• Proceedings of the KOSOMBE Conference
• /
• v.1996 no.05
• /
• pp.232-234
• /
• 1996

Cadiac output estimation scheme of LVAD using multi-dimensional interpolation technique was introduced in this paper. This paper also show appropriate input -output data for estimation. Experimental results show our approach is a good one for the estimation of nonlinear hemodynamics.

• International Journal of Quality Innovation
• /
• v.3 no.2
• /
• pp.113-131
• /
• 2002

The relationships among multi-dimensional data (such as medical examination data) with ambiguity and variation are difficult to explore. The traditional approach to building a data classification system requires the formulation of rules by which the input data can be analyzed. The formulation of such rules is very difficult with large sets of input data. This paper first describes two classification approaches using back-propagation (BP) neural network and Mahalanobis distance (MD) classifier, and then proposes two classification approaches for multi-dimensional feature selection. The first one proposed is a feature selection procedure from the trained back-propagation (BP) neural network. The basic idea of this procedure is to compare the multiplication weights between input and hidden layer and hidden and output layer. In order to simplify the structure, only the multiplication weights of large absolute values are used. The second approach is Mahalanobis-Taguchi system (MTS) originally suggested by Dr. Taguchi. The MTS performs Taguchi's fractional factorial design based on the Mahalanobis distance as a performance metric. We combine the automatic thresholding with MD: it can deal with a reduced model, which is the focus of this paper In this work, two case studies will be used as examples to compare and discuss the complete and reduced models employing BP neural network and MD classifier. The implementation results show that proposed approaches are effective and powerful for the classification.

• Nuclear Engineering and Technology
• /
• v.52 no.2
• /
• pp.230-237
• /
• 2020

In this part, an implicit time dependent solution is presented for the Boltzmann transport equation discretized by the analytic coarse mesh finite difference method (ACMFD) over the spatial domain as well as the simplified P₃ (SP₃) for the angular variable. In the first part of this work we proposed a SP₃-ACMFD approach to solve the static eigenvalue equations which provide the initial conditions for temp dependent equations. Having solved the 3D multi-group SP₃-ACMFD static equations, an implicit approach is resorted to ensure stability of time steps. An exponential behavior is assumed in transverse integrated equations to establish a relationship between flux moments and currents. Also, analytic integration is benefited for the time-dependent solution of precursor concentration equations. Finally, a multi-channel one-phase thermal hydraulic model is coupled to the proposed methodology. Transient equations are then solved at each step using the GMRES technique. To show the sufficiency of proposed transient SP₃-ACMFD approximation for a full core analysis, a comparison is made using transport peers as the reference. To further demonstrate superiority, results are compared with a 3D multi-group transient diffusion solver developed as a byproduct of this work. Outcomes confirm that the idea can be considered as an economic interim approach which is superior to the diffusion approximation, and comparable with transport in results.

• Journal of Ocean Engineering and Technology
• /
• v.8 no.1
• /
• pp.23-32
• /
• 1994

A numerical procedure is described for simultaneously predicting the motion and structural responses of tension leg platforms (TLPs) in multi-directional irregular waves. The developed numerical approach is based on a combination of a three dimensional source distribution method, the finite element method for structurally treating the space frame elements and a spectral analysis technique of directional waves. The spectral description for the linear responses of a structure in the frequency domain is sufficient to completely define the responses. This is because both the wave inputs and the responses are stationary Gaussian ran dom process of which the statistical properties in the amplitude domain are well known. The hydrodynamic interactions among TLP members, such as columns and pontoons, are included in the motion and structural analysis. The effect of wave directionality has been pointed out on the first order motion, tether forces and structural responses of a TLP in multi-directional irregular waves.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.21 no.12
• /
• pp.2055-2067
• /
• 1997

Values of process parameters in sheet metal forming can be estimated by various one-step inverse methods. One-step inverse methods based on deformation theory, however, cause some amount of error. The amount of error is generally increased as the deformation path becomes more complex. As a remedy, a new three dimensional multi-step inverse method is introduced for optimum design of blank shapes and strain distributions from desired final shapes. The approach extends a one-step inverse method to a multi-step inverse method in order to reduce the amount of error. The algorithm developed is applied to square cup drawing to confirm its validity by demonstrating reasonably accurate numerical results. Rapid calculation with this algorithm enables easy determination of an initial blank of sheet metal forming.

• Clean Technology
• /
• v.7 no.2
• /
• pp.99-107
• /
• 2001

Optimization methodologies have been proposed of find the best environment-friendly recycling pathways of plastic materials based on life-cycle assessment (LCA) methodology. The main difficulty in conducting this optimization study is that multiple environmental burdens have to be considered simultaneously as the cost functions. Instead of generating conservative Pareto or noninferior solutions following multi-objective optimization approaches, we have proposed some practical criteria on how to combine the different environmental burdens into a single measure. The obtained single objective optimization problem can then be solved by conventional nonlinear programming techniques or, more effectively, by a tree search method based on decision flows. The latter method reduces multi-dimensional optimization problems to a set of one-dimensional problems in series. It is expected the suggested tree search approach can be applied to many LCA studies as a new promising optimization tool.

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 2005.05a
• /
• pp.262-265
• /
• 2005

A shape optimization of flow guide is accomplished to minimize the wear rate of die in three-dimensional flat-die extrusion processes. In order to achieve the balanced flow and the uniformed distribution of the effective strain during the extrusion, a multi-objective optimization is implemented. During the process of optimization formulation, the flow balance and the deviation of strain is considered as constrained conditions. The proposed approach is applied to an extrusion of H section. Through the optimization, it has been confirmed that the wear rate of die can be minimized satisfying the constraint.

• Communications for Statistical Applications and Methods
• /
• v.28 no.3
• /
• pp.267-279
• /
• 2021

A regression with multi-dimensional responses is quite common nowadays in the so-called big data era. In such regression, to relieve the curse of dimension due to high-dimension of responses, the dimension reduction of predictors is essential in analysis. Sufficient dimension reduction provides effective tools for the reduction, but there are few sufficient dimension reduction methodologies for multivariate regression. To fill this gap, we newly propose two fused slice-based inverse regression methods. The proposed approaches are robust to the numbers of clusters or slices and improve the estimation results over existing methods by fusing many kernel matrices. Numerical studies are presented and are compared with existing methods. Real data analysis confirms practical usefulness of the proposed methods.

• Journal of Digital Convergence
• /
• v.15 no.1
• /
• pp.247-253
• /
• 2017

Feature selection or variable selection is a data mining scheme for selecting highly relevant features with target concept from high dimensional data. It decreases dimensionality of data, and makes it easy to analyze clusters or classification. A feature selection scheme requires an evaluation function. Most of current evaluation functions are based on statistics or information theory, and they can evaluate only for single feature (one-dimensional data). However, features have interactions between them, and require evaluation function for multi-dimensional data for efficient feature selection. In this study, we propose modification of FSDD evaluation function for utilizing evaluation of multiple features using extended distance function. Original FSDD is just possible for single feature evaluation. Proposed approach may be expected to be applied on other single feature evaluation method.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.21 no.48
• /
• pp.123-132
• /
• 1998

The group formation problem of the machine and part is a very important issue in the planning stage of cellular manufacturing systems. This paper investigates Self-Organizing Map(SOM) neural networks approach to machine-part grouping problem. We present a two-phase algorithm based on SOM for grouping parts and machines. SOM can learn from complex, multi-dimensional data and transform them into visually decipherable clusters. Output layer in SOM network is one-dimensional structure and the number of output node has been increased sufficiently to spread out the input vectors in the order of similarity. The proposed algorithm performs remarkably well in comparison with many other algorithms for the well-known problems shown in previous papers.