• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.29 no.6 s.237
• /
• pp.835-841
• /
• 2005

This study proposes a new optimization algorithm which is combined with genetic algorithm and ANOM. This improved genetic algorithm is not only faster than the simple genetic algorithm, but also gives a more accurate solution. The optimizing ability and convergence rate of a new optimization algorithm is identified by using a evaluation function which have several local optimum and an optimum design of rocker arm shaft. The calculation results are compared with the simple genetic algorithm.

• Proceedings of the KSME Conference
• /
• 2000.04a
• /
• pp.708-713
• /
• 2000

The design of experiment(DOE) is getting more attention in the engineering community since it is easy to understand and apply. Recently, engineering designers are adopting DOE with orthogonal arrays when they want to design products in a discrete design space. In this research, a design flow with orthogonal arrays is defined fur structural design according to the general DOE. The design problem is defined as a general structural optimization problem. Sensitivity information is evaluated by the analysis of variance(ANOVA), and an optimum design is determined from analysis of means(ANOM). Interactions between design variables are investigated to achieve additivity which should be valid in DOE. When strong interactions exit, a method is proposed. Some methods to consider the problem are suggested.

• Proceedings of the KSME Conference
• /
• 2001.06a
• /
• pp.770-775
• /
• 2001

Steel bar breaking system is a component of a sled test system for automobile crashworthiness. It is a recent idea for the sled test. In a sled test, a crash pulse is given as a input made from a real test. The steel bar breaking system is designed to generate a certain crash pulse. Orthogonal arrays from design of experiments (DOE) are employed. The factors of the array are panel thickness and the number of steel bars, and the levels are candidate values of them. A simulation is utilized for the crash analysis. A commercial system called LS/DYNA3D is adopted. A test system is designed based on the results.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.18 no.2
• /
• pp.278-291
• /
• 1994

A computer-aided robust design (CARD) technique is developed to search for the design variables, optimal as well as robust in the sense of Taguchi method. The CARD technique can effectively handle inequality problems by employing the variable penalty method, and dynamic problems with many design variables and/or with mixed discrete and continuous variables. It is also capable of providing contributions of each design variables to the object funtion and information for future designs. As the illustrative examples, two dynamic systems, engine mounting system and in-line feeder, are treated.

• Proceedings of the KSME Conference
• /
• 2003.11a
• /
• pp.1660-1665
• /
• 2003

On this study, we improved the efficiency applying algorithm that is repeatedly using orthogonal array in discrete design space and filling a defect of gradient method in continuous design space. we showed optimal ply angle that maximized 1st natural frequency of CFRP laminated composite cantilever plate by each aspect ratio. A finite element analysis on the CFRP laminated composite cantilever plate using orthogonal array is carried out, and the results are compared with those obtained by modal testing.

• The Transactions of the Korean Institute of Electrical Engineers B
• /
• v.54 no.11
• /
• pp.505-510
• /
• 2005

On this study, we optimized maximizing the thrust force at the TFLM(Transverse Flux Linear Motor) using design of experiments by the table of orthogonal array and the analysis of means(ANOM), and classified the most contributive design factor maximizing the thrust force at the TFLM by analysis of variance(ANOVA). From now on, we are going to apply the required technique to design various uses and shapes of the TFLM.

• Journal of Power System Engineering
• /
• v.6 no.2
• /
• pp.68-72
• /
• 2002

Design of experiment was applied to analyze the shrinkage characteristics of the bobbin molded by injection molding. Among lots of design and processing conditions, the thickness of a bobbin and cooling conditions of a mold were considered. The temperature difference between top and bottom parts of the bobbin was considered as the objective to minimize the shrinkage of a bobbin. Optimal thickness of a bobbin was 2.0mm at the part of body and 1.5mm at the part of wing, respectively. Optimal cooling conditions such as cooling time and coolant inlet temperature were 12 second and $12^{\circ}C$, respectively.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2006.05a
• /
• pp.395-396
• /
• 2006

This paper presents the optimal design of Robot-Arm part use Design of Experiment(DOE). The DOE(Design of Experiment)was conducted to find out main effect factors for design of Robot-Arm part. In this design of Robot-Arm, 5 control factors include numbers of 4 level are selected and we make out L16 orthogonal array. Using this orthogonal array, find out optimal value and main effect factors of object function for design of Robot-Arm part by 16 times of test. We evidence this optimal value by using CATIA V5 Analysis.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2006.05a
• /
• pp.786-792
• /
• 2006

The design of experiment(DOE) with orthogonal arrays is adopted when the engineering design is needed in a discrete design space. In this research, a design process with orthogonal array is peformed to obtain the optimum design which satisfy the frequency target of the steering system. The optimum design is determined from the analysis of means(ANOM) and sensitivity information about design variables is evaluated by the analysis of variance(ANOVA). Interactions between design variables are investigated to achieve additivity which should be valid in using orthogonal array. It is shown that when strong interactions exist, the DOE process with orthogonal array considering interaction is introduced to find out optimum design.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.7 no.5
• /
• pp.121-129
• /
• 1999

Recently, the regulations from the govemment and the concems of the people give rise to the interest in exhaust noise of passenger car as much as other vehicles. The exact analysis of various mufflers is needed to reduce the level of exhaust noise. In this paper, we propose a design to improve the mufflers capacity by reducing noise of exhaust system combining Taguchi method and fractional factorial design. In order to measure the performance of a muffler, the performance prediction software which is developed by the Dept. of Automotive Engineering at Hanyang University is used. From the current muffler system we select control factors such as lenght and radius of each component that are thought to be effective on capacity of muffler. Factors are arranged using L18, L27 table of orthogonal array and the fractional factorial design for analysis. We find some significant interaction effects using 1/3 fractional factorial design and accomplish the reduction of noise from the muffler.