• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2008.04a
• /
• pp.306-310
• /
• 2008

Fuzzy regression, a nonparametric method, can be quite useful in estimating the relationships among variables where the available data are very limited and imprecise. It can also serve as a sound methodology that can be applied to a variety of management and engineering problems where variables are interacting in an uncertain, qualitative, and fuzzy way. A close examination of the fuzzy regression algorithm reveals that the resulting possibility distribution of fuzzy parameters, which makes this technique attractive in a fuzzy environment, is dependent upon an h parameter value. The h value, which is between 0 and 1, is referred to as the degree of fit of the estimated fuzzy linear model to the given data, and is subjectively selected by a decision maker (DM) as an input to the model. The selection of a proper value of h is important in fuzzy regression, because it determines the range of the posibility ditributions of the fuzzy parameters. In this paper, we discuss the interdependent relationship among the h value, membership function shape, and the spreads of fuzzy parameters in fuzzy linear regression with fuzzy input-output using shape-preserving operations.

• Journal of the Optical Society of Korea
• /
• v.16 no.4
• /
• pp.372-380
• /
• 2012

The optical method Digital Speckle Correlation Measurement (DSCM) has been extensively applied due its capability to measure the entire displacement field over a body surface. A formula of displacement measurement errors by the gradient-based DSCM method was derived. The errors were found to explicitly relate to the image grayscale errors consisting of sub-pixel interpolation algorithm errors, image noise, and subset deformation mismatch at each point of the subset. A power-law dependence of the standard deviation of displacement measurement errors on the subset size was established when the subset deformation was rigid body translation and random image noise was dominant and it was confirmed by both the numerical and experimental results. In a gradient-based algorithm the basic assumption is rigid body translation of the interrogated subsets, however, this is in contradiction to the real circumstances where strains exist. Numerical and experimental results also indicated that, subset shape function mismatch was dominant when the order of the assumed subset shape function was lower than that of the actual subset deformation field and the power-law dependence clearly broke down. The power-law relationship further leads to a simple criterion for choosing a suitable subset size, image quality, sub-pixel algorithm, and subset shape function for DSCM.

• Transactions of Materials Processing
• /
• v.8 no.5
• /
• pp.454-464
• /
• 1999

A new blank design method is proposed to predict the optimum initial blank shape in the sheet metal forming process. The rollback method for blank shape design takes the difference between the deformed blank contour and the target contour shape into account. the minimization object function R is proposed. Based on the method, a computer program composed of blank design module, FE-analysis module and mesh generation module is developed. The rollback method is applied to square cup, reentrant cross section, L-shaped cup drawing process with the flange of uniform size around its periphery to confirm its validity. The optimum initial blank shape is obtained from an arbitrary blank shape after several modifications. Good agreements are recognized between the numerical results and the published experimental results for initial blank shape and thickness strain distribution. It is concluded that the rollback method is an effective and convenient method for an optimum blank shape design.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2006.11a
• /
• pp.694-698
• /
• 2006

This paper presents an optimal design process using beads on a body panel to improve interior noise of a passenger vehicle. Except modification of structural members, it is difficult to find effective countermeasures that can work for the intermediate frequency range from 100 Hz to 300 Hz which lies between the booming and low medium frequency. In this study, it is a major goal to find additional counter-measures for this intermediate frequency range by performing optimal design of beads on body panels. The proposed method for design optimization consists of 4 sub-steps, that is, a) problem definition, b) cause analysis, c) countermeasure development and d) validation. The objective function is minimization of interior noise level. The major design variables are the geometrical shape of a bead and combination of beads on the critical panels. Sensitivity analysis and optimization are performed according to the predefined process for an optimal design. It is verified that the proposed design decreases the level of noise transfer function above 5 dB.

• Journal of the Korean Data and Information Science Society
• /
• v.11 no.2
• /
• pp.181-188
• /
• 2000

The shape of an M-estimation function is generally determined in the sense of either/both maximizing efficiency of an M-estimator at the model or/and bounding the influence function of an M-estimator. We propose an empirical method of choosing a value of the bending constant in Huber's ${\psi}-function$, which is the most widely used M-estimation function when estimating the location parameter.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.11 no.3
• /
• pp.138-144
• /
• 2003

Recently, the weight reduction of vehicle influences its environment problems and performances. It is a trend that a lot of parts have been currently changed to an aluminum alloy from steel materials. In this study, the shape optimization using an orthogonal array is performed to determine the design of the knuckle which is a part of suspension system. With the material of the weight reduction was achieved by satisfying the constraints of a strength requirement. The orthogonal array of $L_{18}$ is introduced to find the optimum design variables that considers the shape of the knuckle. The characteristic function composed of the objective and the construct is defined to the feasibility. Comparing to the weight of the initial design with steel materials that of optimum design with aluminum alloy material is reduced by 60%.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.27 no.9
• /
• pp.1351-1359
• /
• 2003

This study presents a numerical procedure to optimize shape of streamwise periodic ribs mounted on both of the principal walls to enhance turbulent heat transfer in a rectangular channel flow. The response surface method is used as an optimization technique. The optimization is based on Navier-Stokes analysis of flow and heat transfer with $k-{\varepsilon}$ turbulence model. The width-to-height ratio of a rib, rib height-to-channel height ratio and rib pitch to rib height ratio are chosen as design variables. The object function is defined as a function of heat transfer coefficient and friction drag coefficient with weighting factor. Optimum shapes of the rib have been obtained for the range of 0.02 to 0.1 of weighting factor.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.8 no.1
• /
• pp.135-147
• /
• 2000

Section design of vehicle structure has been developed by two methods. One is the section property method which uses section property as a design variable. This method shows the tendency of an optimum section approximately. The other method is the section shape method which utilizes geometric parameter of section as a design variable. Practical solutions are obtained by this method. However, it is very expensive for large-scale problems due to the large number of geometric parameters. These two methods are compared through several sample problems. The finite element method is used for the structural and sensitivity analyses. The results are analyzed based on the number of function evaluations, the quality of cost function, the complexity of programing, and etc. The applications of both methods are also discussed.

• KIEE International Transactions on Electrophysics and Applications
• /
• v.5C no.4
• /
• pp.155-160
• /
• 2005

This paper addresses the discharge characteristics of cross-linked polyethylene according to void by the Weibull function. It analyzes discharge number and amount of discharge using Weibull distribution to identify the inter-relationship between partial discharge and defect. We detected a 10 second discharge. The applied voltage increased by 1 [kV] at discharge inception voltage. As a result, in a no-void specimen, the shape parameter was consistent according to the increase of voltage, whereas, in a void specimen, it increased according to the increase of voltage. As the result, the shape parameter expressed a fixed value at no-void specimen. However, in void specimen, according to increasing voltage shape parameter rapidly increases.

• Journal of the Society of Naval Architects of Korea
• /
• v.43 no.5 s.149
• /
• pp.550-559
• /
• 2006

A design procedure for a ship with minimum total resistance was developed using a numerical optimization method called SQP(Sequential Quadratic Programming) and a CFD technique based on the Rankine source panel method with the nonlinear free surface boundary conditions. During the whole optimization process the geometry of the hull shape was represented based on the NURBS(Non-uniform rational B-spline) technique and the modification of the hull shape was controlled using the Bell-shaped distribution function to keep the fairness of the hull shape before and after the hull modification. The numerical analysis was carried out using 4000TEU container ship in the towing tank facility installed in the Pusan national university to know the validity of the developed algorithm for this study. As the results of the numerical analysis it proved that the resistance of the optimized hull is conspicuously reduced in comparison with the original hull in a wave-making resistance point of view.