• Computers and Concrete
• /
• 제26권2호
• /
• pp.175-183
• /
• 2020

Fly ash has become an important component of concrete as supplementary cementitious material with the development of concrete technology. To make use of fly ash efficiently, four types of fly ash with particle size distributions that are in conformity with four functions, namely, S.Tsivilis, Andersen, Normal and F distribution, respectively, were prepared. The four particle size distributions as functions of the strength and pore structure of concrete were thereafter constructed and investigated. The results showed that the compressive and flexural strength of concrete with the fly ash that conforming to S.Tsivilis, Normal, F distribution increased by 5-10 MPa and 1-2 MPa, respectively, compared to the reference sample at 28 d. The pore structure of the concrete was improved, in which the total porosity of concrete decreased by 2-5% at 28 d. With regarding to the fly ash with Andersen distribution, it was however not conducive to the strength development of concrete. Regression model based on the grey multiple linear regression theory was proved to be efficient to predict the strength of concrete, according to the characteristic parameters of particle size and pore structure of the fly ash.

• 한국분무공학회지
• /
• 제20권4호
• /
• pp.247-253
• /
• 2015

Recently, the engine calibration technique has been much more complicated than that of the past engine case in order to satisfy the strict emission regulations. The current calibration method for the diesel engine which has an increasing market is both costly and time-consuming. New engine calibration method is required to develop for high-quality diesel engines with low cost and release it at the appropriate time. This study provides the optimal calibrating technique for complex engine systems using statistical modeling and numerical optimization. Firstly, it design a test plan based on Design of Experiments, a V-optimality methodology which is suitable looking for set-points, and determine the shape of test engine response. Secondly, it uses functions to make linear regression model for data analysis and optimization to fit the models of engines behavior. Finally, it generates the optimal calibrations obtained directly from empirical engine models using Grey Relational Analysis and compares the calibrations with data. This method can develop a process for systematically identifying the optimal balance of engine emissions.

• International Journal of Industrial Entomology and Biomaterials
• /
• 제22권1호
• /
• pp.11-15
• /
• 2011

Impact of pruning date, shoot age and weather parameters on the severity and development of grey leaf spot (Pseudocercospora mori) of mulberry was studied. The disease severity (%) increased with increase in shoot age irrespective of pruning date. Maximum disease severity was observed in plants pruned during second week of October and minimum in plants pruned during last week of December. Significant (P<0.05) influence of date of pruning, shoot age and their interaction was observed on the severity of the disease. Apparent infection rate (r) was significantly higher during plant growth period from day-48 to day-55. Average apparent rate was higher in plants pruned during first week of September and least in plants pruned during third and fourth week of December. Multiple regression analysis revealed contribution of various combinations of weather parameters on the disease severity. A linear prediction model [$Y=66.05+(-1.39)x_1+(-0.219)x_4$] with significant $R^2$ was developed for prediction of the disease under natural epiphytotic condition.

• 농업과학연구
• /
• 제7권2호
• /
• pp.131-144
• /
• 1980

수자원개발(水資源開發)의 계획(計劃) 및 설계(設計)를 위한 하천(河川)의 월유출량(月流出量) 추정방법(推定方法)인 전월단위(全月單位)의 사변수(四變數) 선형회귀모형(線型回歸模型)을 일반화(一般化)하여 유출량(流出量) 기록(記錄)의 Extension, 무계기(無計器) 하천(河川)의 유출량(流出量) 추정(推定) 등(等)에 이용(利用)할 수 있도록 하였으며 금강(錦江) 수계(水系)에 적용(適用)한 결과(結果)는 다음과 같다. 1. 선형회귀모형(線型回歸模型)은 물수지방정식(收支方程式)을 중첩(重疊)의 원리(原理)가 적용(適用)되는 선형(線型)(linear)으로 취급(取扱)하여 통계적(統計的)으로 모형(模型)을 설정(設定)하고 유역(流域) Response의 물리적(物理的) System 및 그 변화(變化)를 연역적(演譯的)으로 정성적(定性的)(qualitatively), 개략적(槪略的)(lumped)인 해석(解析)을 하려는 것이다. 각(各) 회귀계수(回歸係數)들이 각(各) parameter들의 phisical properties의 의미(意味)를 내포(內包)하는 일종(一種)의 grey box로 해석(解析)하려는 statistically deterministic model이다. 2. 금강(錦江) 수계(水系)의 4개(個) 수문지점(水文地點)의 선형회귀모형(線型回歸模型)의 방정식(方程式)은 다음과 같다. 통계적(統計的) 기준(基準)에 따라 판정(判定)한 결과(結果) 고도(高度)의 유의성(有意性)이 있는 모형(模型)으로 판정(判定)되었으며 특(特)히 유출량(流出量) 기록(記錄)이 짧은 경우에도 이용(利用)할 수 있다. 각(各) parameter들의 회귀계수(回歸係數)는 유역면적(流域面積)에 따라 질서(秩序)있게 변화(變化)하는 것을 알 수 있다. 즉(卽) 강우량(降雨量)(Pn)의 경우 유역(流域)이 커질수록 interception, detention storage에 의(依)한 손실(損失)도 커지며, 토양수분변화량(土壤水分變化量) (Qn-1)의 경우 유역(流域)이 커질수록 유역(流域)의 저류능(貯溜能)이 커지므로 기저유출(基底流出)도 커지며, 증발산량(蒸發散量)(En)의 경우 유역이 커질수록 Coverage의 나지화(裸地化) 면적(面積)이 커지므로 증발산량(蒸發散量) 손실(損失)도 커진다. 3. 선형회귀모형(線型回歸模型)의 정성적(定性的)인 phisical properties가 유역면적(流域面積)에 따라 변화(變化)하는데 착안(着眼)하여 모형(模型)을 일반화(一般化)하여 수계별(水系別) 유역면적별(流域面積別)로 회귀계수(回歸係數)를 구(求)하여 무계기(無計器) 하천(河川)에서도 월유출량(月流出量) 추정(推定) 모형(模型)을 설정(設定)할 수 있게 하였다. 금강수계(錦江水系)의 선형회귀모형(線型回歸模型)의 일반화도표(一般化圖表)는 다음 Fig.10과 같다.