• Proceedings of the KIEE Conference
• /
• 2000.07d
• /
• pp.2744-2746
• /
• 2000

일반화된 회귀 신경망을 이용하여 반도체 공정 최적화를 위한 플라즈마를 모델링한다. 플라즈마는 Box-W린son 실험계획표에 의해 특성화되었으며, 여기에서 변화시킨 인자로는 소스전력, 압력, 척지지대의 위치, 그리고 염소의 유량이다. 총 24회의 실험이 수행이 되었으며, 플라즈마 변수는 Langmuir Probe를 이용하여 측정하였다. 측정된 주요 플라즈마 변수로는 전자밀도, 전자온도, 그리고 플라즈마 전위이다. 폭변수를 점진적으로 증가시켜 회귀신경망을 최적화하였으며. 최적화된 모델은 통계적인 반응표면모델과 비교하였다. 비교 결과, 회귀신경망은 반응표면모델에 상응하는 예측능력을 보이고 있음을 알 수 있었다.

• International Journal of Highway Engineering
• /
• v.4 no.1 s.11
• /
• pp.123-134
• /
• 2002

Special pavement test sections were selected to develop a distress prediction model on asphalt pavement of National Highway. Experimental design was conducted for the selection of LTPP sections on in-service pavement(new and overlaid pavement) using several variables affecting pavement performance. Preliminary sections that satisfied the design template were chosen from the national highway database, and final selection was fixed through field inspection. The number of monitoring section is 95 including 47 overlaid pavement. A pavement distress data such as crack and rutting were collected for two years. An interim pavement performance analysis was peformed to show feasibility of performance monitoring program. Data related pavement such as traffic, weather, material characteristic and crack etc. should be collected for next project years and distress prediction model will be developed through the statistical analysis.

• Journal of Korean Society for Quality Management
• /
• v.4 no.1
• /
• pp.25-30
• /
• 1976

• Ocean and Polar Research
• /
• v.26 no.2
• /
• pp.333-339
• /
• 2004

This paper is concerned with an experimental investigation about tractive performance of a tracked vehicle on extremely soft soil. A tracked vehicle model with principal dimensions of $0.9\;m(L)\;{\times}\;0.75\;m(B)\;{\times}\;0.4\;m(H)$ and the weight of 167 kg was constructed with a pair of driving chain links driven by two AC-servo motors. The tracks are configured with detachable grousers with variable span. Deep seabed was simulated by means of bentonite-water mixture in a soil bin of $6.0\;m(L)\;{\times}\;3.7\;m(B)\;{\times}\;0.7\;m(H)$. Slip of vehicle and driving torque of motor were measured with respect to experimental variables; grouser span, grouser chevron angle, driving speed, drawbar-pull weight, position of center-of-gravity and weight. $L_8$ orthogonal array is adopted for DOE (Design Of Experiment). The effects of experiment variables on traction performance are evaluated.

• Korean Journal of Materials Research
• /
• v.3 no.5
• /
• pp.514-520
• /
• 1993

In this paper the cure cycle of carbon fiber/phenolic resin was investigated by the Taguchi Method in an experimental design. Experiments were systematically performed using $L_{18}(2^1 \times 3_7)$ orthorgonal array table of the experimental design. In the experimental design, eight compression molding parameters (heating rate, pressing temperature, pressing rate, molding pressure, curing temperature, dwell time at curing temperature, cooling rate and degassing) were considered and the effects of the parameters on the flexural strength and the apparent porosity of carbon fiber/phenolic composites were investigated. The analysis of variance for the experimental results indicated that molding pressure and curing temperature are the most significant parmeters in the flexural strength and the apparent porosity of carbon fiber/phenolic resin composites, respectively.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.10 no.1
• /
• pp.143-152
• /
• 2002

European side impact analysis of an electric vehicle was done using the robust design method. In order to minimize VC as well as rib deflection, the injury response table which consists of rib deflection and VC response table has been introduced. The sensitivities and interactions are almost the same when it was compared with those of rib deflection and VC response table. Using internal energy of the factors, the starting time of dummy rib deflection and the contact average velocity, the internal energy and time-velocity response table were introduced. It is shown that the results of the new response tables have the similar characteristics to those of the Injury response table. It is suggested that the internal energy and time-velocity response table should be utilized to minimize injuries.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.28 no.7
• /
• pp.885-895
• /
• 2004

Generally, automatic design is carried out with computer simulation and the simulation models are established by investigating the correlations between the simulation and real experiments. Therefore, the experiment results are utilized as complimentary data although they are considered to be precise. Orthogonal arrays have been adopted for discrete design. A method is proposed to directly exploit the experiment results in the design process with orthogonal arrays. Experiments are allocated to some rows of an orthogonal array and computer simulations are allocated to the others. A rule for the allocation is found to keep the orthogonality. Error analysis of the design results is performed. Mathematical examples are made to verify the validity of the proposed method. Error models are defined with the examples and the design solutions from the examples are discussed.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.6
• /
• pp.557-564
• /
• 2019

The characteristics of the torque and torque ripple of Interior Permanent Magnet Synchronous Motor(IPMSM) are influenced by the size and position of the rotor magnet and the size of the stator slot. This paper deals with the optimal design method for improving torque and torque ripplerate for IPMSM using Response Surface Methodology(RSM). Two objective functions of torque output and torque ripple were derived from the sensitivity analysis by Plackett-Burmann(PB) for the characteristic variables affecting torque and torque ripple. Secondary characteristic variables were selected from the derived objective function and RSM secondary regression model function was estimated by the experiment schedule and analysis results according to the Central Composite Design (CCD). The reliability of the secondary regression model was verified using ANOVA table. The analysis according to the experimental schedule was verified by JMAG(Ver. 18.0) which is Finite Element Method(FEM) software. The torque output of IPMSM applied with final characteristic variables was increased torque output by 11.5 % and the torque ripplerate was reduced by 9.1 %.

• Proceedings of the KIEE Conference
• /
• 2007.07a
• /
• pp.1805-1806
• /
• 2007

Silicon nitride (SiN) 박막을 플라즈마 응용화학기상법을 이용하여 증착하였다. SiN박막의 전하밀도는 일반화된 회귀 신경망 (GRNN)을 이용하여 모델링하였다. PECVD 공정은 Box Wilson 실험계획표를 이용하여 수행하였다. GRNN 모델의 예측수행은 유전자 알고리즘 (GA)을 이용하여 최적화하였다. 최적화한 GA-GRNN 모델은 종래의 GRNN 모델과 비교하여, 약55%정도의 예측성능의 향상을 보였다.

• Proceedings of the Korean Institute of Surface Engineering Conference
• /
• 2009.10a
• /
• pp.233-234
• /
• 2009

증착된 silicon nitride (Sin) 박막의 수명 시간을 예측하는 신경망 모델을 개발하였다. SiN 박막은 플라즈마 화학기상 증착방식을 이용하여 증착되었다. 증착 공정은 통계적인 실험계획표를 이용하여 수행되었고, 신경망 모델의 예측 성능은 유전자 알고리즘을 이용하여 최적화하였다. 수명시간은 다른 박막특성 (굴절률, 증착률, 전하밀도)의 영향을 상당히 받았으며, 특히 굴절률과 전하밀도는 높은 증착률에서 증가시킬 때 수명시간을 최대화할 수 있었다.