하수처리장 유출수의 수질 예측은 수질 사고의 사전 대응 및 처리장의 안정적인 운영을 위해 필수적인 요소이다. 최근 머신러닝을 활용한 예측 모델링에서 예측 성능 향상과 과적합 방지를 위해 다양한 교차 검증법과 하이퍼파라미터 최적화 기법이 활용되고 있으나, 하수처리장 데이터는 시간적 의존성과 급격한 변동성이 내재되어 있어 과적합에 취약하고 안정적인 모델 구축에 어려움이 따른다. 본 연구에서는 이러한 데이터 특성을 효과적으로 반영할 수 있는 최적의 모델링 파이프라인을 구축하고자 하였으며, XGBoost 모델을 기반으로 유출수 내 총질소 농도를 예측하였다. 예측 성능 평가 지표로는 평균 제곱근 오차(Root Mean Square Error, RMSE), 결정계수(coefficient of determination, R2), RMSE 오차 개선율(the rate of improvement on RMSE, RIRRMSE) 그리고 계산 시간을 사용하였다. 기본적인 Hold-out 방식의 성능을 기준으로 K-fold, 시계열 교차 검증(Time Series Cross Validation, TSCV), 블록 시계열 교차 검증(Blocked Time Series Cross Validation, BTSCV) 기법의 예측 성능을 분석한 결과, BTSCV는 인접한 데이터만을 고려하는 방식으로 시간적 의존성과 급변 특성을 효과적으로 반영하여 가장 높은 RIR(36.37%)을 기록하였다. 또한, 하이퍼파라미터 최적화(그리드 서치와 베이지안 최적화) 기법과의 다양한 교차 검증법의 조합을 통해 모델 성능과 과적합 방지 효과를 분석한 결과, BTSCV와 베이지안 최적화의 결합은 짧은 계산 시간(364.64초)과 함께 가장 높은 RIR(64.93%)을 보였으며, 훈련 및 평가 데이터 간 성능 차이도 최소화되어 일반화된 예측 모델로서의 효과성이 입증되었다. 따라서 본 연구는 하수처리장 시계열 데이터의 특성에 적합한 BTSCV와 베이지안 최적화 기법을 결합한 모델링 파이프라인 전략을 제안하며, 향후 실시간 수질 모니터링 및 하수처리장 운영 효율성 제고에 기여할 수 있을 것으로 기대된다.

Domestic 105㎜ HE (High Explosive) shell is composed of three parts that are Fuze, Projectile and Propellants. Among three parts, propelling charge of propellants part consists of single base propellants. It has been known that the lifespan of single base propellants is affected by a storage period. These are because Nitrocellulose (NC) which is the main component of propelling gunpowder can be naturally decomposed to unstable substances similar with other nitric acid ester. Even though it cannot be prevented fundamentally from being disassembled, a decomposition product (NO2, NO3, and HNO3) and tranquillizer DPA (Diphenylamine) having high reactivity are added into a propellant to restrain induction of automatic catalysis by a decomposition product. The decay rate of the tranquillizer is also affected by a production rate of the decomposition product of NC. Therefore, an accurate prediction of the Self-Life is required to ensure against risks such as explosion. Hereupon, this paper presents a new methodology to estimate the shelf-life of single base propellants using data of ASRP (Ammunition Stockpile Reliability Program) to domestic 105mm HE (propelling charge of propellants part). We selected four attributes that are inferred to have influence on distribution of the DPA amount in a propellant from the ASRP dataset through data mining processes. Then the selected attributes were used as independent variables in a regression analysis in order to estimate the shelf-life of single base propellants. 1)

Domestic 105㎜ HE (High Explosive) shell is divided into three parts (Fuze, Projectile and Propellants). Among three parts, propelling charge of propellants part consists of single base propellants. The lifespan of single base propellants is affected by a storage period. These are because Nitrocellulose (NC) which is the main component of propelling gunpowder can be naturally decomposed to unstable substances similar with other nitric acid ester. Even though it cannot be prevented fundamentally from being disassembled, a decomposition product (NO2, NO3, and HNO3) and tranquillizer DPA (Diphenylamine) having high reactivity are added into a propellant to restrain induction of automatic catalysis by a decomposition product. The decay rate of the tranquillizer is also affected by a production rate of the decomposition product of NC. Therefore, an accurate prediction of the Self-Life is required so that it is ensure against risks. Hereupon, this paper presents a new methodology to estimate the shelf-life of single base propellants using data of ASRP (Ammunition Stockpile Reliability Program) to domestic 105mm HE (propelling charge of propellants part). We selected four attributes that is inferred to have influence on distribution of the DPA amount in a propellant from the ASRP dataset through data mining processes. Then the selected attributes were used as independent variables in a regression analysis in order to estimate the shelf-life of single base propellants.

본 연구에서는 이미지 감상자가 선호하는 화질의 객관적 평가 항목들의 범위를 구체화하고 실제 이미지를 기반으로 화질의 선호도를 측정하는 방식을 따랐다. 즉, 무기준법(No-Reference)을 기반으로 하고 화질 평가 요소를 다이내믹 레인지, 컬러, 콘트라스트로 규정하였다. 샘플 사진 수집은 인터넷 갤러리에서 추천수 30회 이상을 기준으로 감상자들이 선호하는 풍경사진 200장의 이미지를 선정하였다. 그리고 세 가지 객관적 화질 평가 항목에 의한 정규분포분석을 통하여 총점을 백점 기준으로 환산하여 최종적으로 예상되는 화질의 선호도를 측정하였다. 본 실험에서 적용한 실제 사진 샘플의 다이내믹 레인지 측정값은 10 stop, LAB 평균은 L:54.7, A:2.96, B:-15.84, RSC 콘트라스트는 376.9로 나타났다. 총 200장 샘플 사진의 정규분포 z값은 다이내믹 레인지가 0.21, LAB 평균이 LAB 평균이 L:0.15, A:0.38, B:0.13, RSC 콘트라스트가 0.08을 나타냈다. 표준정규분포 표에서 위의 z값이 나타날 확률을 상위 백분율로 나타내면 실제 사진 샘플의 다이내믹 레인지는 8.32%, LAB 평균은 LAB 평균은 L:5.96%, A:14.8%, B:5.17%, RSC 콘트라스트는 3.19%를 나타낸다. 즉 화질 평가 모형에 사용된 실제 사진 샘플의 상위 백분율을 100점으로 환산한 다이내믹 레인지는 91.68점, LAB 평균은 91.36점, RSC 콘트라스트는 96.81점이다. 따라서 세 가지 객관적 화질 평가 항목에 의한 총점의 평균은 94.99점으로 나타낼 수 있었다. 즉 본 연구에서 제안한 화질 측정 모형을 통하여 실제 사진 샘플에 대한 선호도를 수치적으로 측정할 수 있었다. 이와 같은 연구를 통하여 이미지 감상자가 선호하는 화질의 재현 성능 범위를 구체화하고 예상되는 화질의 선호도를 수치화하는 실용적인 연구 결과를 제안하였다.

Among ammunitions which are stored in a war field, the lifespan of propelling gunpowder is affected by storage environment such as storage temperature, humidity, and exposure to sunlight. These are because Nitrocellulose (NC) which is the main component of propelling gunpowder can be naturally disassembled to unstable substances similar with other nitric acid ester. We can’t prevent it fundamentally from being disassembled, but to restrain induction of automatic disassembly by decomposition product, a decom- position product (NO2, NO3, and HNO3) and tranquillizer DPA (Diphenylamine), having high reactivity, are added into a propellant. For this, it will decrease the velocity of tranquillizer which can also affect the velocity of producing the decomposition product of NC, storage temperature or humidity of propelling gunpowder is higher, drop of tranquillizer content is much faster. Therefore, to extend storage lifespan of propelling gunpowder, it is really important to control storage temperature or humidity inside the magazine. Hereupon, according to the manufacture of small scale model magazine and the result of performing experiments and measuring variation of inside temperature (storage temperature), using roof types that have a steel slate structure of magazine among ground magazines, this research shows the differences in details.

Among ammunitions which are stored in a war field, the lifespan of propelling gunpowder is affected by storage environment such as storage temperature, humidity, and exposure to sunlight. These are because Nitrocellulose (NC) which is the main component of propelling gunpowder can be naturally disassembled to unstable substances similar with other nitric acid ester. We can’t prevent it fundamentally from being disassembled, but to restrain induction of automatic disassembly by decomposition product, a decomposition product (NO2, NO3, and HNO3) and tranquillizer DPA (Diphenylamine), having high reactivity, are added into a propellant. For this, it will decrease the velocity of tranquillizer which can also affect the velocity of producing the decomposition product of NC, storage temperature or humidity of propelling gunpowder is higher, drop of tranquillizer content is much faster. Therefore, to extend storage lifespan of propelling gunpowder, it is really important to control storage temperature or humidity inside the magazine. Hereupon, according to the manufacture of small scale model magazine and the result of performing experiments and measuring variation of inside temperature(storage temperature), using roof types that have a steel slate structure of magazine among ground magazines, this research shows the differences in details.

This paper presents an analytical approach to evaluate a possibility of progressive collapse of reinforced concrete frame structure. In order to simulate adequate behavior of the structure, we considered the bond failure of lap spliced bars due to concrete crack in a plastic hinge region. Analysis result gave that there was additional deflection due to the bond failure of bar in beams at the node where the increasing displacement load was applied. The stress was concentrated to the beam where a column was removed, and as it goes up to higher stories, the moment change of an upper story appeared less.

This paper presents an analytical approach to evaluate a possibility of progressive collapse of reinforced concrete frame structure. In order to simulate adequate behavior of the structure, we considered the bond failure of lap spliced bars due to concrete crack in a plastic hinge region. Bond characteristic was modeled in the analysis by using previous bond-slip relation and the effect of it was evaluated. Analysis result gave that there was additional deflection due to the bond failure of bar in beams at the node where the increasing displacement load was applied.