고성능 콘크리트(HPC) 압축강도는 추가적인 시멘트질 재료의 사용으로 인해 예측하기 어렵고, 개선된 예측 모델의 개발이 필수적 이다. 따라서, 본 연구의 목적은 배깅과 스태킹을 결합한 앙상블 기법을 사용하여 HPC 압축강도 예측 모델을 개발하는 것이다. 이 논 문의 핵심적 기여는 기존 앙상블 기법인 배깅과 스태킹을 통합하여 새로운 앙상블 기법을 제시하고, 단일 기계학습 모델의 문제점을 해결하여 모델 예측 성능을 높이고자 한다. 단일 기계학습법으로 비선형 회귀분석, 서포트 벡터 머신, 인공신경망, 가우시안 프로세스 회귀를 사용하고, 앙상블 기법으로 배깅, 스태킹을 이용하였다. 결과적으로 본 연구에서 제안된 모델이 단일 기계학습 모델, 배깅 및 스태킹 모델보다 높은 정확도를 보였다. 이는 대표적인 4가지 성능 지표 비교를 통해 확인하였고, 제안된 방법의 유효성을 검증하였다.

Impact damage tolerance is an important design requirement for composite structures. In this study, the effect of post impact damage and hole size of the composite sandwich skin / sandwich with core specimen on compressive strength of the laminate was analyzed. Three specimen tests were performed in this research. Two tests were carried out on pure bending test specimens subjected to impact damage to the skin and specimen with a hole in one of its skin as a damage. Through this test, we compared the reduction of compressive strength due to the size of skin damage and the size of the hole. Also, core-free specimen with an open hole under uniaxial loading were tested to produce reference data for comparison with the series tested earlier. As results of the tests, the sandwich beams with damage size and open hole are almost identical, and we concluded that the prediction of compressive strength reduction after impact of the sandwich skin structure can be predicted using an analytical model assuming skin open hole as impact inputs.

Concrete has recently been modified to have various performance and properties. However, the conventional method for predicting the compressive strength of concrete has been suggested by considering only a few influential factors. so, In this study, nine influential factors (W/B ratio, Water, Cement, Aggregate(Coarse, Fine), Fly ash, Blast furnace slag, Curing temperature, and humidity) of papers opened for 10 years were collected at 4 conferences in order to know the various correlations among data and the tendency of data. The selected mixture and compressive strength data were used for learning the Deep Learning Algorithm to derive a prediction model. The purpose of this study is to suggest a method of constructing a prediction model that predicts the compression strength with high accuracy based on Deep Learning Algorithms.

Prediction of compressive strength of concrete by Maturity Method is applied in construction site. However, due to the use of wired type high-priced equipment, economic efficiency and workability are falling. In this study, a newly developed concrete embedded wireless sensor is used to perform a mock-up test. Next, the concrete compressive strength of the Maturity Method is predicted using Saul and Plowman's function as measured temperature data. The predicted concrete strength at the beginning of the age was the actual strength and stiffness, but the error rate was less than 1% at 28th day.

This study was evaluated compressive strength of age 28 days of binary blended concrete according to there type of superplasticizer and there type of w/c. In addition, we are proposed modification prediction model equation that can reflect efficiency of water reducing and influence of binders using Lyse equation to predict the compression strength through the conventional W/C.

콘크리트구조물의 진단에 사용되는 비파괴실험법들은 구조물에 손상을 입히지 않고 구조물의 결함이나 강도를 추정할 수 있다는 장점이 있지만 추정값에 대한 신뢰성이 떨어진다는 문제점이 있다. 본 연구에서는 이러한 문제점을 해결하기 위해 2가지 배합으로 총 180개의 공시체를 제작하였고, P파와 S파에 의한 초음파속도 측정, 종진동과 변형진동에 의한 충격공진법 총 4가지의 비파괴실험을 실시하였다. 그리고 실제압축강도 측정을 통해 비파괴실험 결과값의 신뢰성을 비교 분석하였다. 각 비파괴실험법의 결과값에 대한 통계적 분석결과 변동계수값이 가장 낮은 실험법은 S파에 의한 초음파속도법으로 가장 안정적인 관측이 가능한 것으로 나타났다. 한편, 실제압축강도와의 관계를 통해 압축강도 4개의 압축강도 추정식을 제안하였으며 S파에 의한 초음파속도법의 결정계수값이 가장 높은 것으로 나타났다. 향후 다양한 배합조건에 따른 비파괴실험 신뢰성에 대한 보완 연구가 필요할 것으로 판단된다.

Concrete with blast furnace slag (BFS) shows varied strength development properties different from normal concrete. Therefore, a precise prediction of compressive strength using a full maturity model is desired. The purpose of this study is to predict the compressive strength of concrete with BFS by calculating the apparent activation energy (Ea) and rate constant (kT) for each BFS replacement ratio. The method of Carino Model is used in this study for predicting compressive strength of concrete with BFS.

Concrete with blast furnace slag (BFS) shows varied strength development properties different from normal concrete. Therefore, a precise prediction of compressive strength using a full maturity model is desired. The purpose of this study is to predict the compressive strength of concrete with BFS by calculating the apparent activation energy (Ea) and rate constant (kT) for each BFS replacement ratio. The method of Carino Model is used in this study for predicting compressive strength of concrete with BFS.

The purpose of this study is inspection of concrete using impact resonance method and UPV(Ultrasonic Pulse Velocity) method. The frequency polygon of non-destructive testing result shows that non-destructive testing is closely related to compressive strength of concrete.

The purpose of this study is to more accurately predict the compressive strength of concrete according to mixing proportion using UPV(Ultrasonic Pulse Velocity) method. The equation to predict the compressive strength of concrete is proposed based on the results of UPV method.

In this study, rational prediction models for the effective compressive strengths of HSC corner and interior columns with intervening NSC slabs are developed. A structural analogy between HSC column-NSC slab joint and brick masonry is used to develop the prediction models. In addition, the aspect ratio of slab thickness to column dimension and the surrounding slab confinement effect are considered in the models. The proposed prediction model is verified by comparison with experimental results and various prediction expressions. As a result, with average test-to-predicted ratios of 1.00 for HSC corner columns and 1.09 for interior columns, the proposed equation provides superior predictions over all of the existing effective strength prediction approaches including KCI structural concrete design code(2012).

This study investigated the compressive strength of concrete by using ultrasonic pulse velocity method, and proposed the equation able to predict the compressive strength of concrete.

본 연구에서는 화산재를 건설재료로 활용하기 위하여 백두산, 한라산의 화산재 및 다공성의 제올라이트에 시멘트 및 메타카올린을 첨가한 공시체에 대한 재령 0일, 7일, 28일 배합비별 압축강도 특성 데이터를 바탕으로 인공신경망 모델에 적용하여 학습, 예측함으로써 강도예측을 위한 인공신경망의 적용 가능성을 평가하였다. 인공신경망 모델에는 역전파 학습알고리즘(back-propagation learning algorithm)이 적용되었으며, 다양한 입력변수를 달리한 최적의 인공신경망 조건에서 학습을 시행하였다. 또한, 다양한 배합조건이 일축압축강도에 미치는 영향에 대한 민감도 분석을 실시하였다. 이러한 연구를 통해 얻어진 결과물은 화산재를 활용한 블록의 일축압축강도 특성을 파악하는데 좋은 툴이 될 것으로 기대된다.

Temperature of fresh concrete can be effectively used to predict the strength of concrete being cured and make an informed decision for stripping the molds. A hygrothermograph and thermo-couple sensors that require an extensive wiring have been applied to measure a temperature of concrete at the early stage of the curing process on site. Therefor, this study on the strength prediction using Maturity is mainly focused on, but the study on the concrete mixing blast furnace slag powder is insufficient. The purpose of this study is to investigate the relationships between compressive strength and equivalent age by Maturity function and is to compare and examine the strength prediction of concrete mixing Blast Furnace Slag Power using ACI and Logistic Curve prediction equation.