• 제목/요약/키워드: sequential experimental design

검색결과 87건 처리시간 0.029초

품질 향상에 적용되는 전산 실험의 계획과 분석 (Design and Analysis of Computer Experiments with An Application to Quality Improvement)

  • Jung Wook Sim;Jeong Soo Park;Jong Sung Bae
    • 응용통계연구
    • /
    • 제7권1호
    • /
    • pp.83-102
    • /
    • 1994
  • 컴퓨터 시뮬레이션 실험을 이용한 제반 연구의 효율성을 높이기 위한 통계적 실험 계획법으로서 최적 실험법과 라틴 하이퍼큐브 계획법에 대하여 연구하여 최적 라틴 하이퍼큐브 계획법을 제시하였다. 또한 전산 실험 자료의 분석을 위하여, 공간적 예측모형을 택하여 자료로부터의 모수추정과 이 모형에 적합한 예측방법 및 최적 실험 계획법 등이 고려되었다. 최적 라틴 하이퍼큐브 실험계획법을 구성하기 위한 2단계 (2점 교환법 및 뉴톤방법) 알고리즘과 그것에 의한 결과를 제시하였고, 나아가 축차적(최적) 라틴 하이퍼큐브 계획법의 구축을 위한 한 방법을 제시하였다. 이와같은 접근법은 주요인 그림과 축차적인 계획 및 분석을 이용하여 집적회로 계획의 최적화 문제로 응용되어 결국 품질향상에 도움이 되도록 하는 실예를 통하여 그 실제적 적용성이 예증되었다.

  • PDF

이식형 약물 주입 장치용 마이크로 튜브 펌프 (Micro Tube Pump for an Implantable Drug Infusion Device)

  • 백두진;박준우;이덕희;이철한;홍소영;조영호
    • 한국유체기계학회 논문집
    • /
    • 제11권1호
    • /
    • pp.40-45
    • /
    • 2008
  • This paper proposes a tube pump composed of small-sized cams and followers for an implantable intrathecal drug infusion device. Each followers is driven by a cam and liquid is discharged by a sequential reciprocal motion of the followers. The advantage of this structure is that it allows the pump to be clean and valveless. To design a small-sized, low power pump some analysis were performed to determine the design parameters of the cam, follower and the tube. To verify the feasibility of the experiment, a prototype was manufactured and its operating characteristics were investigated. Experimental results were in accordance with the expected results obtained from analysis.

Effective Gas Identification Model based on Fuzzy Logic and Hybrid Genetic Algorithms

  • Bang, Yonug-Keun;Byun, Hyung-Gi;Lee, Chul-Heui
    • 센서학회지
    • /
    • 제21권5호
    • /
    • pp.329-338
    • /
    • 2012
  • This paper presents an effective design method for a gas identification system. The design method adopted the sequential combination between the hybrid genetic algorithms and the TSK fuzzy logic system. First, the sensor grouping method by hybrid genetic algorithms led the effective dimensional reduction as well as effective pattern analysis from a large volume of pattern dimensions. Second, the fuzzy identification sub-models allowed handling the uncertainty of the sensor data extensively. By these advantages, the proposed identification model demonstrated high accuracy rates for identifying the five different types of gases; it was confirmed throughout the experimental trials.

Feature curve extraction from point clouds via developable strip intersection

  • Lee, Kai Wah;Bo, Pengbo
    • Journal of Computational Design and Engineering
    • /
    • 제3권2호
    • /
    • pp.102-111
    • /
    • 2016
  • In this paper, we study the problem of computing smooth feature curves from CAD type point clouds models. The proposed method reconstructs feature curves from the intersections of developable strip pairs which approximate the regions along both sides of the features. The generation of developable surfaces is based on a linear approximation of the given point cloud through a variational shape approximation approach. A line segment sequencing algorithm is proposed for collecting feature line segments into different feature sequences as well as sequential groups of data points. A developable surface approximation procedure is employed to refine incident approximation planes of data points into developable strips. Some experimental results are included to demonstrate the performance of the proposed method.

반응표면방법론을 이용한 BLDC전동기의 코깅토크 저감에 관한 연구 (A study on Reduction of Cogging Torque for BLDC Motor Using Response Surface Methodology Optimization)

  • 김영균;이근호;홍정표
    • 대한전기학회논문지:전기기기및에너지변환시스템부문B
    • /
    • 제51권2호
    • /
    • pp.55-60
    • /
    • 2002
  • This paper presents an optimization procedure by using Response Surface Methodology(RSM) to determine design parameters for reducing cogging torque. RSM is achieved through using the experimental design method in combination with Finite Element Method and adapted to make analytical model for a complex problem considering a lot of interaction of these parameters. Sequential Quadratic Problem (SQP) method is used to solve the relsulting of constrained nonlinear optimization problem.

Optimal Design of Laminate Composites with Gradient Structure for Weight Reduction

  • Back, Sung-Ki;Kang, Tae-Jin;Lee, Kyung-Woo
    • 한국복합재료학회:학술대회논문집
    • /
    • 한국복합재료학회 1999년도 추계학술발표대회 논문집
    • /
    • pp.68-72
    • /
    • 1999
  • In an effort to construct a structure under the design principle of minimal use of materials for maximum performances, a discrete gradient structure has been introduced in laminate composite systems. Using a sequential linear programming method, the gradient structure of composites to maximize the buckling load was optimized in terms of fiber volume fraction and thickness of each layer. Theoretical optimization results were then verified with experimental ones. The buckling load of laminate composite showed maximum value with the outmost [$0^{\circ}$] layer concentrated by almost all the fibers when the ratio of length to width(aspect ratio) was less than 1.0. But when the aspect ratio was 2.0, the optimum was determined in a structure where the thickness and fiber volume fraction were well balanced in each layer. From the optimization of gradient structure, the optimal fiber volume fraction and thickness of each layer were proposed. Experimental results agreed well with the theoretical ones. Gradient structures have also shown an advantage in the weight reduction of composites compared with the conventional homogeneous structures.

  • PDF

Statistical Optimization of Medium Composition for Growth of Leuconostoc citreum

  • Kim, Hyun;Eom, Hyun-Ju;Lee, Jun-Soo;Seo, Jin-Ho;Han, Nam-Soo
    • 한국생물공학회:학술대회논문집
    • /
    • 한국생물공학회 2005년도 생물공학의 동향(XVI)
    • /
    • pp.68-72
    • /
    • 2005
  • Leuconostoc citreum is one of the representative strains of Leuconostoc spp. that show fast growth rates in fermented vegetables. Sequential experimental designs including the Plackett-Burman design, fractional factorial design, steepest ascent analysis, central composite design and response surface methodology were introduced tooptimize and improve the medium for Leuconostoc citreum. Fifteen medium ingredients were examined and glucose (20 g/l), yeast extract (12.5 g/l), sodium acetate trihydrate (6.12 g/l), potassium phosphate (42.55 g/l) and dibasic ammonium citrate (4.12 g/l)were chosen as the best components to give a critical and positive effect for cell-growth. The biomass was increased to 2.79 g/l (169%), compared to the 1.65 g/l in MRS medium.

  • PDF

Statistical Optimization of Medium Composition for Growth of Leuconostoc citreum

  • Kim, Hyun;Eom, Hyun-Ju;Lee, Jun-soo;Han, Jin-soo;Han, Nam-Soo
    • Biotechnology and Bioprocess Engineering:BBE
    • /
    • 제9권4호
    • /
    • pp.278-284
    • /
    • 2004
  • Leuconostoc citreum is one of the representative strains of Leuconostoc spp. that show fast growth rates in fermented vegetables. Sequential experimental designs including the Plackett-Burman design, fractional factorial design, steepest ascent analysis, central composite design and response surface methodology were introduced to optimize and improve the medium for L. citreum. Fifteen medium ingredients were examined and glucose ($20 g/\ell$), yeast extract ($12.5g/\ell$), sodium acetate trihydrate ($6.12g/\ell$), potassium phosphate ($42.55g/\ell$), and dibasic ammonium citrate ($4.12g/\ell$), were chosen as the best components to give a critical and positive effect for cell-growth. The biomass was increased to ($2.79g/\ell$), (169%), compared to the $1.65g/\ell$ in MRS medium.

Systemic Statistical Optimization of Astaxanthin Inducing Methods in Haematococcus pluvialis cells -Statistical Optimization of Astaxanthin Production in Haematococcus

  • Kim, Sun-Hyoung;Jeong, Sung Eun;Hong, Seong-Joo;Lee, Choul-Gyun
    • 한국해양바이오학회지
    • /
    • 제6권1호
    • /
    • pp.31-40
    • /
    • 2014
  • The production of astaxanthin in the microalga Haematococcus pluvialis has been investigated using a sequential methodology based on the application of two types of statistical designs. The employed preliminary experiment was a fractional factorial design $2^6$ in which the factors studied were: excessive irradiance and nitrate starvation, phosphate deficiency, acetate supplementation, salt stress, and elevated temperature. The experimental results indicate that the amount of astaxanthin accumulation in the cells can be enhanced by excessive irradiance and nitrate starvation whereas the other factors tested did not yield any enhancement. In the subsequent experiment, a central composite design was applied with four variables, light intensity, nitrate, phosphate, and acetate, at five levels each. The optimal conditions for the highest astaxanthin production were found to be $1040{\mu}E/(m^2{\cdot}s)$ light intensity, 0.04 g/L nitrate, 0.31 g/L phosphate, 0.05 g/L acetate concentration.

아세틸렌 흡착공정용 MOF-235 합성 최적화를 위한 실험 계획법 적용 (An Application of Design of Experiments for Optimization of MOF-235 Synthesis for Acetylene Adsorption Process)

  • 조형민;유계상
    • 공업화학
    • /
    • 제31권4호
    • /
    • pp.377-382
    • /
    • 2020
  • 아세틸렌 흡착 공정을 위한 MOF-235 합성의 최적화를 위해 순차적인 실험 계획법을 사용하였다. 이를 위하여 두 가지 실험 계획법이 적용되었는데, screening을 위한 2단계 요인 설계와 반응표면 분석법 중에 하나인 중심합성 계획이다. 본 연구에서는 23 요인 설계법을 이용하여 MOF-235의 결정도에 대한 합성 온도, 합성 시간 및 혼합 속도의 영향을 평가하였다. MINITAB 19 소프트웨어에 따라 설계된 16번의 MOF-235 합성 실험을 수행하였다. XRD 분석을 바탕으로 Half-Normal, Pareto, Residual, Main 및 Interaction 효과를 구하였다. 시험 결과의 분산 분석(ANOVA)을 통해 합성 온도 및 시간이 MOF-235의 결정도에 중요한 영향을 미친다는 것을 분석하였다. 이후, 중심합성 계획법을 이용하여 아세틸렌 흡착성능 최적화를 MOF-235의 선정된 합성 조건을 바탕으로 수행하였다. 설계된 9번의 흡착실험을 통해 도출된 결과를 2차 모델 방정식을 이용하여 계산하였다. 아세틸렌의 최대 흡착 용량(18.7 mmol/g)은 86.3 ℃ 및 28.7 h의 최적의 조건에서 합성된 MOF-235에서 얻을 수 있다고 예측하였다.