• Computers and Concrete
• /
• 제12권6호
• /
• pp.785-802
• /
• 2013

Ready-mixed soil material, known as a kind of controlled low-strength material, is a new way of soil cement combination. It can be used as backfill materials. In this paper, artificial neural network and nonlinear regression approach were applied to predict the compressive strength of ready-mixed soil material containing Portland cement, slag, sand, and soil in mixture. The data used for analyzing were obtained from our testing program. In the experiment, we carried out a mix design with three proportions of sand to soil (e.g., 6:4, 5:5, and 4:6). In addition, blast furnace slag partially replaced cement to improve workability, whereas the water-to-binder ratio was fixed. Testing was conducted on samples to estimate its engineering properties as per ASTM such as flowability, strength, and pulse velocity. Based on testing data, the empirical pulse velocity-strength correlation was established by regression method. Next, three topologies of neural network were developed to predict the strength, namely ANN-I, ANN-II, and ANN-III. The first two models are back-propagation feed-forward networks, and the other one is radial basis neural network. The results show that the compressive strength of ready-mixed soil material can be well-predicted from neural networks. Among all currently proposed neural network models, the ANN-I gives the best prediction because it is closest to the actual strength. Moreover, considering combination of pulse velocity and other factors, viz. curing time, and material contents in mixture, the proposed neural networks offer better evaluation than interpolated from pulse velocity only.

• 응용통계연구
• /
• 제27권5호
• /
• pp.743-753
• /
• 2014

Kim 등 (2006)과 Kim 등 (2009)은 2002년에 (사)볼런티어 21에서 조사한 설문자료에 기초하여 우리나라 개인의 기부횟수에 영향을 주는 유의적 설명변수를 보고한 바 있다. 본고에서는 Kim 등 (2006)과 Kim 등 (2009)의 계산오류를 발견하여 이를 수정하고, 아울러 Kim 등 (2009)이 적용한 0이 팽창된 포아송 모형에 로지스틱 회귀모형을 추가하였다. 동 로지스틱 모형으로 기부행위(0, 1)에 영향을 주는 설명변수를 식별하고, 아울러 기부횟수가 작은 군(群)과 큰 군(群)을 판별하여 주는 설명변수를 식별하고자 한다.

• Applied Biological Chemistry
• /
• 제40권3호
• /
• pp.225-231
• /
• 1997

임의로 선정되어진 국산두부콩과 수입두부콩들의 물성 및 색도에 대한 최적화이론을 적용하였다. 제조한 두부의 파손강도와 응력완화현상은 rheometer를 사용하여 측정하였으며, 색도는 colorimeter를 이용하였다. 각 두부콩들의 효과는 수식화된 non-linear canonical regression mpdel로 표현하였으며 각 콩들의 기여도는 trace plot으로 나타내었다. 국산콩은 그 첨가량이 증가할수록 두부의 파손강도는 다른 수입콩보다 우월하게 나타나 두부조직의 강도에는 뛰어남을 보여주었으나 두부조직의 점탄성에 있어서는 약간 뒤떨어짐을 보여주었으며 색감에서는 크게 뒤지는 것으로 나타나서 선정된 물성반응에 대하여 선별적인 경쟁력을 가짐을 보여주었다. 그러나 국산콩의 흡습능력은 수율에 결과적으로 영향을 주며 이를 새로운 반응치로 최적화이론에 적용하였고 그 결과 가격경쟁과 반응에 대한 취약점을 보완하여 최적화된 결과에 기여함을 보여주었다. 이와같은 국산두부콩과 수입두부콩의 최적화된 배합비율은 수율과 가격이라는 두 양립되는 목적함수로써 발산함을 보여 주었으며 이를 수렴시키기 위하여서는 가증치도입과 국산콩의 흡습능력조절에 대한 가공공정의 새로운 연구가 필요하며 이에 대한 조절이 품질향상에 기여할 수 있음을 보여 주었다.

• 콘크리트학회논문집
• /
• 제22권3호
• /
• pp.335-343
• /
• 2010

콘크리트 내부의 철근부식은 구조물의 안전성에 큰 영향을 주므로, 목표 내구수명동안 구 조물의 성능을 확보하려는 연구가 활발하게 진행되고 있다. 이 연구는 대도시나 지하구조물에서 중요하게 평가되는 탄산화에 대하여, 유전자 알고리즘을 적용한 콘크리트 배합기법에 대한 연구이다. 이를 위해, 배합인자에 따른 이산화탄소 확산계수를 문헌조사를 통하여 분석하였으며, 습도를 고려한 최적 함수식을 회귀분석을 통하여 도출하였다. 최적 함수식은 12개의 실험자료에 대하여, 물-시멘트비, 단위 시멘트량, 잔골재율, 단위 굵은골재량, 그리고 상대습도를 포함하도록 고려하였으며, 유전자 알고리즘을 통하여, 주어진 이산화탄소 확산계수에 대한 콘크리트 배합을 도출하였다. 3개의 배합에 대하여 검증한 결과, 10% 미만의 상대오차를 보이며 주어진 배합을 잘 추정하였다. 최종적으로 서로 다른 환경과 설계 제원을 가지는 콘크리트 구조물을 가정하여, 목표 확산계수와 단위 시멘트량을 계산하였으며, 이를 이용하여 배합을 추정하였다. 제안된 기법은 주어진 확산계수와 배합을 잘 추정하였으며, 다양한 배합인자 및 혼화재료가 고려된 실험 자료를 이용한다면 더욱 합리적인 배합 기법으로 발전할 것이다.

• Bulletin of the Korean Chemical Society
• /
• 제29권8호
• /
• pp.1549-1553
• /
• 2008

A molecular imprinted polymer (MIP) using (+)catechin ((+)C) as a template and acrylamide (AM) as a functional monomer was prepared. Acetonitrile was used as the porogen with ethylene glycol dimethacrylate (EGDMA) as the crosslinker and 2,2'-azobis(isobutyronitrile) (AIBN) as the initiator. The adsorption isotherms in the MIP were measured and the parameters of the equilibrium isotherms were estimated by linear and nonlinear regression analyses. The linear equation for original concentration and adsorpted concentrations was then expressed, and the adsorption equilibrium data were correlated into Langmuir, Freundlich, quadratic, and Langmuir Extension isotherm models. The mixture compounds of (+)C and epicatechin (EC) show competitive adsorption on specific binding sites of the (+)catechin-MIP. The adsorption concentrations of (+)C, epicatechin (EC), epicatechin gallate (ECG), and epigallocatechin gallate (EGCG), on the (+)catechin-molecular imprinted polymer were compared. Through the analysis, the (+)catechin-molecular imprinted polymer showed higher adsorption ability than blank polymer which was synthesized molecular imprinted polymer without (+)catechin. Furthermore, the competitive Langmuir isotherms were applied to the mixture compounds of (+)C and EC.

• 동아시아식생활학회지
• /
• 제22권6호
• /
• pp.782-794
• /
• 2012

The objective of this study was to determine the optimal conditions for preparation of cooked brown rice by blending brown rice, white rice and glutinous rice to improve the palatability. Formulations composed of brown rice (10~100%), white rice (0~90%) and glutinous rice (0~90%) were generated from an extreme-vertices of mixture experimental design, which showed ten experimental points for brown rice, with white rice and glutinous rice as the independent variables. The sensory evaluation, color, and texture profile analysis (TPA) of cooked brown rice and pasting characteristics of blending cereals flour were measured as response variables. Regression analysis showed that all responsible variables fit linear, quadratic or special cubic models (p<0.1), except for the cohesiveness of TPA. The goals of optimization of the blending ratio of brown rice, white rice and glutinous rice were given as appearance, flavor, texture and overall acceptability (lower: 5.50, target: 6.62). The optimal conditions were determined to be 34.55% brown rice, 42.71% white rice and 22.74% glutinous rice.

• Industrial Engineering and Management Systems
• /
• 제15권1호
• /
• pp.110-121
• /
• 2016

With the growing interest in skincare and maintenance, there are increasing numbers of studies on the classification of skin type and the factors influencing each type. This study presents a novel methodology by using data mining, for the determination of the relationships between skin type, lifestyle, and patterns of cosmetic utilization. Eight skin-specific factors, which are moisture, sebum in U-zone (both cheeks), sebum in T-zone (forehead, nose, and chin), pore, melanin, wrinkle, acne, hemoglobin, were measured in 1,246 subjects living in South Korea, in conjunction with a questionnaire survey analyzing their lifestyles and pattern of cosmetic utilization. Using various multivariate statistical methods and data mining techniques, we classified the skin types based on the skin-specific values, determined the relationship between skin type and lifestyle, and accordingly sorted the subjects into clusters. Logistic regression analysis revealed gender-related differences in the skin; therefore, separate analyses were performed for males and females. Using the Gaussian Mixture Modeling (GMM) technique, we classified the subjects based on skin type (two male and four female). Using the ANOVA and decision tree techniques, we attempted to characterize the relationship between each skin type and the lifestyles of the subjects. Menstruation, eating habits, stress, and smoking were identified as the major factors affecting the skin.

• Smart Structures and Systems
• /
• 제24권6호
• /
• pp.733-744
• /
• 2019

Wind speed forecasting is critical for a variety of engineering tasks, such as wind energy harvesting, scheduling of a wind power system, and dynamic control of structures (e.g., wind turbine, bridge, and building). Wind speed, which has characteristics of random, nonlinear and uncertainty, is difficult to forecast. Nowadays, machine learning approaches (generalized regression neural network (GRNN), back propagation neural network (BPNN), and extreme learning machine (ELM)) are widely used for wind speed forecasting. In this study, two schemes are proposed to improve the forecasting performance of machine learning approaches. One is that optimization algorithms, i.e., cross validation (CV), genetic algorithm (GA), and particle swarm optimization (PSO), are used to automatically find the optimal model parameters. The other is that the combination of different machine learning methods is proposed by finite mixture (FM) method. Specifically, CV-GRNN, GA-BPNN, PSO-ELM belong to optimization algorithm-assisted machine learning approaches, and FM is a hybrid machine learning approach consisting of GRNN, BPNN, and ELM. The effectiveness of these machine learning methods in wind speed forecasting are fully investigated by one-year field monitoring data, and their performance is comprehensively compared.

• Computers and Concrete
• /
• 제32권3호
• /
• pp.233-246
• /
• 2023

Accurate prediction of concrete compressive strength can minimize the need for extensive, time-consuming, and costly mixture optimization testing and analysis. This study attempts to enhance the prediction accuracy of compressive strength using stacking ensemble machine learning (ML) with feature engineering techniques. Seven alternative ML models of increasing complexity were implemented and compared, including linear regression, SVM, decision tree, multiple layer perceptron, random forest, Xgboost and Adaboost. To further improve the prediction accuracy, a ML pipeline was proposed in which the feature engineering technique was implemented, and a two-layer stacked model was developed. The k-fold cross-validation approach was employed to optimize model parameters and train the stacked model. The stacked model showed superior performance in predicting concrete compressive strength with a correlation of determination (R²) of 0.985. Feature (i.e., variable) importance was determined to demonstrate how useful the synthetic features are in prediction and provide better interpretability of the data and the model. The methodology in this study promotes a more thorough assessment of alternative ML algorithms and rather than focusing on any single ML model type for concrete compressive strength prediction.

• Nuclear Engineering and Technology
• /
• 제54권11호
• /
• pp.4159-4169
• /
• 2022

Hydrogen mitigation using Passive Autocatalytic Recombiners (PARs) has been widely accepted methodology inside reactor containment of accident struck Nuclear Power Plants. They reduce hydrogen concentration inside reactor containment by recombining it with oxygen from containment air on catalyst surfaces at ambient temperatures. Exothermic heat of reaction drives the product steam upwards, establishing natural convection around PAR, thus invoking homogenisation inside containment. CFD models resolving individual catalyst plate channels of PAR provide good insight about temperature and hydrogen recombination. But very thin catalyst plates compared to large dimensions of the enclosures involved result in intensive calculations. Hence, empirical correlations specific to PARs being modelled are often used in integral containment studies. In this work, an experimentally validated CFD model of PAR has been employed for developing an empirical correlation for Indian PAR. For this purpose, detailed parametric study involving different gas mixture variables at PAR inlet has been performed. For each case, respective values of gas mixture variables at recombiner outlet have been tabulated. The obtained data matrix has then been processed using regression analysis to obtain a set of correlations between inlet and outlet variables. The empirical correlation thus developed, can be easily plugged into commercially available CFD software.