• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.41 no.2
• /
• pp.123-131
• /
• 2021

The buffer is a key component of an engineered barrier system that safeguards the disposal of high-level radioactive waste. Buffers are located between disposal canisters and host rock, and they can restrain the release of radionuclides and protect canisters from the inflow of ground water. Since considerable heat is released from a disposal canister to the surrounding buffer, the thermal conductivity of the buffer is a very important parameter in the entire disposal safety. For this reason, a lot of research has been conducted on thermal conductivity prediction models that consider various factors. In this study, the thermal conductivity of a buffer is estimated using the machine learning methods of: linear regression, decision tree, support vector machine (SVM), ensemble, Gaussian process regression (GPR), neural network, deep belief network, and genetic programming. In the results, the machine learning methods such as ensemble, genetic programming, SVM with cubic parameter, and GPR showed better performance compared with the regression model, with the ensemble with XGBoost and Gaussian process regression models showing best performance.

• Korean Journal of Construction Engineering and Management
• /
• v.22 no.5
• /
• pp.73-85
• /
• 2021

This study proposes an efficient management direction for Korean construction accident cases through a deep learning-based text data classification model. A deep learning model was developed, which categorizes five categories of construction accidents: fall, electric shock, flying object, collapse, and narrowness, which are representative accident types of KOSHA. After initial model tests, the classification accuracy of fall disasters was relatively high, while other types were classified as fall disasters. Through these results, it was analyzed that 1) specific accident-causing behavior, 2) similar sentence structure, and 3) complex accidents corresponding to multiple types affect the results. Two accuracy improvement experiments were then conducted: 1) reclassification, 2) elimination. As a result, the classification performance improved with 185.7% when eliminating complex accidents. Through this, the multicollinearity of complex accidents, including the contents of multiple accident types, was resolved. In conclusion, this study suggests the necessity to independently manage complex accidents while preparing a system to describe the situation of future accidents in detail.

• Korean Chemical Engineering Research
• /
• v.59 no.2
• /
• pp.209-218
• /
• 2021

First principle-based modeling studies have been performed to improve the heat exchange efficiency of ORV and optimize operation, but the heat transfer coefficient of ORV is an irregular system according to time and location, and it undergoes a complex modeling process. In this study, FNN, LSTM, and AutoML-based modeling were performed to confirm the effectiveness of data-based modeling for complex systems. The prediction accuracy indicated high performance in the order of LSTM > AutoML > FNN in MSE. The performance of AutoML, an automatic design method for machine learning models, was superior to developed FNN, and the total time required for model development was 1/15 compared to LSTM, showing the possibility of using AutoML. The prediction of NG and seawater discharged temperatures using LSTM and AutoML showed an error of less than 0.5K. Using the predictive model, real-time optimization of the amount of LNG vaporized that can be processed using ORV in winter is performed, confirming that up to 23.5% of LNG can be additionally processed, and an ORV optimal operation guideline based on the developed dynamic prediction model was presented.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.1
• /
• pp.777-785
• /
• 2021

Many utility lines are buried underground to provide various functions of the city. Because historical records are not managed systematically, damage has occurred during excavation. In addition, the demand for an underground facility management system is increasing as the aerial underground project is progressing. By attaching an electronic recognition sensor to an underground facility, such as pipelines, the management history and site conditions can be carefully managed. Therefore, in this study, electronic recognition sensors, such as BLE Beacon, UHF RFID, geomagnetic sensor, and commercial marker, were tested to analyze the strengths, weaknesses, and field applicability through a pilot project. According to the limited research results collected through two pilot projects, the installation depth is most important to demonstrate the performance of the electronic reader. In addition, because it should be used in urban areas, the influence of environmental interference should be minimized, and there should be no performance degradation over time. In the case of the geomagnetic recognizer, the effect of environmental interference was large, and performance degradation occurred over time using the BLE Beacon. In the field situation, where the installation depth can be controlled to less than 40cm, the utility of the battery-free UHF RFID was the best.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.2
• /
• pp.651-658
• /
• 2021

Because bolts with anti-loosening coatings are used mainly for joining safety-related components in automobiles, accurate automatic screening of these coatings is essential to detect defects efficiently. The performance of the convolutional neural network (CNN) used in a previous study [Identification of bolt coating defects using CNN and Grad-CAM] increased with increasing number of data for the analysis of image patterns and characteristics. On the other hand, obtaining the necessary amount of data for coated bolts is difficult, making training time-consuming. In this paper, resorting to the same VGG16 model as in a previous study, transfer learning was applied to decrease the training time and achieve the same or better accuracy with fewer data. The classifier was trained, considering the number of training data for this study and its similarity with ImageNet data. In conjunction with the fully connected layer, the highest accuracy was achieved (95%). To enhance the performance further, the last convolution layer and the classifier were fine-tuned, which resulted in a 2% increase in accuracy (97%). This shows that the learning time can be reduced by transfer learning and fine-tuning while maintaining a high screening accuracy.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.20 no.1
• /
• pp.174-192
• /
• 2021

The LIDAR sensor, which provides higher cognitive performance than cameras and radar, is difficult to apply to ADAS or autonomous driving because of its high price. On the other hand, as the price is decreasing rapidly, expectations are rising to improve existing autonomous driving functions by taking advantage of the LIDAR sensor. In level 3 autonomous vehicles, when a dangerous situation in the cognitive module occurs due to a sensor defect or sensor limit, the driver must take control of the vehicle for manual driving. If the driver does not respond to the request, the system must automatically kick in and implement a minimum risk maneuver to maintain the risk within a tolerable level. In this study, based on this background, a LIDAR-based LKS MRM algorithm was developed for the case when the normal operation of LKS was not possible due to troubles in the cognitive system. From point cloud data collected by LIDAR, the algorithm generates the trajectory of the vehicle in front through object clustering and converts it to the target waypoints of its own. Hence, if the camera-based LKS is not operating normally, LIDAR-based path tracking control is performed as MRM. The HAZOP method was used to identify the risk sources in the LKS cognitive systems. B, and based on this, test scenarios were derived and used in the validation process by simulation. The simulation results indicated that the LIDAR-based LKS MRM algorithm of this study prevents lane departure in dangerous situations caused by various problems or difficulties in the LKS cognitive systems and could prevent possible traffic accidents.

• Korean Journal of Remote Sensing
• /
• v.37 no.3
• /
• pp.627-636
• /
• 2021

Although many studies on ship detection using synthetic aperture radar (SAR) satellite images are being conducted around the world, there are still very few employing SAR microsatellites, as most of the microsatellites are optical satellites. Recently, the ICEYE and Capella Space have embarked on the development of microsatellites with SAR sensor, and similar projects are being initiated globally in line with the flow of the new space era [e.g., for the ICEYE: 18 satellites (~2021); Capella Space: 36 satellites (~2023); and the Coast Guard SAR: 32 satellites in the early development stage]. In preparation for these new systems, it is important to review the SAR microsatellite system and the recent advances in this technology. Accordingly, in this paper, the current status and characteristics of optical and SAR microsatellite constellation operation are described, and studies using them are investigated. In addition, based on the status and characteristics of the representative SAR microsatellites, specifically the ICEYE and Capella systems, methods for using SAR microsatellite data for ship detection applications are described. Our results confirm that the SAR microsatellites operate as a constellation and have the advantages of short revisit cycles and quick provision of high-resolution images. With this technology, we expect SAR microsatellites to contribute greatly to the monitoring a wide-area target vessel, in which the spatiotemporal resolution of the imagery is especially important.

• Journal of Korea Society of Waste Management
• /
• v.35 no.7
• /
• pp.670-682
• /
• 2018

Engineered nanomaterials (ENMs) can be released to humans and the environment through the generation of waste containing engineered nanomaterials (WCNMs) and the use and disposal of nano-products. Nanoparticles can also be introduced intentionally or unintentionally into waste streams. This study examined WCNMs in domestic industries, and target nanomaterials, such as silicon dioxide, titanium oxide, zinc oxide, nano silver, and carbon nanotubes (CNTs), were selected. We tested 48 samples, such as dust, sludge, ash, and by-products from manufacturing facilities and waste treatment facilities. We analyzed leaching and content concentrations for heavy metals and hazardous constituents of the waste. Chemical compositions were also measured by XRD and XRF, and the unique properties of nano-waste were identified by using a particle size distribution analyzer and TEM. The dust and sludge generated from manufacturing facilities and the use of nanomaterials showed higher concentrations of metals such as lead, arsenic, chromium, barium, and zinc. Oiled cloths from facilities using nano silver revealed high concentrations of copper, and the leaching concentrations of copper and lead in fly ash were higher than those in bottom ash. In XRF measurements at the facilities, we detected compounds such as silicon dioxide, sulfur trioxide, calcium oxide, titanium dioxide, and zinc oxide. We found several chemicals such as calcium oxide and silicon dioxide in the bottom ash of waste incinerators.

• Journal of Korean Society of Disaster and Security
• /
• v.13 no.4
• /
• pp.25-36
• /
• 2020

An increasing frequency and intensity of natural disasters have been observed due to climate change. To better prepare for these, the MOIS (ministry of the interior and safety) announced a comprehensive plan for minimizing damages associated with natural disasters, including drought and heavy snowfall. The spatial-temporal pattern of snowfall is greatly influenced by temperature and geographical features. Heavy snowfalls are often observed in Gangwon-do, surrounded by mountains, whereas less snowfall is dominant in the southern part of the country due to relatively high temperatures. Thus, snow depth data often contains zeros that can lead to difficulties in the selection of probability distribution and estimation of the parameters. A generalized mixture distribution approach to a maximum snow depth series over the southern part of Korea (i.e., Changwon, Tongyeoung, Jinju weather stations) are located is proposed to better estimate a threshold (𝛿) classifying discrete and continuous distribution parts. The model parameters, including the threshold in the mixture model, are effectively estimated within a Bayesian modeling framework, and the uncertainty associated with the parameters is also provided. Comparing to the Daegwallyeong weather station, It was found that the proposed model is more effective for the regions in which less snow depth is observed.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.12
• /
• pp.1-9
• /
• 2020

As pet dogs rapidly increase in number, abandoned and lost dogs are also increasing in number. In Korea, animal registration has been in force since 2014, but the registration rate is not high owing to safety and effectiveness issues. Biometrics is attracting attention as an alternative. In order to increase the recognition rate from biometrics, it is necessary to collect biometric images in the same form as much as possible-from the face. This paper proposes a method to determine whether a dog is facing front or not in a real-time video. The proposed method detects the dog's eyes and nose using deep learning, and extracts five types of directional face information through the relative size and position of the detected face. Then, a machine learning classifier determines whether the dog is facing front or not. We used 2,000 dog images for learning, verification, and testing. YOLOv3 and YOLOv4 were used to detect the eyes and nose, and Multi-layer Perceptron (MLP), Random Forest (RF), and the Support Vector Machine (SVM) were used as classifiers. When YOLOv4 and the RF classifier were used with all five types of the proposed face orientation information, the face recognition rate was best, at 95.25%, and we found that real-time processing is possible.