• Journal of the Korean Statistical Society
• /
• v.34 no.2
• /
• pp.141-151
• /
• 2005

Under the assumption that the three-level factors are quantitative, the linear effects are taken more attention than the quadratic effects of the interaction terms. Webb (1971) presented some small incomplete factorial designs that are mixed two- and three-level designs with 20 or fewer runs. The designs provided the estimating linear-by-linear components of interactions between the three-level factors; moreover, they could also offer estimation of interactions that interest the experiments. Webb used ad hoc methods to find these plans; hence, there was still no unified structure to those experiments. In this paper, we develop the methods to construct the $2^{n}3^3$ and $2^{1}3^3$ designs. The designs constructed by these methods not only supply orthogonal estimates of all the main effects but also permit estimation of all the two-factor interactions not involving the quadratic effects. Furthermore, the designs we find are nearly orthogonal.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.20 no.1
• /
• pp.222-231
• /
• 1996

Temperature distribution on the flank face in orthogonal turning with cutting tool of high speed steel is studied by using a finite element method and experiments. Experiments are carried out to verify the validity of the temperature measurement by using a thermoelectric couple junciton imbedded in a cutting tool of high speed steel. Good agreement is obtained between the analytical results and the experimental ones for the temperature distributions on flank face of cutting tool with igh speed steel. The analytical results show that the temperature on the top flank face of a tool is higher because of the difference of the friction velocity on each face of the tool.

• Wind and Structures
• /
• v.3 no.2
• /
• pp.97-115
• /
• 2000

The technique of proper orthogonal decomposition is potentially useful in specifying the fluctuating surface pressure field around structures. However there has been a degree of controversy over whether or not the calculated modes have physical meanings. This paper addresses this issue through consideration of the results of full scale experiments, and through an analytical investigation. It is concluded that the lower, most energetic modes are likely to reflect different fluctuating flow mechanisms, although no mode is likely to be associated with just one flow mechanism or vice versa. The higher, less energetic modes are likely to represent interactions between different flow mechanisms, and to be significantly affected by the number of measurement points and measurement errors. The paper concludes with a brief description of the application of POD to the problem of building ventilation, and the calculation of cladding pressures.

• Journal of the Korean Society for Precision Engineering
• /
• v.12 no.10
• /
• pp.102-112
• /
• 1995

Temperature distribution on the rake face and flank face in orthogonal turning with cutting tool of high speed steel is studied by using a finite element method and experiments. Experiments are carried out to verify the validity of the temperature measurement by using a thermoelectric couple junction imbedded in a cutting tool of high speed steel. Good agreement is obtained between the analytical results and the experimental ones for the temperature distributions on both the rake face and flank face of cutting tool with high speed steel. The analytical results show that the temperature on the top flank face of a tool is higher than it on the top rake face of the tool because of the difference of the friction velocity on each face of the tool.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.10 no.6
• /
• pp.70-76
• /
• 2011

This paper presents an optimization procedure for reducing warpage of injection molded part by using a volumetric shrinkage deviation as an objective function. A design of experiments based on orthogonal arrays was used in the optimization procedure, and the entire optimization was performed through a two stage process - a preliminary experimentation and a principal experimentation. Proposed optimization method was applied to the design of a CPU-base part in computer. With the moderate number of experiments, an optimal molding condition for uniform distribution of volumetric shrinkage was obtained, as a result, the warpage of the molded part was significantly reduced.

• Proceedings of the Korean Society For Composite Materials Conference
• /
• 2001.05a
• /
• pp.205-210
• /
• 2001

In this work, virtual material characterization of 3D orthogonal woven composites is performed to predict the elastic properties by a full scale FEA. To model the complex geometry of 3D orthogonal woven composites, an accurate unit structure is first prepared. The unit structure includes warp yarns, filler yarns, stuffer yams and resin regions and reveals the geometrical characteristics. For this virtual experiments by using finite element analysis, parallel multifrontal solver is utilized and the computed elastic properties are compared to available experimental results and the other analytical results. It is founded that a good agreement between material properties obtained from virtual characterization and experimental results. Using the method of this virtual material characterization, the effects of inconsistent filler yarn distribution on the in-plane shear modulus and filler yarn waviness on the transverse Young's modulus are investigated. Especially, the stiffness knockdown of 3D woven composite structures is simulated by virtual characterization. Considering these results, the virtual material characterization of composite materials can be used for designing the 3D complex composite structures and may supplement the actual experiments.

• Journal of Mechanical Science and Technology
• /
• v.20 no.3
• /
• pp.366-375
• /
• 2006

In this paper, it is intended to introduce a method to solve multi-objective optimization problems and to evaluate its performance. In order to verify the performance of this method it is applied for a vertical roller mill for Portland cement. A design process is defined with the compromise decision support problem concept and a design process consists of two steps: the design of experiments and mathematical programming. In this process, a designer decides an object that the objective function is going to pursuit and a non-linear optimization is performed composing objective constraints with practical constraints. In this method, response surfaces are used to model objectives (stress, deflection and weight) and the optimization is performed for each of the objectives while handling the remaining ones as constraints. The response surfaces are constructed using orthogonal polynomials, and orthogonal array as design of experiment, with analysis of variance for variable selection. In addition, it establishes the relative influence of the design variables in the objectives variability. The constrained optimization problems are solved using sequential quadratic programming. From the results, it is found that the method in this paper is a very effective and powerful for the multi-objective optimization of various practical design problems. It provides, moreover, a reference of design to judge the amount of excess or shortage from the final object.

• Journal of Power System Engineering
• /
• v.13 no.5
• /
• pp.89-94
• /
• 2009

In this work, the optimal process has been designed by $L_9(3^4)$ orthogonal array and analysis of variance(ANOVA) for thermal-sprayed Ni-based hard coating. Ni-based hard coatings were fabricated by flame spray process on steel substrate. Then, the hardness test and observation of microstructure of the coatings were performed. The results of hardness test were analyzed by ANOVA. The ANOVA results demonstrated that the acetylene gas flow had the greatest effect on hardness of the coatings. The oxygen gas flow was found to have a neglecting effect. From these results, the optimal combination of the flame spray parameters could be predicted. The calculated hardness of the coatings by ANOVA was found to lie close to that of confirmation experimental result. Thus, it was considered that design of experiments design using orthogonal array and ANOVA was useful to determine optimal process of thermal-sprayed Ni-based hard coating.

• Journal of Drive and Control
• /
• v.12 no.4
• /
• pp.8-14
• /
• 2015

A quick-acting hydraulic fuse, which is mainly composed of a poppet, a seat, and a spring, must be designed to minimize the leaked oil volume during fuse operation on a line rupture. The optimal design parameters of a quick-acting hydraulic fuse were searched using the design of experiments method and the complex method. First, the $L_{50}(5^4)$ orthogonal array is used to find the robust minimum point among the 625 points of design variables. The search range can then be narrowed around the robust minimum point. Second, the $L_{25}(5^4)$ orthogonal array is used to obtain the variations of the design variables in the narrowed search range. The variations of design variables are used to set the structure of a polynomial equation representing the leakage oil volume of the quick-acting hydraulic fuse. The least squares method is then applied to obtain the coefficients of polynomial equation. Finally, the complex method is used to find the optimal design parameters where the objective function is described by the polynomial equation.

• Journal of Computing Science and Engineering
• /
• v.4 no.2
• /
• pp.97-109
• /
• 2010

Nonnegative matrix factorization (NMF) is a popular method for multivariate analysis of nonnegative data, which is to decompose a data matrix into a product of two factor matrices with all entries restricted to be nonnegative. NMF was shown to be useful in a task of clustering (especially document clustering), but in some cases NMF produces the results inappropriate to the clustering problems. In this paper, we present an algorithm for orthogonal nonnegative matrix factorization, where an orthogonality constraint is imposed on the nonnegative decomposition of a term-document matrix. The result of orthogonal NMF can be clearly interpreted for the clustering problems, and also the performance of clustering is usually better than that of the NMF. We develop multiplicative updates directly from true gradient on Stiefel manifold, whereas existing algorithms consider additive orthogonality constraints. Experiments on several different document data sets show our orthogonal NMF algorithms perform better in a task of clustering, compared to the standard NMF and an existing orthogonal NMF.