilities on floodplains or to evaluate their stability, numerical analysis should be performed in advance. In addition, it is crucial to address the drying and wetting processes generated by water level fluctuations. This study uses a Nays2D model, which analyzes drying and wetting, to examine its applicability to simple terrain in which such fluctuations occur and to natural rivers in which drying occurs. The results of applying this model to sites of actual typhoon events are compared with values measured at water level observatories. Through this comparison, it is determined that values of coefficient of determination ($R^2$), mean absolute error (MAE), and root-mean-square error (RMSE) are 0.988, 0.208, and 0.239, respectively, thus showing a statistically high correlation. In addition, the results are used to calculate flood risk indices for evaluation of such risk for water-friendly facilities constructed on floodplains.