• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.3
• /
• pp.231-240
• /
• 2001

Conventional expert systems has been criticized due to its lack of capability to adapt to the changing decision-making environments. In literature, many methods have been proposed to make expert systems more environment-adaptive by incorporating fuzzy logic and neural networks. The objective of this paper is to propose a new approach to building a self-evolving expert system inference mechanism by integrating fuzzy neural network and fuzzy rule extraction technique. The main recipe of our proposed approach is to fuzzify the training data, train them by a fuzzy neural network, extract a set of fuzzy rules from the trained network, organize a knowledge base, and refine the fuzzy rules by applying a pruning algorithm when the decision-making environments are detected to be changed significantly. To prove the validity, we tested our proposed self-evolving expert systems inference mechanism by using the bankruptcy data, and compared its results with the conventional neural network. Non-parametric statistical analysis of the experimental results showed that our proposed approach is valid significantly.

• Steel and Composite Structures
• /
• v.19 no.5
• /
• pp.1237-1258
• /
• 2015

This paper proposes a simple inelastic analysis approach to efficiently map out the complete nonlinear post-collapse (strain-softening) response and the maximum load capacity of axially loaded concrete encased steel composite columns (stub and slender). The scheme simultaneously incorporates the influences of difficult instabilizing phenomena such as concrete confinement, initial geometric imperfection, geometric nonlinearity, buckling of reinforcement bars and local buckling of structural steel, on the overall behavior of the composite columns. The proposed numerical method adopts fiber element discretization and an iterative M${\ddot{u}}$ller's algorithm with an additional adaptive technique that robustly yields solution convergence. The accuracy of the proposed analysis scheme is validated through comparisons with various available experimental benchmarks. Finally, a parametric study of various key parameters on the overall behaviors of the composite columns is conducted.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.1 s.33
• /
• pp.35-44
• /
• 2005

The degree reduction of B-splines is necessary in exchanging parametric curves and surfaces of the different geometric modeling systems because some systems limit the supported maximal degree. In this paper, We provide an our experimental results in approximate conversion for B-splines apply to degree reduction. We utilize the existing Bezier degree reduction methods, and analyze the methods. Also, knot removal algorithm is used to reduce data in the degree reduction Process.

• Journal of the Korea Society of Computer and Information
• /
• v.8 no.1
• /
• pp.7-12
• /
• 2003

Because some systems limit the supported maximal degree, the degree reduction of NURBS is necessary in Parametric curves and surfaces of the different geometric modeling systems. Therefore an approximate degree reduction method of NURBS curves was introduced in this research. Also the existing Eck's B$\'{e}$zier degree reduction method and knot removal algorithm were used to reduce data in the degree reduction process. Finally we found that this method was stable, efficient for implementations, and easy to use algorithms.

• Korean Journal of Computational Design and Engineering
• /
• v.2 no.2
• /
• pp.122-129
• /
• 1997

Ball rolling blending is a popular technique for blending between parametric surfaces. The ball rolling blend surface is conceptually a trajectory of a ball rolling between two base sufaces. It is constructed by sweeping a circular arc along a ball contact curve pair. Since a ball rolling blend surfaces does not have a polynomial form like a Bezier surface patch, it is impossible to apply this method directly to a commercial CAD/CAM system. In this paper an algorithm is developed to approximate a ball rolling blend surface into Bezier surface patches. Least square method is applied to obtain proper Bezier surface patches under a given tolerance. The Bezier surface patches have degree three or more and guarantee VC1-continuity.

• Journal of Digital Contents Society
• /
• v.5 no.3
• /
• pp.176-179
• /
• 2004

Superquadrics can represent various and complex 3D object with only wome parameters(size, position, deformation etc). So if we use both superquadrics and deformed superquadrics, we can reoresent more realistic 3D object which are existed in real world. In this paper we use the Z-buffer algorithm and stencil buffer together because this is very useful when the wuperquadric primitives are combined. The fundamental ideas are illustrated with a number of tables and figures.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.25 no.12
• /
• pp.654-659
• /
• 2013

This paper presents a numerical optimization of louvered fins to enhance the JF factor in terms of the design parameters, including the fin pitch, the number of louvers, the louver angle, the fin thickness, and the re-direction louver length. We carried out a parametric study to select the three most important parameters affecting the JF factor, which were the fin pitch, number of louvers, and the louver angle. We optimally designed the louvered fin by using 3rd-order full factorial design, the kriging method, and a micro genetic algorithm. Consequently, the JF factor of the optimum model increased by 16% compared to that of the base model. Moreover, the optimum model reduced the pressure drop by 17% with a comparable heat transfer rate.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.37 no.4
• /
• pp.239-244
• /
• 2014

The conventional clustering approaches are mostly based on minimizing total dissimilarity of input and output. However, the clustering approach may not be helpful in some cases of clustering decision making units (DMUs) with production feature converting multiple inputs into multiple outputs because it does not care converting functions. Data envelopment analysis (DEA) has been widely applied for efficiency estimation of such DMUs since it has non-parametric characteristics. We propose a new clustering method to identify groups of DMUs that are similar in terms of their input-output profiles. A real world example is given to explain the use and effectiveness of the proposed method. And we calculate similarity value between its result and the result of a conventional clustering method applied to the example. After the efficiency value was added to input of K-means algorithm, we calculate new similarity value and compare it with the previous one.

• Proceedings of the KIEE Conference
• /
• 2007.10a
• /
• pp.131-132
• /
• 2007

This paper presents a numerical solution of PID controller design. The complete set of PID controllers stabilizing an LTI plant that solely determined from the knowledge of the frequency response and the number of unstable poles of the plant. An illustrative example is given to describe implementation of the design algorithm and exhibit graphical displays of the feasible regions of the design parameters.

• Smart Structures and Systems
• /
• v.8 no.2
• /
• pp.191-204
• /
• 2011

A technique for damage localization and assessment based on measurements of both eigenvectors curvatures and eigenfrequencies is proposed. The procedure is based on two successive steps: a model independent localization, based on changes of modal curvatures, and the solution of a one-dimensional minimization problem to evaluate damage intensity. The observability properties of damage parameters is discussed and, accordingly, a suitable change of coordinates is introduced. The proposed technique is illustrated with reference to a cantilever Euler beam endowed with a set of piezoelectric transducers. To assess the robustness of the algorithm, a parametric study of the identification errors with respect to the number of transducers and to the number of considered modal quantities is carried out with both clean and noise-corrupted data.