In this study, the positions of Cs-137 gamma ray source are estimated from the plastic scintillating fiber bundle sensor with length of 5 m, using machine learning data analysis. Seven strands of plastic scintillating fibers are bundled by black shrink tube and two photomultiplier tubes are used as a gamma ray sensing and light measuring devices, respectively. The dose rate of Cs-137 used in this study is 6 μSv·h−1. For the machine learning modeling, Keras framework in a Python environment is used. The algorithm chosen to construct machine learning model is regression with 15,000 number of nodes in each hidden layer. The pulse-shaped signals measured by photomultiplier tubes are saved as discrete digits and each pulse data consists of 1,024 number of them. Measurements are conducted separately to create machine learning data used in training and test processes. Measurement times were different for obtaining training and test data which were 1 minute and 5 seconds, respectively. It is because sufficient number of data are needed in case of training data, while the measurement time of test data implies the actual measuring time. The machine learning model is designated to estimate the source positions using the information about time difference of the pulses which are created simultaneously by the interaction of gamma ray and plastic scintillating fiber sensor. To evaluate whether the double-trained machine learning model shows enhancement in accuracy of source position estimation, the reference model is constructed using training data with one-time learning process. The double-trained machine learning model is designed to construct first model and create a second training data using the training error and predetermined coefficient. The second training data are used to construct a final model. Both reference model and double-trained models constructed with different coefficients are evaluated with test data. The evaluation result shows that the average values calculated for all measured position in each model are different from 7.21 to 1.44 cm. As a result, by constructing the double-trained machine learning model, the final accuracy shows 80% of improvement ratio. Further study will be conducted to evaluate whether the double-trained machine learning model is applicable to other data obtained from measurement of gamma ray sources with different energy and set a methodology to find optimal coefficient.

Barrier effect model developed by CRIEPI is used for the estimation of rate of radioactive material release from a transport cask submerged in the ocean. If the containment boundary of cask is broken in an accident during maritime transportation, the sea water comes into the cask cavity and the leaching of radioactive material occurs. If the release of radioactive material thorough the opening of the containment boundary of cask is less than the leaching rate of the radioactive material inside the cask, then the release rate is controlled by the saturation limit of the sea water inside the cask cavity. In this study, the release rate estimation using the barrier effect model is compared with the model used in other codes, such as MARINRAD. And by parameter study, important factors that affect the releaser rate are identified and prioritized. It is shown that the gap generated in the containment boundary is the key parameter that determine the release rate of the radioactive material and the leaching rate is the dominant parameter to determine the saturation time of the cavity sea water.

부공력감쇠는 풍직각방향의 와류공진을 예측하는데 있어서 매우 중요한 요소이다. 부공력감쇠는 진동유발하중 또는 피드백 하중을 구성하는 주요인자로 와류진동이 급격히 발현되는 현상을 설명하는 도구이기도 하다. 본 연구에서는 공력감쇠의 수학적 모델 을 제시하고 와류유발하중모델과 함께 와류진동을 예측하는 프러세스를 제안한다. 직사각형단면에 대한 공기력진동실험을 수행하여, 계측된 가속도로부터 공력감쇠와 와류유발하중을 추정하고 이에 기반하여 공력감쇠모델과 와류유발하중모델을 구축하는 과정을 다룬 다. 최종적으로 공력감쇠모델과 와류유발하중 모델에 대한 재해석을 통하여 가속도응답을 구하고 계측된 가속도와 비교하여 모델의 진동예측성능을 평가한다. 본 연구에서 제안된 와류하중모델의 진동예측성능을 평가한 결과 안정적이며, 신뢰도가 높은 와류진동예측 이 가능함을 알 수 있었다.

해운 시황을 예측하는 것은 중요한 문제이다. 투자 방식의 결정, 선대 편성 방법, 운임 등을 결정하기 위한 판단 근거가 되며 이는 기업의 이익과 생존에 큰 영향을 미치기 때문이다. 이를 위해 본 연구에서는 기계학습 모델인 장단기 메모리 및 간소화된 장단기 메모리 구조의 Gated Recurrent Units를 활용하여 컨테이너선의 해상운임 예측 모델을 제안한다. 운임 예측 대상은 중국 컨테이너 운임지수 (CCFI)이며, 2003년 3월부터 2020년 5월까지의 CCFI 데이터를 학습에 사용하였다. 각 모델에 따라 2020년 6월 이후의 CCFI를 예측한 후 실 제 CCFI와 비교, 분석하였다. 실험 모델은 하이퍼 파라메터의 설정에 따라 총 6개의 모델을 설계하였다. 또한 전통적인 분석 방법과의 성 능을 비교하기 위해 ARIMA 모델도 실험에 추가하였다. 최적 모델은 두 가지 방법에 따라 선정하였다. 첫 번째 방법으로 각 모델을 10회 반복 실험하여 얻은 RMSE의 평균값이 가장 작은 모델을 선정하는 것이다. 두 번째 방법으로는 모든 실험에서 가장 낮은 RMSE를 기록한 모델을 선정하는 것이다. 실험 결과 전통적 시계열 예측모델인 ARIMA 모델과 비교하여 딥러닝 모델의 정확도를 입증하였으며, 정확한 예측모델을 통해 운임 변동의 위험관리 능력을 제고시키는데 기여했다. 반면 코로나19와 같은 외부 효과에 따른 운임의 급격한 변화상황이 발생한 경우, 예측모델의 정확도가 감소하는 한계점을 나타냈다. 제안된 모델 중 GRU1 모델이 두 가지 평가 방법 모두에서 가장 낮은 RMSE(69.55, 49.35)를 기록하며 최적 모델로 선정되었다.

PURPOSES : For vehicle-alone accidents with a high mortality rate, it is necessary to analyze the factors influencing the severity of roadside fixed-object traffic accidents. METHODS : A total of 313 roadside fixed obstacle traffic accidents, variables related to fixed obstacles, and variables related to road geometry were collected. The estimation model was constructed with data collected using an ordinal probit regression model. RESULTS : Piers, vertical slopes, and distances between roads and objects were the primary causes of increased accident severity. CONCLUSIONS : Countermeasures, such as object removal, relocation, clear zones, frangibles, breakaway poles, etc., are required for accident-prone or dangerous points.

Identifying the water circulation status is one of the indispensable processes for watershed management in an urban area. Recently, various water circulation models have been developed to simulate the water circulation, but it takes a lot of time and cost to make a water circulation model that could adapt the characteristics of the watershed. This paper aims to develop a water circulation state estimation model that could easily calculate the status of water circulation in an urban watershed by using multiple linear regression analysis. The study watershed is a watershed in Seoul that applied the impermeable area ratio in 1962 and 2000. And, It was divided into 73 watersheds in order to consider changes in water circulation status according to the urban characteristic factors. The input data of the SHER(Similar Hydrologic Element Response) model, a water circulation model, were used as data for the urban characteristic factors of each watershed. A total of seven factors were considered as urban characteristic factors. Those factors included annual precipitation, watershed area, average land-surface slope, impervious surface ratio, coefficient of saturated permeability, hydraulic gradient of groundwater surface, and length of contact line with downstream block. With significance probabilities (or p-values) of 0.05 and below, all five models showed significant results in estimating the water circulation status such as the surface runoff rate and the evapotranspiration rate. The model that was applied all seven urban characteristics factors, can calculate the most similar results such as the existing water circulation model. The water circulation estimation model developed in this study is not only useful to simply estimate the water circulation status of ungauged watersheds but can also provide data for parameter calibration and validation.

본 논문에서는 고리 1호기 해체 비용 추정을 위해 외국 원자력발전소 해체 비용 데이터를 현가화한 후 원자력발전소 해체 비용 추정 회귀 분석모델을 개발하였다. 이 모델 개발에 사용된 데이터는 해체 또는 진행 중인 BWR 13기, PWR 16기의 해체 비용 데이터이다. 회귀 분석모델 도출을 위해, 해체 비용을 종속변수로 정하고, 해체 원전의 운전 특성을 반영할 수 있게 고 안된 Contamination factor와 해체 기간을 독립변수로 선정하였다. 빅데이터 분석 도구인 R language의 통계패키지를 이용 하여 회귀 분석모델을 도출하였다. 이 회귀 분석 모델을 적용하여 고리 1호기 해체 비용을 예측한 결과, 미화 663.40~928.32 백만 달러, 한화 약 7,828.12억~1조 954.18억 원이 소요될 것으로 예측되었다.

In this paper, we studied the applicability of data driven model, AAKR(Auto-Associative Kernel Regression) for generator power loss estimation. Correlation analysis performed on 39 turbine system variables for dataset construction and then 13 variables were selected as highly correlated with generator power output. For a memory vector, 95~100% thermal power section data were used and data at normal power condition for 3 month were extracted for query vector. Analysis result shows that 9 variables show good prediction between measured and estimated data, 2 variables show good correlation but with small bias and 2 variables show increasing difference and low correlation with the passage of time, which assumed to be cause of electric output loss.

The estimation of heat source model is very important for heat transfer analysis with finite element method. Part I of this study used adaptive simulated annealing which is one of the global optimization algorithm for anticipating the parameters of the Goldak model. Although the analysis with 3D model which depicted the real situation produced the correct answer, that took too much time with moving heat source model based on Fortran and Abaqus. This research suggests the procedure which can reduce time with maintaining quality of analysis. The lead time with 2D model is reduced by 90% comparing that of 3D model, the temperature distribution is similar to each other. That is based on the saturation of heat transfer among the direction of heat source movement. Adaptive simulated annealing with 2D model can be used to estimate more proper heat source model and which could enhance to reduce the resources and time for experiments.

Anticipation of welding deformation with finite element method is a very interested topic in the industries, adequate heat source model is essential for concluding reasonable results. This study is related to estimate the parameters of Goldak heat source model, and global optimization algorithm is applied to this research. The heat affected zone (HAZ) boundary line of bead on plate (BOP) welding is used as the target, parameters of heat sources are used as the variables. Adaptive simulated annealing is applied and the optimal result is obtained out of 1,000 candidates. The convergence of finite element method and the global optimization is meaningful for estimation of welding deformation, which could enhance to reduce the resources and time for experiments.

The stochastic point-source model has been widely used in generating artificial ground motions, which can be used to develop a ground motion prediction equation and to evaluate the seismic risk of structures. This model mainly consists of three different functions representing source, path, and site effects. The path effect is used to emulate decay in ground motion in accordance with distance from the source. In the stochastic point-source model, the path attenuation effect is taken into account by using the geometrical attenuation effect and the inelastic attenuation effect. The aim of this study is to develop accurate equations of ground motion attenuation in the Korean peninsula. In this study, attenuation was estimated and validated by using a stochastic point source model and observed ground motion recordings for the Korean peninsula.

Reliability analysis of the components frequently starts with the data that manufacturer provides. If enough failure data are collected from the field operations, the reliability should be recomputed and updated on the basis of the field failure data. However, when the failure time record for a component contains only a few observations, all statistical methodologies are limited. In this case, where the failure records for multiple number of identical components are available, a valid alternative is combining all the data from each component into one data set with enough sample size and utilizing the useful information in the censored data. The ROK Navy has been operating multiple Patrol Killer Guided missiles (PKGs) for several years. The Korea Multi-Function Control Console (KMFCC) is one of key components in PKG combat system. The maintenance record for the KMFCC contains less than ten failure observations and a censored datum. This paper proposes a Bayesian approach with a Dirichlet mixture model to estimate failure time density for KMFCC. Trends test for each component record indicated that null hypothesis, that failure occurrence is renewal process, is not rejected. Since the KMFCCs have been functioning under different operating environment, the failure time distribution may be a composition of a number of unknown distributions, i.e. a mixture distribution, rather than a single distribution. The Dirichlet mixture model was coded as probabilistic programming in Python using PyMC3. Then Markov Chain Monte Carlo (MCMC) sampling technique employed in PyMC3 probabilistically estimated the parameters’ posterior distribution through the Dirichlet mixture model. The simulation results revealed that the mixture models provide superior fits to the combined data set over single models.

본 연구는 한반도에서 Proto3 모형을 수행하기 위하여 필요한 3가지 기후인자 지도, 1. 식물 내한성 구역, 2. 쾨펜- 가이거 기후구, 3. 연누적강수량 영역에 대한 한반도 지도를 1 km 해상도로 제작하였다. Worldclim V2와 한반도 30 년 평균 기상대 데이터를 이용하여 한반도 최저극값온도 격자를 제작하였으며, 이를 활용하여 한반도 식물 내한성 구역 격자지도를 제작하여 Proto3에 이용하였다. 쾨펜-가 이거 기후구 지도는 Beck et al. (2018)의 쾨펜-가이거 기후구 지도를 Proto3에 적합하게 변형하였으며, 연누적강수량 영역은 Worldclim V2를 활용하여 제작하였다. 제작된 Proto3 기후 인자 지도를 활용하여 외래잡초 별나팔꽃의 정착 확산 가능 지역을 예측한 결과, 남한 대부분의 지역과 북한의 황해도를 포함한 남쪽 지방까지 전체 한반도 격자의 50.7%가 정착, 확산 가능지역으로 예측되었다. 본 결과는 별나팔꽃의 위해성 평가 및 관리 체계 수립 등에 활용가능할 것으로 판단된다.

채소작물과 과수작물의 생육에 악영향을 미치는 서리발생을 미리 예측하기 위해 모형을 구축하고 채소 주산지에 적용해 보았다. 서리 발생 전날에 관측되는 다양한 기상인자들(최저기온, 18시 기온, 21시 기온, 24시 기온, 평균풍속, 18시 풍속, 21시 풍속, 구름량, 5일간 강수량, 3일간 강수량, 상대습도, 이슬점온도, 초상최저기온, 지면온도)을 수집 하고, 그 중에서 서리발생에 유의한 영향이 있다고 판단되는 변수들을 통계적 방법(T-test, Random Forest, Multicollinearity test, Akaike Informaiton Criteria, 그리고 Wilk’s lambda values)을 통해 선택하였다. 여러 통계적 방법을 통해 선택된 유의한 기상 인자는 24시 기온, 구름량, 이슬점온도, 21시 풍속 이였으며, 이 기상인자를 기계학습법의 한 종류인 랜덤 포레스트에 적용하여 서리 발생 예측 모형을 구축하였다. 이렇게 구축 된 서리 발생예측 모형의 정확도는 70.6%로 나타났으며, 이 모형을 가을배추와 가을무의 주산지인 홍성과 서산에 적용하였을 때 65.2%와 78.6%로 나타났다.

PURPOSES : This study aimed to estimate road pavement life expectancy using Bayesian Markov Mixture Hazard Model, to support infrastructure asset management. In addition, the life expectancies for the pavement condition index were compared among regional construction and management administrations. METHODS : Eleven years of National Highway road pavement monitoring data fused with ESAL (Equivalent Single Axle Loads), SNP (Structural Number of Pavement, an indicator of structural capacity), and average low temperature, total rainfall, and de-icing were used for the deterioration modeling. Deterioration modeling was performed through the Bayesian Markov Mixture Hazard Model. RESULTS : The expected life expectancy of the crack was estimated at 12.28 to 18.51 years, rut depth was estimated at 15.93 to 25.3 years, and the International Roughness Index was estimated at 10.44 to 14.33 years. It was also confirmed that the heterogeneity factor proposed in the Bayesian Markov Mixture Hazard Model could be used to analyze group characteristics and differences in the benchmark. CONCLUSIONS: This study provided important information in that it compared the life expectancies and structural characteristics of the pavement condition indexes among regional construction and management administrations. Based on this result, it is expected that a pavement structure design and maintenance strategy suitable for deterioration characteristics among regional construction and management administrations will be established.