기후변화와 식품공급망의 복잡성 증대로 식품 위해요소 의 발생 경로와 패턴이 다변화됨에 따라, 과학적 예측과 선 제적 개입이 가능한 예방형 식품안전 관리체계의 필요성이 대두되고 있다. 본 연구는 기후·환경 요인이 식품 위해요소 에 미치는 영향을 분석함으로써, 기후 민감성이 높은 위해 요소를 식별하고 예측 가능성과 주요 환경인자를 도출하였 다. 아울러 국내외 데이터 기반 위해예측 시스템의 운영 사 례를 비교·분석함으로써, 식품위해예측센터의 실질적 운영 과 역할을 위한 발전방향을 제시하였다. 본 연구를 통해 향 후 식품위해예측센터가 식품안전 정책의 과학화와 지능화 를 이끄는 전략적 플랫폼으로 기능하고, 예방 중심의 관리 체계로의 전환을 유도할 수 있도록 실효적 토대와 정책적 방향성을 제공하고자 한다.

As conventional road traffic noise prediction models are designed to estimate long-term representative noise levels, capturing fine-scale noise fluctuations caused by real-world traffic dynamics is challenging. A previous study proposed a microscopic road traffic noise model (MTN) can calculate time-series noise levels with a resolution of 1 s using the concept of a moving noise source. In this study, two experiments were conducted to verify the accuracy of the noise prediction of the model. First, by comparing the calculated noise levels of two conventional road traffic noise models and the MTN in a simple road simulation environment, it was confirmed that the calculation error was within 3 dB(A) when calculating the 1-h equivalent noise level. Second, an experiment was conducted to verify the noise prediction error of the MTN on six actual roads. A comparison of the calculated noise level using the MTN based on traffic data collected from actual roads with the measured noise level on real roads showed that the calculated noise level achieved a mean absolute error (MAE) of 1.88 dB(A) from the equivalent noise level and 1.28 dB(A) from the maximum noise level. This was similar to the MAE of the foreign road traffic noise models. However, when the location of the receiver is within 10 m of the road, an error of more than 3 dB(A) occurs because of the simplicity of the MTN propagation model, which remains a problem that must be solved in the future. This study proved that the noise level calculation using the MTN is similar to the noise of an actual road environment. Additionally, the continuous development of the MTN is expected to make it an effective alternative for the management of road noise.

본 연구는 미국흰불나방(Hyphantria cunea)의 국내 잠재 서식지 변화를 분석하기 위해 기후 변화와 토지 피복 변화라는 주요 환경 요인을 종합 적으로 고려하여 서식 적합도를 예측하였다. 먼저, 전 지구적 출현 데이터를 바탕으로 MaxEnt 모델을 구축하여 기후 변화 시나리오에 따른 국내 서식 적합도 변화를 모의하였다. 이후, 토지 피복 변화에 따른 산림 및 시가지 내 미국흰불나방의 PRA를 분석하였다. 연구 결과, 미국흰불나방의 적합 서식지는 태백산맥과 한라산 고산지대를 제외한 한국 대부분 지역에 분포할 것으로 예측되었다. SSP에 기반한 통합 기후-토지 피복 시나리오 에서 미국흰불나방의 PRA는 SSP1-2.6 시나리오에서는 2055s, 2085s가 각각 66,934 km2에서 67,363 km2로 증가한 반면, SSP3-7.0에서는 PRA는 66,676 km2에서 59,696 km2로 크게 감소하는 결과가 나타났다. 그러나 모든 시나리오에서 미국흰불나방 PRA 백분율이 여전히 전 국토 면적의 80%를 초과하기 때문에, 미국흰불나방에 대한 지속적인 방제 및 관리가 필요함을 시사한다. 본 연구는 국내에 광범위하게 퍼진 미국흰불나 방 개체군의 지속적인 관리가 필요함을 강조하며, 이를 토대로 미국흰불나방의 모니터링, 조기 경보, 예방 및 통제, 관리를 위한 기초 자료를 제공한 다. 또한, 기후 변화와 토지 이용 변화가 미국흰불나방의 서식 적합도에 미치는 영향을 종합적으로 분석함으로써, 효과적인 방제 및 관리 전략 수립 에 기여할 것으로 기대된다.

The purpose of this study, when predicting acute oral toxicity using QSAR software, the reliability of the predicted values was studied according to a single functional group or multiple functional groups within a single chemical. Acute oral toxicity is predicted using EPA T.E.S.T S/W for chemicals registered in ChemIDplus. The effect of a combination of specific functional periods on the degree of consistency of predicted values was studied. When some specific functional groups (combinations) exist, it was confirmed that the experimental and predicted values were high and low. It was confirmed that the prediction accuracy was high when the Anion group and the Halogen group were together, and the perdiction accuracy was significantly low when the Nitrile group was present. As a result of accumulating such data and showing reliability in predicting acute oral toxicity with EPA T.E.S.T S/W for 10 SVHC substances without experimental values, the matching rate was derived from at least 0% to 73.33%. It was confirmed that there was some tendency of the QSAR prediction value according to the combination of specific complex functional groups. When 10 SVHC substances without experimental data were predicted to be toxic through T.E.S.T S/W by quantitatively databaseizing the above tendency, 0~73.33% of the results were derived as a result of showing the realiability of the program prediction

대만은 글로벌 반도체 공급망에서 중요한 역할을 하고 있으며, 관련 제품 수출도 원활하고 경제 상황도 비교적 좋다. 그러나 이러한 경제적 이점을 대만의 청년들은 충분히 누리지 못하고 있으며, 우리나라 청년 들과 마찬가지로 일과 삶에서 여러 가지 어려움에 직면해 있다. 특히 기술 불일치로 인하여 취업에 있어 어려움을 겪는 경우가 많다. 본 연구는 이러한 상황을 반영하여 대만 청년층을 중심으로 임금 근로자들의 기술 불일치 영향 요인을 판별분석을 이용하여 분석하였다. 분석자료는 대만 중앙연구원(Academia Sinica)의 ‘Taiwan Social Change Survey’ 7차 자료의 ‘Work Orientation’ 2차 데이터셋을 이용하였다. 판별분석결과, 기술불일치에는 임금근로자로 일하는 대만 청년들의 교육불일치, 우울, 행복, 신기술 수용 성 및 임금과 학력이 유의미한 영향을 미치는 것으로 나타났으며 성별, 결혼여부, 학력, 노종조합 참여 여부는 통계적으로 유의한 효과를 보이는 것으로 나타나지 않았다. 변수의 판별에 미치는 효과의 크기는 교육불일치, 임금, 학력, 신기술 수용성의 순서로 상대적인 중요도를 나타났다. 또한, 기술일치집단의 72.3%, 기술불일치집단의 64.6%가 정확히 분류된 것으로 나타났으며, 판별적중률은 72.9%로 분석되었다. 이러한 연구 결과를 바탕으로 본 연구의 시사점 및 한계점, 그리고 향후 연구방향이 제시되었다.

This study proposes a weighted ensemble deep learning framework for accurately predicting the State of Health (SOH) of lithium-ion batteries. Three distinct model architectures—CNN-LSTM, Transformer-LSTM, and CEEMDAN-BiGRU—are combined using a normalized inverse RMSE-based weighting scheme to enhance predictive performance. Unlike conventional approaches using fixed hyperparameter settings, this study employs Bayesian Optimization via Optuna to automatically tune key hyperparameters such as time steps (range: 10-35) and hidden units (range: 32-128). To ensure robustness and reproducibility, ten independent runs were conducted with different random seeds. Experimental evaluations were performed using the NASA Ames B0047 cell discharge dataset. The ensemble model achieved an average RMSE of 0.01381 with a standard deviation of ±0.00190, outperforming the best single model (CEEMDAN-BiGRU, average RMSE: 0.01487) in both accuracy and stability. Additionally, the ensemble's average inference time of 3.83 seconds demonstrates its practical feasibility for real-time Battery Management System (BMS) integration. The proposed framework effectively leverages complementary model characteristics and automated optimization strategies to provide accurate and stable SOH predictions for lithium-ion batteries.

Republic of Korea is building a multi-layered missile defense system against North Korea’s growing ballistic missile threat. To maximize the intercept performance of a multi-layered missile defense system, it is important to develop an efficient engagement plan that considers the interceptable time/space of each interceptor system for ballistic missiles. To do so, it is necessary to predict the flight trajectory of the ballistic missile, which must be done within a short time considering the short battlefield environment and the speed of the ballistic missile. This study presents a model for rapid trajectory prediction of ballistic missiles using the kinetic characteristics of each flight phase(thrust phase, midcourse phase, and re-entry phase) of ballistic missiles, a method for estimating kinetic information from ballistic missile observation data(time and position), and a mathematical analysis of the equations of motion of ballistic missiles.

본 연구는 환경 요인을 바탕으로 절화용 국화 생장 예측을 위한 최적의 모델을 개발하는 것을 목표로 하였다. 이를 위해 13개의 모델(Linear Regression, Lasso Regression, Ridge Regression, ElasticNet Regression, K-Nearest Neighbors (KNN), Support Vector Regression (SVR), Neural Network, Decision Tree, Random Forest, XGBoost, AdaBoost, CatBoost, Stacking)의 성능을 R2, MAE, RMSE를 평가 지표 로 비교하였다. 단일 모델 중에서는 Decision Tree가 가장 우수한 성능을 보였으며, R2값은 0.90에서 0.91 사이였다. 앙 상블 모델 중에서는 CatBoost가 가장 높은 성능을 보였으며 (R2=0.90~0.92) Random Forest와 XGBoost 또한 유사한 성 능을 보였다. 전체적으로 트리 기반 앙상블 모델이 국화 생장 예측에 적합한 모델로 나타났다.

The COVID-19 pandemic has caused significant disruptions in global air travel demand, presenting new challenges for accurately forecasting passenger volumes. This study analyzes the monthly air passenger demand data from 2010 to 2022 to identify key external factors that influence passenger demand. Our analysis shows that the number of international visitors to Singapore is a critical determinant of passenger demand. Consequently, we propose a SARIMAX (Seasonal AutoRegressive Integrated Moving Average with eXogenous variables) model to forecast monthly air passenger demand at Singapore's Changi Airport, integrating international visitor numbers as an exogenous variable. Through comprehensive model identification and parameter estimation, we select the best SARIMAX configuration. To validate the performance of the model, traditional time series methods such as SARIMA, various exponential smoothing methods, and advanced machine learning methods like LSTM (Long Short-Term Memory) and Prophet were compared for forecasting monthly air passenger demand at Changi Airport in 2023. The results show that the SARIMAX model significantly outperforms all other tested models, achieving the best performance across multiple forecasting metrics, including the Mean Absolute Percentage Error.

Existing reinforced concrete buildings with seismically deficient columns experience reduced structural capacity and lateral resistance due to increased axial loads from green remodeling or vertical extensions aimed at reducing CO2 emissions. Traditional performance assessment methods face limitations due to their complexity. This study aims to develop a machine learning-based model for rapidly assessing seismic performance in reinforced concrete buildings using simplified structural details and seismic data. For this purpose, simple structural details, gravity loads, failure modes, and construction years were utilized as input variables for a specific reinforced concrete moment frame building. These inputs were applied to a computational model, and through nonlinear time history analysis under seismic load data with a 2% probability of exceedance in 50 years, the seismic performance evaluation results based on dynamic responses were used as output data. Using the input-output dataset constructed through this process, performance measurements for classifiers developed using various machine learning methodologies were compared, and the best-fit model (Ensemble) was proposed to predict seismic performance.