• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.667-669
• /
• 2022

최근 고도화 되고 있는 인공지능 기술은 복잡한 데이터 속에서 내재된 인사이트를 발견하여 상태 변화를 진단하고 예측하는데 활용되고 있다. 이러한 첨단 기술을 활용하여, 원자력 발전소에서 공정상태가 비정상 또는 비상 상태로 악화되기 전에 운전원이 인지할 수 있다면 공정상태를 정상으로 회복하는데 도움을 줄 수 있다. 이 논문에서는 공정상태 판단을 위한 딥러닝 모델을 활용하는 지능형 조기 경고 개념을 제안한다. 공정상태의 변화 추세를 예측하는 목적으로 사용하는 지능형 조기 경고는 기존 경보 보다 단순화된 상태 천이 메커니즘을 사용하여 운전원의 부담이 증가하지 않도록 한다. 또한, 사고 시나리오 데이터로 딥러닝 모델로 학습하고 지능형 조기 경고 화면을 구현하여 운전원을 지원하기 위한 구현방향을 제시한다.

• Journal of Korea Water Resources Association
• /
• v.36 no.3 s.134
• /
• pp.375-384
• /
• 2003

Severe drought tends to occur in almost event five years in Korea. Drought responses have been well operated in close collaboration with the central, local government and the water management authorities on the institutional framework. However, the responses are usually post-activities to a drought event. The responses often face difficulties in operating and managing process due to an absence of a drought monitoring system and drought triggers. The objective of this study is to set up drought triggers through a time-spatial interpretation of drought index and the government responses during historical drought events. Drought triggers are divided into four categories: advisory, watch, warning and emergency stage. The range and drought-impacted area of an each stage in triggers have been addressed using drought index. Furthermore, a web-based drought monitoring system is illustrated.

• Journal of the Korean Society of Safety
• /
• v.25 no.3
• /
• pp.40-44
• /
• 2010

North Korea's discharge of water from a dam into the Imjin River that flows through the inter-Korean border swept away a riverside camping site early Sunday morning, killing six people. This tragic incident might have been prevented if the North had given prior notice of the release from its Hwanggang Dam to the South. Investigations are under way to determine the reasons behind the unexpected act. This incident was a man-made disaster not least death of six people. A water level monitoring system(WLMS) of the Imjin River was installed the bridge of Pilseung that operate three public institutions. The WLMS of the A institution set up warning siren and broadcasting as the water level has been rising the bridge of Pilseung in the Imjin River. But the A institution's system was already out of side before discharged of water a dam into the Imjin River and the operators were culpable negligence. The B institution's office employee on charges of negligence that might have contributed to the tragedy and one of the A institution's employees ignored 26 warning messages on the WLMS. This tragic incident was a man-made disaster not least death of six people and might have been prevented if the WLMS was normally operated and the system operators must be worked a tight.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.4
• /
• pp.31-39
• /
• 2012

On rainy days, water films form on wet road surfaces and reduce the braking force of vehicles, which often ends up in accidents. For safe driving, the road information signage provides information on road and weather conditions warning drivers of wet road conditions. Still, current information on road conditions is neither localized nor detailed but universal. The present study used the images on CCTVs installed on roads to compare the images of road surfaces in an attempt to suggest a mechanism determining factors that hamper safe driving based on the images. In the image comparison, a normal road image taken on a sunny day is used as an original image, against which road conditions occurring on rainy days are categorized and determined on a case-by-case basis to provide drivers with early warning for the sake of safe driving.

• Journal of The Geomorphological Association of Korea
• /
• v.15 no.2
• /
• pp.25-35
• /
• 2008

Typhoon Ewiniar and Bilis followed by heavy rainfall in July 2006 triggered massive slope failures and debris flows along the Osaek valley within Seoraksan National Park. Since national road 44 is constructed along the fault-line, the susceptibility of hazard in the area is very high. Debris flows in Osaekcheon are mobilized from landslides near the ridgelines and peaks when heavy rainfall elevates pore pressure and adds weight to the hillslopes, causing failure. Stream flows falling onto the existing colluvium or channel-margin deposits also trigger debris flows. Steep slopes constructed along the road and thin regolith in the slope is the main reason for the landslide in the upper stream. In middle reaches of stream, under-fit drainage utilities and narrow bridges cause the overflow, this then triggers debris flow. Overflowing and erosion in the channel margin deposits is main reasons for the debris flow. The intensities and frequencies of heavy rainfall are certain to increase, so early warning and management system for the landslide-related hazard is urgently needed.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.12
• /
• pp.3904-3922
• /
• 2022

As a research hotspot, pedestrian detection has a wide range of applications in the field of computer vision in recent years. However, current pedestrian detection methods have problems such as insufficient detection accuracy and large models that are not suitable for large-scale deployment. In view of these problems mentioned above, a lightweight pedestrian detection and early warning method using a new model called you only look once (Yolov5) is proposed in this paper, which utilizing advantages of Yolov5s model to achieve accurate and fast pedestrian recognition. In addition, this paper also optimizes the loss function of the batch normalization (BN) layer. After sparsification, pruning and fine-tuning, got a lot of optimization, the size of the model on the edge of the computing power is lower equipment can be deployed. Finally, from the experimental data presented in this paper, under the training of the road pedestrian dataset that we collected and processed independently, the Yolov5s model has certain advantages in terms of precision and other indicators compared with traditional single shot multiBox detector (SSD) model and fast region-convolutional neural network (Fast R-CNN) model. After pruning and lightweight, the size of training model is greatly reduced without a significant reduction in accuracy, and the final precision reaches 87%, while the model size is reduced to 7,723 KB.

• Journal of Advanced Navigation Technology
• /
• v.28 no.2
• /
• pp.185-192
• /
• 2024

The aviation has improved safety through technological innovation and strengthened flight safety through safety regulations and supervision by aviation authorities. As the industry's safety approach has evolved into a systematic approach to the aircraft system, airlines have established a safety management system. Technical defects or abnormal data in an aircraft can be warning signs that could lead to an accident, and the risk of an accident can be reduced by identifying and responding to these signs early. Therefore, management of abnormal warning signs is an essential element in promoting data-based decision-making and enhancing the operational efficiency and safety level of airlines. In this study, we present a model to statistically analyze quick access recorder (QAR) flight data in the preliminary analysis stage to analyze the patterns and causes of crab landing events that can lead to runway departures when landing an aircraft, and provide a precursor to a landing event. We aim to identify signs and causes and contribute to increasing the efficiency of safety management.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2009.09a
• /
• pp.117-129
• /
• 2009

This paper studied the safety management network system of infrastructure which constructed smart sensors, closed-circuit television(CCTV) and monitoring system. This safety management of infrastructure applied to bridge, cut slop and tunnel, embankment etc. The system applied to technologies of standardization guidelines, data acquirement technologies, data analysis and judgment technologies, system integration setup technology, and IT technologies. It was constructed safety management network system of various infrastructure to improve efficient management and operation for many infrastructure. Integrated safety management network system of infrastructure consisted of the real-time structural health monitoring system of each infrastructure, integrated control center, measured data transmission using i of tet web-based, collecting data using sf ver, early alarm system which the dangerous event of infrastructure occurred. Integrated control center consisted of conference room, control room to manage and analysis the data, server room to present the measured data and to collect the raw data. Early alarm system proposed realization of warning and response within 5 minute or less through development of sensor-based progress report and propagation automation system using the media such as MMS, VMS, EMS, FMS, SMS and web services of report and propagation. Based on this, the most effective u-Infrastructure Safety Management System is expected to be stably established at a less cost, thus making people's life more comfortable. Information obtained from such systems could be useful for maintenance or structural safety evaluation of existing structures, rapid evaluation of conditions of damaged structures after an earthquake, estimation of residual life of structures, repair and retrofitting of structures, maintenance, management or rehabilitation of historical structures.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.19 no.3
• /
• pp.153-163
• /
• 2017

The objective of this study is to enhance the model's speed of estimating weather variables (e.g., minimum/maximum temperature, sunshine hour, PRISM (Parameter-elevation Regression on Independent Slopes Model) based precipitation), which are applied to the Agrometeorological Early Warning System (http://www.agmet.kr). The current process of weather estimation is operated on high-performance multi-core CPUs that have 8 physical cores and 16 logical threads. Nonetheless, the server is not even dedicated to the handling of a single county, indicating that very high overhead is involved in calculating the 10 counties of the Seomjin River Basin. In order to reduce such overhead, several cache and parallelization techniques were used to measure the performance and to check the applicability. Results are as follows: (1) for simple calculations such as Growing Degree Days accumulation, the time required for Input and Output (I/O) is significantly greater than that for calculation, suggesting the need of a technique which reduces disk I/O bottlenecks; (2) when there are many I/O, it is advantageous to distribute them on several servers. However, each server must have a cache for input data so that it does not compete for the same resource; and (3) GPU-based parallel processing method is most suitable for models such as PRISM with large computation loads.

• Journal of Korean Society of Environmental Engineers
• /
• v.34 no.12
• /
• pp.828-834
• /
• 2012

The ecotoxicity detection device for preliminary test (Test jig) was manufactured to develop the biological early warning system using Vibrio fischeri. In this study, the ecotoxicity detection charateristics of the Test jig was investigated for 6 heavy metals (Cr, Zn, Pb, Cd, Cu and Hg). It was observed that relative luminescence unit (RLU) of Vibrio fischeri constantly decreased by the concentrations of the tested heavy metals. In contrast with other heavy metals, RLUs of Pb and Hg constantly decreased even at low concentrations. RLU of Hg drastically decreased when its concentration increased from 0.13 mg/L to 0.25 mg/L. $EC_{50}$ values of Cr, Zn, Pb and Cd gradually decreased with exposure time, whereas there was no significant change in $EC_{50}$ values of Cu and Hg with time. On the other hand, $EC_{50}$ values between the Test jig and Reference device were compared to evaluate the ecotoxicity detection performance of the Test jig. No big difference was found in $EC_{50}$ vlaues between the two devices, indicating that the Test jig could be applied as the ecotoxicity detection device for heavy metals.