• 소성∙가공
• /
• 제30권3호
• /
• pp.125-133
• /
• 2021

In this study, response surface method (RSM), back propagation neural network (BPNN), and genetic algorithm (GA) were used for modeling and multi-objective optimization of the parameters of AA5052-H32 in incremental sheet forming (ISF). The goal of optimization is to determine the maximum forming angle and minimum surface roughness, while varying the production process parameters, such as tool diameter, tool spindle speed, step depth, and tool feed rate. A Box-Behnken experimental design (BBD) was used to develop an RSM model and BPNN model to model the variations in the forming angle and surface roughness based on variations in process parameters. Subsequently, the RSM model was used as the fitness function for multi-objective optimization of the ISF process the GA. The results showed that RSM and BPNN can be effectively used to control the forming angle and surface roughness. The optimized Pareto front produced by the GA can be utilized as a rational design guide for practical applications of AA5052 in the ISF process

• 지질공학
• /
• 제13권3호
• /
• pp.369-378
• /
• 2003

캐오리나이트, 일라이트, 녹리석의 중금속(Pb, Cu, Zn, Cd) 흡착에 다양한 실험적 변수의 영향을 박스-벤켄(Box-Benken)모델을 이용한 통계 모델링을 실시하였다 중금속 흡착에 미치는 변수로 pH, 중금속의 초기농도, HCO3 (혹은 K)농도를 변수로 하고, 각 실험변수의 농도를 3차원으로 설정하여 모델에 근거한 벳치실험과 3차원 통계모델링을 실시하였다. 아울러 일라이트에 한하여 pH변화에 따른 중금속의 흡착특성을 알아보았다. 통계적 해를 얻기 위하여 SAS 프로그램을 이용하여 표면반응 분석을 통하여 실시하였다 중금속의 흡착에 영향을 미치는 가장 큰 변수는 pH이며, 그다음 중금속의 초기 농도와 중탄산(혹은 K성분)의 농도가 실험적 조건에 따라서 서로 다른 영향력을 보인다. 점토광물에 따른 중금속 흡착능력의 차이는 크지 않지만 일라이트의 경우 중금속의 초기농도에 따라서 흡착율에 상당한 차이를 보인다. 일라이트의 중금속 흡착에 K 성분이 흡착경쟁에 영향을 미치는 것으로 나타났다. 중금속간의 흡착경쟁은 Cu>Pb>>Zn>Cd의 순서를 보인다. 중금속 흡착에 점토광물의 광물학적 영향보다는 수용액의 화학적 조건이 더 큰 영향을 미치는 것으로 나타났다. 통계모델에 근거한 흡착모델링은 흡착결과를 3차원으로 효과적으로 표현할 수 있으며, 실험적 노력을 경감할 수 있는 효과적인 방법으로 판단된다.

• 한국해양공학회지
• /
• 제34권4호
• /
• pp.253-262
• /
• 2020

This study provides a comparative analysis of experiments-based enhancements and sensitivity evaluations for the structural design of an automatic ocean salt collector under various load conditions. The sizing variables of the structural members were considered as design factors. The strength and weight performances were selected as output responses. The design of experiments used in the comparative study consisted of the orthogonal array design, Box-Behnken design, and central composite design. The response surface model, one of the metamodels, was applied to the approximate model generation. The design enhancement performance metrics, including numerical costs and weight minimization, according to the design of experiments, were compared from the best design case results. The central composite design method showed the most enhanced design results for the structural design of the automatic ocean salt collector.

• 한국식생활문화학회지
• /
• 제29권6호
• /
• pp.623-636
• /
• 2014

This study was performed to determine the optimal composition of muffins added with dropwort powder, egg and sugar. The experiment was designed based on BBD (Box-behnken design), and evaluation was carried out by means of RSM (response surface methodology), consisting of 13 experimental points with three replicates each for three independent variables dropwort powder, eggs and sugar. Using F-test, height, moisture, 1,1-diphenyl-2picriylhydrazyl (DPPH) radical scavenging activity, L-value, a-value, springiness, cohesiveness, chewiness, brittleness, color, and appearance acceptability were expressed as a linear model, whereas pH, b-value, flavor, taste, and texture were expressed as a quadratic model. After swallow and overall acceptability were expressed as a 2FI model. As dropwort powder content increased, 1,1-diphenyl-2picriylhydrazyl (DPPH) radical scavenging activity increased. Increasing amounts of dropwort powder led to reduction of sensory scores for color, appearance, flavor, taste, and texture quality. The optimum formulation determined by the numerical and graphical methods were similar: dropwort powder 6 g, egg 77 g, and sugar 60 g.

• 반도체디스플레이기술학회지
• /
• 제18권2호
• /
• pp.48-52
• /
• 2019

Virtual metrology, which is one of APC techniques, is a method to predict characteristics of manufactured films using machine learning with saving time and resources. As the photoresist is no longer a mask material for use in high aspect ratios as the CD is reduced, hard mask is introduced to solve such problems. Among many types of hard mask materials, amorphous carbon layer(ACL) is widely investigated due to its advantages of high etch selectivity than conventional photoresist, high optical transmittance, easy deposition process, and removability by oxygen plasma. In this study, VM using different machine learning algorithms is applied to predict the thickness of ACL and trained models are evaluated which model shows best prediction performance. ACL specimens are deposited by plasma enhanced chemical vapor deposition(PECVD) with four different process parameters(Pressure, RF power, $C_3H_6$ gas flow, $N_2$ gas flow). Gradient boosting regression(GBR) algorithm, random forest regression(RFR) algorithm, and neural network(NN) are selected for modeling. The model using gradient boosting algorithm shows most proper performance with higher R-squared value. A model for predicting the thickness of the ACL film within the abovementioned conditions has been successfully constructed.

• 동아시아식생활학회지
• /
• 제23권6호
• /
• pp.799-809
• /
• 2013

This study was performed to determine the optimal composition of a makgeolli administered nuruk, water and Asparagus cochinchinensis. The experiment was designed base on BBD (box-behnken design), and an evaluation was carried out by means of RSM (response surface methodology), which included 15 experimental points with 3 replicates for the three independent variables nuruk, water and Asparagus cochinchinensis. The nuruk levels were tested in a range of 10~30 g, the water levels in a range 120~180% and Asparagus cochinchinensis was tested within a range of 2~6% by the weight of cooked-rice. Using the F-test, brix and appearance were expressed as a linear model, whereas the pH, acidity, DPPH radical scavenging, L-value, savory taste, taste, fresh aroma, after swallow and overall acceptability were expressed as a quadratic model. Increased amount of Asparagus cochinchinensis led to the reduction of the sensory scores for appearance, flavor, taste, texture and overall quality. The optimum formulation by numerical and graphical method were similar: nuruk 24.50 g, water 174.95% and Asparagus cochinchinensis 2.40%.

• Journal of Microbiology and Biotechnology
• /
• 제19권7호
• /
• pp.690-697
• /
• 2009

In the present study, production of compactin by Penicillium brevicompactum WA 2315 was studied. In the first step, various precultural parameters were studied by substituting one factor at a time. Subsequently, the effect of maltodextrin DE 18 on compactin production was studied. The optimized parameters gave maximum compactin production of 850 ${\mu}g/gds$as compared with 678 ${\mu}g/gds$before optimization. Statistical study was performed to further improve the production and develop a robust model. An improved yield of 950 ${\mu}g/gds$was obtained using the conditions proposed by the experimental model. The present study emphasizes the importauce of precultural and nutritional parameters on the production of compactin, and further confirms the usefulness of solid-state fermentation for the production of industrially important secondary metabolites. It also confirms that complex nitrogen sources such as oil cakes can be used for the production of compactin.

• 한국버섯학회지
• /
• 제21권3호
• /
• pp.110-117
• /
• 2023

Cordycepin (3'-deoxyadenosine) is a nucleoside analog known for its diverse range of biological activities. This study investigated the effect of different types of sawdust on the production of the bioactive compound cordycepin. The results of the study showed that different types of wood sawdust affected the biosynthesis of cordycepin and a significant increase was observed when the conventional SDB medium was replaced with 1% NaOH treated pine sawdust. To optimize cordycepin production from Paecilomyces tenuipes in a medium containing 1% NaOH-pretreated pine sawdust, we employed Response Surface Methodology (RSM) in its Box-Behnken design (BBD) canonical form. The optimal conditions were determined as follows: a particle size of 109.5111-mesh (140 ㎛) for 1% NaOH-pretreated pine sawdust, an input weight of 21.1679 g/L, and an incubation time of 73.8423 hours. According to our model, this combination is expected to yield a maximum cordycepin content of 896.1428 ㎍/mL. Experimental validation of this prediction was performed using the suggested optimal conditions, resulting in an average cordycepin content of 922.6771 ㎍/mL across three replicates, thus confirming the model's accuracy.

• 대한환경공학회지
• /
• 제30권6호
• /
• pp.642-652
• /
• 2008

본 연구는 축산폐수 중 COD$_{Cr}$을 응집공정 후 Photo-Fenton 공정에 의해 산화분해 최적조건 및 제거 예측식에 수립에 관한 연구이다. 본 연구는 축산폐수 중 COD$_{Cr}$에 대한 Photo-Fenton 산화반응을 이용하여 이들 분해특성을 2차원 반응모델로 추정하기 위해 중심합설계법 대안으로 많이 사용되고 있는 박스-벤켄법(Box-Behnken method)을 이용하였다. 최적조건 수립을 위한 입력변수, 즉 3가지 변수(Fe(II)(x$_1$), $H_2O_2(x_2)$, pH(x$_3$)) 등을 램덤화, 반복화 및 블록화 원리에 따라 실험설계하여 반응값에 대한 예측식을 수학적으로 산출하였다. 수학적 및 통계적으로 산출된 예측식은 Y = 79.3 + 15.61x$_1$ - 7.31x$_2$ - 4.26x$_3$ - 18x$_1{^2}$ - 10x$_2{^2}$ - 11.9x$_3{^2}$ + 2.49x$_1x_2$ - 4.4x$_2x_3$ - 1.65x$_1x_3$와 같이 얻을 수 있었고 COD$^{Cr}$ 제거율(%)의 실측치에 대한 예측치의 적합도(goodness of fit) 검증시 결정계수(coefficient of determination: R$^2$) 0.96으로 에측식을 충분히 설명할 수 있었다. 예측 모형에 대한 최소제곱추정법으로 적합된 반응표면에서 1차 선형항(linear term)은 Fe(II)(x$_1$), $H_2O_2(x_2)$, 그리고 pH(x$_3$)은 상승작용(synergistic effect)으로 반응모델에 유의한 차이를 보였으며(p < 0.001) 그러나 교호항(cross-product term)은 $H_2O_2$ $\times$ pH(x$_2x_3$)와 순수이차항(quadratic terms)의 Fe(II) $\times$ Fe(II)(x$_1{^2}$), $H_2O_2$ $\times$ H$_2O_2$(x$_2{^2}$) 그리고 pH $times$ pH(x$_3{^2}$) 등은 대립적인(감쇠)(antagonistic effect) 작용으로 반응모델에 유의한 차이를 보였다(p < 0.01). 반응 모델에 대한 예측식 수립 후 COD$_{Cr}$ 제거율(%)의 최적조건을 도출하기 위해 정준분석(canonical analysis)와 능선분석(ridge analysis)에 이용한 결과 반응값(결과값: Y)은 84 $\pm$ 0.95%, COD$_{Cr}$ 최적처리를 위한 변수들의 조건은 Fe(II)(X$_1$) = 0.0146 mM, $H_2O_2$(X$_2$) = 0.0867 mM 그리고 pH(X$_3$) = 4.704 등의 결과를 얻을 수 있었다. 또한 이들 최적조건을 이용하여 재현성을 통한 모델검증 결과 높은 신뢰성을 보였다.

• 생명과학회지
• /
• 제22권1호
• /
• pp.74-79
• /
• 2012

본 실험은 효소들을 이용하여 볏짚으로부터 수용성 탄수화물을 생산할 때 각 효소들의 상호작용을 탐색하기 위하여 수행되었다. 볏짚의 수준에 따른 각 효소들의 수용성 탄수화물 생산 활성을 조사한 결과, 볏짚의 수준이 높을 경우(100 g/l)에는 cellulase와 hemicellulase가 서로 유사한 활성을 나타낸 반면, 볏짚의 수준이 50 g/l로 낮아질 경우, hemicellulase의 활성은 낮아지지만 cellulase는 높은 활성으로 유지되었다. 효소반응액에 포함된 볏짚의 수준, hemicellulase 및 xylanase의 상호작용을 Box Behnken design과 반응표면모형을 이용하여 분석한 결과 hemicellulase과 xylanase 간의 상호작용이 발견되었으며, 각각의 효소들을 0.55 mg/ml와 0.65 mg/ml로 혼합하는 것이 볏짚의 수준과는 상관없이 가장 많은 수용성 탄수화물을 생산하는 결과를 나타내었다. 따라서, 결과적으로 볏짚의 효소처리를 통한 수용성 탄수화물의 생산에 있어 cellulase가 가장 큰 효과를 나타내며, hemicellulase과 xylanase는 서로 상호작용을 통하여 수용성 탄수화물 생성량을 향상시킨다는 알 수 있었다.