• 한국수자원학회논문집
• /
• 제56권10호
• /
• pp.679-690
• /
• 2023

누수는 용수공급시스템 내에서 발생할 수 있는 대표적인 비정상상황 중 하나이다. 누수는 관로가 매설된 이후부터 잠재적으로 발생할 수 있으며 발생 직후부터 즉시 경제적 및 수리학적 피해를 입을 수 있기 때문에 이를 적시에 감지하고 탐지하는 것이 중요하다. 하지만 시스템이 지하에 매설되어 있어 이를 빠르게 인지하는 것은 쉽지 않으며 인지한다 하여도 복구하기 위해서는 상대적으로 많은 가용자산이 요구된다. 따라서 다중 누수가 발생할 시 누수규모 및 위치에 따라 복구 우선순위에 대한 우선순위를 선정해야 할 필요성이 있으며 최적의 복구전략이 도출되어 이를 수행할 시 시스템의 탄력성 측면에 있어 유리함을 가질 수 있다. 본 연구에서는 프로그램 기반 모의 누수를 발생시켜 비정상상황 시나리오를 구축하였으며 이에 따라 딥러닝 기반 모델로 누수탐사를 수행하였다. 탐사 결과로 얻어지는 누수위치와 누수량은 이 후 누수복구 우선순위를 위한 요소로써 활용되며 타 요소와 함께 최적의 누수복구 시나리오를 도출하였다.

• 지구물리와물리탐사
• /
• 제19권4호
• /
• pp.200-211
• /
• 2016

횡단 전위 배열(cross potential array)과 동일열 전위배열(direct potential array) 자료를 뒷받침하여 다양한 수변 구조물의 누수 경로 예측을 가능하게 하는 배열을 제시하고 이를 D-Lux array 라고 명명하였다. 또한 D-Lux array 자료를 색으로 가시화한 하나의 행렬로 정리하여 D-Lux view라고 제시하고 D-Lux view에서 관찰되는 저 전위차 이상대의 위치로 누수 구역의 위치를 해석하였다. D-Lux view의 보다 직관적인 해석을 위해서 D-Lux array 자료와 동일열 전위 배열 자료를 함께 사용하여 각 지점들 사이의 전위차 자료를 각 지점의 전위값으로 역산하고 이를 이용하여 등전위 분포도를 작성하였다. 등전위 분포도는 그래프나 D-Lux view에서 알 수 없었던 누수의 유입구, 유출구 뿐만 아니라 경로까지 예측 가능하게 하였다. 수조 실험과 수치 해석으로 예비 탐사를 실시한 후 현장 탐사로 콘크리트 보와 필 댐에 대한 적용이 이루어졌다. 그 결과, 콘크리트 보와 필 댐에 대해 누수 경로 탐지가 가능함을 확인하였다.

• Nuclear Engineering and Technology
• /
• 제53권2호
• /
• pp.532-548
• /
• 2021

Nuclear power plants contain several monitoring systems that can identify the in-vessel phenomena of a severe accident (SA). Though a lot of analysis and research is carried out on SA, right from the development of the nuclear industry, not all the possible circumstances are taken into consideration. Therefore, to improve the efficacy of the safety of nuclear power plants, additional analytical studies are needed that can directly monitor severe accident phenomena. This paper presents an interacting multiple model (IMM) based fault detection and diagnosis (FDD) approach for the identification of in-vessel phenomena to provide the accident propagation information using reactor vessel (RV) out-wall temperature distribution during severe accidents in a nuclear power plant. The estimation of wall temperature is treated as a state estimation problem where the time-varying wall temperature is estimated using IMM employing three multiple models for temperature evolution. From the estimated RV out-wall temperature and rate of temperature, the in-vessel phenomena are identified such as core meltdown, corium relocation, reactor vessel damage, reflooding, etc. We tested the proposed method with five different types of SA scenarios and the results show that the proposed method has estimated the outer wall temperature with good accuracy.

• 한국농공학회:학술대회논문집
• /
• 한국농공학회 1998년도 학술발표회 발표논문집
• /
• pp.331-337
• /
• 1998

This study was carried out to derive optimal design floods by Weibull-3 distribution with the annual maximum series at seven watersheds along Man, Nagdong, Geum, Yeongsan and Seomjin river systems. Adequacy for the analysis of flood data used in this study was acknowledged by the tests of Independence, Homogeneity, detection of Outliers. Parameters were estimated by the Methods of Moments and L-Moments. Design floods obtained by Methods of Moments and L-Moments using different methods for plotting positions in Weibull-3 distribution were compared by the rotative mean error and relative absolute error. It has shown that design floods derived by the method of L-moments using Weibull plotting position formula in Weibull-3 distribution are much closer to those of the observed data in comparison with those obtained by method of moments using different formulas for plotting positions in view of relative mean and relative absolute error.

• 전기학회논문지
• /
• 제61권11호
• /
• pp.1758-1764
• /
• 2012

A large number of wire rope has been used in various inderstiries as Cranes and Elevators from expanding the scale of the industrial market. But now, the management of wire rope is used as manually operated by rope replacement from over time or after the accident.It is caused to major accidents as well as economic losses and personal injury. Therefore its time to need periodic fault diagnosis of wire rope or supply of real-time monitoring system. Currently, there are several methods has been reported for fault diagnosis method of the wire rope, to find out the feature point from extracting method is becoming more common compared to time wave and model-based system. This method has implemented a deterministic modeling like the observer and neural network through considering the state of the system as a deterministic signal. However, the out-put of real system has probability characteristics, and if it is used as a current method on this system, the performance will be decreased at the real time. And if the random noise is occurred from unstable measure/experiment environment in wire rope system, diagnostic criterion becomes unclear and accuracy of diagnosis becomes blurred. Thus, more sophisticated techniques are required rather than deterministic fault diagnosis algorithm. In this paper, we developed the fault diagnosis of the wire rope using probability density estimation techniques algorithm. At first, The steady-state wire rope fault signal detection is defined as the probability model through probability distribution estimate. Wire rope defects signal is detected by a hall sensor in real-time, it is estimated by proposed probability estimation algorithm. we judge whether wire rope has defection or not using the error value from comparing two probability distribution.

• 한국동물위생학회지
• /
• 제41권4호
• /
• pp.245-250
• /
• 2018

Lawsonia intracellularis is the pathogenic agent of porcine proliferative enteritis (PPE). The bacterial pathogen infects the intestinal crypt cells which causes hyperplasia of the infected cells and leads to the process of intestinal pathogenesis. PPE includes some clinical maninfestations, including acute hemorrhagic diarrhea with sudden death in growing pigs and porcine intestinal adenomatosis, to a chronic diarrhea with reduced productivity of the infected pigs. The purpose of the present studies were carried out to determine L. intracellularis in livestock transport car of slaughterhouse. Distribution of L. intracellularis in livestock transport car were conducted using real-time polymerase chain reaction (real-time PCR) testing method, total 300 samples. Of 300 samples, 119 (39.7%) were detected as positive to L. intracellularis in livestock transport car. In seasonal analysis, 42 (28.0%) out of 150 samples in spring and summer season. 77 (51.3%) out of 150 sample in autumn and winter season. In regional analysis, 53 (88.3%) out of 60 cars and the detection ratio showed that regional variation in livestock transport car.

• 인터넷정보학회논문지
• /
• 제21권5호
• /
• pp.57-65
• /
• 2020

At present, the existing virus recognition systems usually use signature approach to detect malicious executable files, but these methods often fail to detect new and invisible malware. At the same time, some methods try to use more general features to detect malware, and achieve some success. Moreover, machine learning-based approaches are applied to detect malware, which depend on features extracted from malicious codes. However, the different distribution of features oftraining and testing datasets also impacts the effectiveness of the detection models. And the generation oflabeled datasets need to spend a significant amount time, which degrades the performance of the learning method. In this paper, we use transfer learning to detect new and previously unseen malware. We first extract the features of Portable Executable (PE) files, then combine transfer learning training model with KNN approachto detect the new and unseen malware. We also evaluate the detection performance of a classifier in terms of precision, recall, F1, and so on. The experimental results demonstrate that proposed method with high detection rates andcan be anticipated to carry out as well in the real-world environment.

• Journal of Biosystems Engineering
• /
• 제41권3호
• /
• pp.263-272
• /
• 2016

Purpose: Pulse crop damage due to wild birds is a serious problem, to the extent that the rate of damage during the period of time between seeding and the stage of cotyledon reaches 45.4% on average. This study investigated a method of fundamentally blocking birds from eating crops by conducting vinyl mulching after seeding and identifying the growing locations for beans to perform punching. Methods: Infrared (IR) sensors that could measure the temperature without contact were used to recognize the locations of soybean cotyledons below vinyl mulch. To expand the measurable range, 10 IR sensors were arranged in a linear array. A sliding mechanical device was used to reconstruct the two-dimensional spatial variance information of targets. Spatial interpolation was applied to the two-dimensional temperature distribution information measured in real time to improve the resolution of the bean coleoptile locations. The temperature distributions above the vinyl mulch for five species of soybeans over a period of six days from the appearance of the cotyledon stage were analyzed. Results: During the experimental period, cases where bean cotyledons did and did not come into contact with the bottom of the vinyl mulch were both observed, and depended on the degree of growth of the bean cotyledons. Although the locations of bean cotyledons could be estimated through temperature distribution analyses in cases where they came into contact with the bottom of the vinyl mulch, this estimation showed somewhat large errors according to the time that had passed after the cotyledon stage. The detection results were similar for similar types of crops. Thus, this method could be applied to crops with similar growth patterns. According to the results of 360 experiments that were conducted (five species of bean ${\times}$ six days ${\times}$ four speed levels ${\times}$ three repetitions), the location detection performance had an accuracy of 36.9%, and the range of location errors was 0-4.9 cm (RMSE = 3.1 cm). During a period of 3-5 days after the cotyledon stage, the location detection performance had an accuracy of 59% (RMSE = 3.9 cm). Conclusions: In the present study, to fundamentally solve the problem of damage to beans from birds in the early stage after seeding, a working method was proposed in which punching is carried out after seeding, thereby breaking away from the existing method in which seeding is carried out after punching. Methods for the accurate detection of soybean growing locations were studied to allow punching to promote the continuous growth of soybeans that had reached the cotyledon stage. Through experiments using multiple IR sensors and a sliding mechanical device, it was found that the locations of the crop could be partially identified 3-5 days after reaching the cotyledon stage regardless of the kind of pulse crop. It can be concluded that additional studies of robust detection methods considering environmental factors and factors for crop growth are necessary.

• 대한수학회논문집
• /
• 제16권3호
• /
• pp.459-485
• /
• 2001

The aim of EIT (electrical impedance tomography) system is to image cross-section conductivity distribution of a human body by means of both generating and sensing electrodes attached on to the surface of the body, where currents are injected and voltages are measured. EIT has been suffered from the severe ill-posedness which is caused by the inherent low sensitivity of boundary measurements to any changes of internal tissue conductivity values. With a limited set of current-to-voltage data, figuring out full structure of the conductivity distribution could be extremely difficult at present time, so it could be worthwhile to extract some necessary partial information of the internal conductivity. We try to extract some key patterns of current-to-voltage data that furnish some core information on the conductivity distribution such s location and size. This overview provides our recent observation on the location search and the size estimation.

• 수산해양기술연구
• /
• 제56권4호
• /
• pp.384-394
• /
• 2020

This paper proposes an outlier detection model based on machine learning that can diagnose the presence or absence of major engine parts through unsupervised learning analysis of main engine big data of a ship. Engine big data of the ship was collected for more than seven months, and expert knowledge and correlation analysis were performed to select features that are closely related to the operation of the main engine. For unsupervised learning analysis, ensemble model wherein many predictive models are strategically combined to increase the model performance, is used for anomaly detection. As a result, the proposed model successfully detected the anomalous engine status from the normal status. To validate our approach, clustering analysis was conducted to find out the different patterns of anomalies the anomalous point. By examining distribution of each cluster, we could successfully find the patterns of anomalies.